Graph Transformers (GTs) have emerged as a promising graph learning tool, leveraging their all-pair connected property to effectively capture global information. To address the over-smoothing problem in deep GNNs, global attention was initially introduced, eliminating the necessity for using deep GNNs. However, through empirical and theoretical analysis, we verify that the introduced global attention exhibits severe over-smoothing, causing node representations to become indistinguishable due to its inherent low-pass filtering. This effect is even stronger than that observed in GNNs. To mitigate this, we propose PageRank Transformer (ParaFormer), which features a PageRank-enhanced attention module designed to mimic the behavior of deep Transformers. We theoretically and empirically demonstrate that ParaFormer mitigates over-smoothing by functioning as an adaptive-pass filter. Experiments show that ParaFormer achieves consistent performance improvements across both node classification and graph classification tasks on 11 datasets ranging from thousands to millions of nodes, validating its efficacy. The supplementary material, including code and appendix, can be found in https://github.com/chaohaoyuan/ParaFormer.

Accurate weather forecasting on Subseasonal-to-Seasonal (S2S) timescale is critical to human society such as agriculture planning and extreme weather preparation. Although data-driven models have become alternatives to computationally intensive Numerical Weather Prediction (NWP) systems, existing Transformer-based approaches suffer from biases due to planar projections distorting the spherical geometry and inadequate handling of vector-scalar variable interactions (e.g., wind velocity vs. temperature). To address these limitations, we propose a graph-based Equivariant and Invariant Message Passing (EIMP) framework that directly processes spherical grid data. It maintains SO(3) equivariant embeddings for vector data and SO(3) invariant embeddings for scalar data, which are interacted by a shared invariant message embedding. Guaranteed equivariant and invariant message aggregation functions are proposed to update embeddings under strict symmetry constraints. Extensive experiments on the Earth Reanalysis 5 (ERA5) reanalysis dataset of 41 years demonstrate the proposed model achieves significant improvement over advanced data-driven models and skillful numerical ECMWF systems. Additionally, we empirically show that EIMP demonstrates geometrically superior predictions and conduct ablation studies to validate the efficacy of its design.

In real-world recommendation scenarios, users typically engage with platforms through multiple types of behavioral interactions. Multi-behavior recommendation algorithms aim to leverage various auxiliary user behaviors to enhance prediction for target behaviors of primary interest (e.g., buy), thereby overcoming performance limitations caused by data sparsity in target behavior records. Current state-of-the-art approaches typically employ hierarchical design following either cascading (e.g., view$\rightarrow$cart$\rightarrow$buy) or parallel (unified$\rightarrow$behavior$\rightarrow$specific components) paradigms, to capture behavioral relationships. However, these methods still face two critical challenges: (1) severe distribution disparities across behaviors, and (2) negative transfer effects caused by noise in auxiliary behaviors. In this paper, we propose a novel model-agnostic Hierarchical Graph Information Bottleneck (HGIB) framework for multi-behavior recommendation to effectively address these challenges. Following information bottleneck principles, our framework optimizes the learning of compact yet sufficient representations that preserve essential information for target behavior prediction while eliminating task-irrelevant redundancies. To further mitigate interaction noise, we introduce a Graph Refinement Encoder (GRE) that dynamically prunes redundant edges through learnable edge dropout mechanisms. We conduct comprehensive experiments on three real-world public datasets, which demonstrate the superior effectiveness of our framework. Beyond these widely used datasets in the academic community, we further expand our evaluation on several real industrial scenarios and conduct an online A/B testing, showing again a significant improvement in multi-behavior recommendations. The source code of our proposed HGIB is available at https://github.com/zhy99426/HGIB.

In the digital era, users typically interact with diverse items across multiple domains (e.g., e-commerce, streaming platforms, and social networks), generating intricate heterogeneous interaction graphs. Leveraging multi-domain data can improve recommendation systems by enriching user insights and mitigating data sparsity in individual domains. However, integrating such multi-domain knowledge for cross-domain recommendation remains challenging due to inherent disparities in user behavior and item characteristics and the risk of negative transfer, where irrelevant or conflicting information from the source domains adversely impacts the target domain's performance. To tackle these challenges, we propose HAGO, a novel framework with Heterogeneous Adaptive Graph coOrdinators, which dynamically integrates multi-domain graphs into a cohesive structure. HAGO adaptively adjusts the connections between coordinators and multi-domain graph nodes to enhance beneficial inter-domain interactions while alleviating negative transfer. Furthermore, we introduce a universal multi-domain graph pre-training strategy alongside HAGO to collaboratively learn high-quality node representations across domains. Being compatible with various graph-based models and pre-training techniques, HAGO demonstrates broad applicability and effectiveness. Extensive experiments show that our framework outperforms state-of-the-art methods in cross-domain recommendation scenarios, underscoring its potential for real-world applications. The source code is available at https://github.com/zhy99426/HAGO.

Attributed Hypergraph Clustering (AHC) aims at partitioning a hypergraph into clusters such that nodes in the same cluster are close to each other with both high connectedness and homogeneous attributes. Existing AHC methods are all based on matrix factorization which may incur a substantial computation cost; more importantly, they inherently require a prior knowledge of the number of clusters as an input which, if inaccurately estimated, shall lead to a significant deterioration in the clustering quality. In this paper, we propose Attributed Hypergraph Representation for Clustering (AHRC), a cluster-number-free hypergraph clustering consisting of an effective integration of the hypergraph topology and node attributes for hypergraph representation, a multi-hop modularity function for optimization, and a hypergraph sparsification for scalable computation. AHRC achieves cutting-edge clustering quality and efficiency: compared to the state-of-the-art (SOTA) AHC method on 10 real hypergraphs, AHRC obtains an average of 20% higher F-measure, 24% higher ARI, 26% higher Jaccard Similarity, 10% higher Purity, and runs 5.5× faster. As a byproduct, the intermediate result of graph representation dramatically boosts the clustering quality of SOTA contrastive-learning-based hypergraph clustering methods, showing the generality of our graph representation.

Query optimization is a complex planning and decision-making problem within the exponentially growing plan space in database management systems (DBMS). Traditional optimization techniques have been extensively studied over decades, leaving limited room for further improvement along this track. Recent developments of Large Language Models (LLMs) have demonstrated their potential in solving complex planning and decision-making problems, such as arithmetic and programmatic tasks. In this paper, we try to explore the potential of LLMs in handling query optimization and propose a tentative LLM-based query optimizer dubbed LLM-QO, established on PostgreSQL's execution engine. In LLM-QO, we formulate query optimization in an autoregressive fashion which directly generates the execution plan without explicit plan enumeration. To investigate the essential input of LLM-QO, we design a customized data recipe named QInstruct to collect the training data from various optimizers and serialize the database's meta data, queries and corresponding plans into a textual format. Based on QInstruct, we implement a two-stage fine-tuning pipeline, Query Instruction Tuning (QIT) and Query Direct Preference Optimization (QDPO), to empower the capability of general-purpose LLMs in handling query optimization. In our experiments, LLM-QO can generate valid and high-quality plans and consistently outperforms both traditional and learned optimizers on three query workloads. Our findings verify that LLMs can be derived as query optimizers where generalization, efficiency and adaptivity deserve further research efforts.

Recently, generative models based on the diffusion process have emerged as a promising direction for automating the design of molecules. However, directly adding continuous Gaussian noise to discrete graphs leads to the problem that the generated data do not conform to the discrete graph data distribution in the training set. Current graph diffusion models either corrupt discrete data through a transition matrix or relax the discrete data to continuous space for the diffusion process. These approaches make it hard to perform extensible conditional generation, such as adapting to text-based conditions, due to the lack of embedding representations and require significant computation resources due to the diffusion process of the bond type matrix. This paper introduces the Hierarchical Graph Latent Diffusion Model (HGLDM), a novel variant of latent diffusion models that overcomes the problem of applying continuous diffusion models directly to discrete graph data. Meanwhile, based on the latent diffusion framework, HGLDM avoids the issues of computational consumption and lack of embeddings for extensible conditional generation. In addition, by comparing the HGLDM with its variant, the Graph Latent Diffusion Model (GLDM), which only has graph-level embeddings, we validate the advantage of the hierarchical graph structure for capturing the relationship between structure information and molecular properties. We evaluate the performance of our model through various conditional generation tasks, demonstrating its superior performance.

Individual personalities significantly influence our perceptions, decisions, and social interactions, which is particularly crucial for gaining insights into human behavior patterns in online social network analysis. Many psychological studies have observed that personalities are strongly reflected in their social behaviors and social environments. Unfortunately, psychological traits like one's personality are high-level and hidden in the innermost corner of data, which is intractable to be uncovered by traditional data mining approaches; The data quality of online social networks is far from sufficient to support such profound psychological analysis, because user behavior records and their attributes are usually very fragmented, missing lots of key information to understand a person in depth; In addition, the social environments in online networks are very complicated, making the interaction patterns between users and their environments underexplored.

Combinatorial medication recommendation (CMR) is a fundamental task of healthcare, which offers opportunities for clinical physicians to provide more precise prescriptions for patients with intricate health conditions, particularly in the scenarios of long-term medical care. Previous research efforts have sought to extract meaningful information from electronic health records (EHRs) to facilitate combinatorial medication recommendations. Existing learning-based approaches further consider the chemical structures of medications, but ignore the textual medication descriptions in which the functionalities are clearly described. Furthermore, the textual knowledge derived from the EHRs of patients remains largely underutilized. To address these issues, we introduce the Natural Language-Assisted Multi-modal Medication Recommendation (NLA-MMR), a multimodal alignment framework designed to learn knowledge from the patient view and medication view jointly. Specifically, NLA-MMR formulates CMR as an alignment problem from patient and medication modalities. In this vein, we employ pretrained language models (PLMs) to extract in-domain knowledge regarding patients and medications, serving as the foundational representation for both modalities. In the medication modality, we exploit both chemical structures and textual descriptions to create medication representations. In the patient modality, we generate the patient representations based on textual descriptions of diagnosis, procedure, and symptom. Extensive experiments conducted on three publicly accessible datasets demonstrate that NLA-MMR achieves new state-of-the-art performance, with a notable average improvement of 4.72% in Jaccard score.

Graph plays a significant role in representing and analyzing complex relationships in real-world applications such as citation networks, social networks, and biological data. Graph intelligence is rapidly becoming a crucial aspect of understanding and exploiting the intricate interconnections within graph data. Recently, large language models (LLMs) and prompt learning techniques have pushed graph intelligence forward, outperforming traditional Graph Neural Network (GNN) pre-training methods and setting new benchmarks for performance. In this tutorial, we begin by offering a comprehensive review and analysis of existing methods that integrate LLMs with graphs. We introduce existing works based on a novel taxonomy that classifies them into three distinct categories according to the roles of LLMs in graph tasks: as enhancers, predictors, or alignment components. Secondly, we introduce a new learning method that utilizes prompting on graphs, offering substantial potential to enhance graph transfer capabilities across diverse tasks and domains. We discuss existing works on graph prompting within a unified framework and introduce our developed tool for executing a variety of graph prompting tasks. Additionally, we discuss the applications of combining Graphs, LLMs, and prompt learning across various tasks, such as urban computing, recommendation systems, and anomaly detection. This lecture-style tutorial is an extension of our original work published in IJCAI 2024[44] and arXiv[77] with the invitation of KDD24.

Large Language Models (LLMs) have revolutionized the fields of computer vision (CV) and natural language processing (NLP). One of the most notable advancements of LLMs is that a single model is trained on vast and diverse datasets spanning multiple domains -- a paradigm we term 'All in One'. This methodology empowers LLMs with super generalization capabilities, facilitating an encompassing comprehension of varied data distributions. Leveraging these capabilities, a single LLM demonstrates remarkable versatility across a variety of domains -- a paradigm we term 'One for All'. However, applying this idea to the graph field remains a formidable challenge, with cross-domain pretraining often resulting in negative transfer. This issue is particularly important in few-shot learning scenarios, where the paucity of training data necessitates the incorporation of external knowledge sources. In response to this challenge, we propose a novel approach called Graph COordinators for PrEtraining (GCOPE), that harnesses the underlying commonalities across diverse graph datasets to enhance few-shot learning. Our novel methodology involves a unification framework that amalgamates disparate graph datasets during the pretraining phase to distill and transfer meaningful knowledge to target tasks. Extensive experiments across multiple graph datasets demonstrate the superior efficacy of our approach. By successfully leveraging the synergistic potential of multiple graph datasets for pretraining, our work stands as a pioneering contribution to the realm of graph foundational model. Code available at https://github.com/cshhzhao/GCOPE.

As a fundamental task in graph data management, maximal clique enumeration (MCE) has attracted extensive attention from both academic and industrial communities due to its wide range of applications. However, MCE is very challenging as the number of maximal cliques may grow exponentially with the number of vertices. The state-of-the-art methods adopt a recursive paradigm to enumerate maximal cliques exhaustively, suffering from a large amount of redundant computation. In this paper, we propose a novel reduction-based framework for MCE, namely RMCE, that aims to reduce the search space and minimize unnecessary computations. The proposed framework RMCE incorporates three kinds of powerful reduction techniques including global reduction, dynamic reduction, and maximality check reduction. Global and dynamic reduction techniques effectively reduce the size of the input graph and dynamically construct subgraphs during the recursive sub-tasks, respectively. The maximality check reduction minimizes the computation for ensuring maximality by utilizing neighborhood dominance between visited vertices. Extensive experiments on 18 real graphs demonstrate that the existing approaches achieve remarkable speedups powered by the proposed techniques.

Scoring systems are commonly seen for platforms in the era of Big Data. From credit scoring systems in financial services to membership scores in E-commerce shopping platforms, platform managers use such systems to guide users towards the encouraged activity pattern, and manage resources more effectively and efficiently. To establish such scoring systems, several “empirical criteria” are first determined, followed by a dedicated top-down design for each score factor, which usually requires enormous effort to adjust and tune the scoring function in the new application scenario. What's worse, many fresh projects usually have no ground truth or any experience to evaluate a reasonable scoring system, making the designing even harder. To reduce the effort of manual adjustment of the scoring function in every new scoring system, we innovatively study the scoring system from the preset empirical criteria without any ground truth and propose a novel framework to improve the system from scratch. In this paper, we propose a “counter-empirical attacking” mechanism that can generate “attacking” behavior traces and try to break the empirical rules of the scoring system. Then an adversarial “enhancer” is applied to evaluate the scoring system and find the improvement strategy. By training the adversarial learning problem, a proper scoring function can be learned to be robust to the attacking activity traces that are trying to violate the empirical criteria. Extensive experiments have been conducted on two scoring systems, including a shared computing resource platform and a financial credit system. The experimental results have validated the effectiveness of our proposed framework.

The task of entity alignment between knowledge graphs (KGs) aims to identify every pair of entities from two different KGs that represent the same entity. Many machine learning-based methods have been proposed for this task. However, to our best knowledge, existing methods all requiremanually craftedseed alignments, which are expensive to obtain. In this paper, we propose the first fully automatic alignment method named AutoAlign, which does not require any manually crafted seed alignments. Specifically, for predicate embeddings, AutoAlign constructs a predicate-proximity-graph with the help of large language models to automatically capture the similarity between predicates across two KGs. For entity embeddings, AutoAlign first computes the entity embeddings of each KG independently using TransE, and then shifts the two KGs' entity embeddings into the same vector space by computing the similarity between entities based on their attributes. Thus, both predicate alignment and entity alignment can be done without manually crafted seed alignments. AutoAlign is not only fully automatic, but also highly effective. Experiments using real-world KGs show that AutoAlign improves the performance of entity alignment significantly compared to state-of-the-art methods. Our source code is available at ruizhang-ai/AutoAlign.

Many patients with chronic diseases resort to multiple medications to relieve various symptoms, which raises concerns about the safety of multiple medication use, as severe drug-drug antagonism can lead to serious adverse effects or even death. This paper presents a Decision Support System, called DSSDDI, based on drug-drug interactions to support doctors prescribing decisions. DSSDDI contains three modules, Drug-Drug Interaction (DDI) module, Medical Decision (MD) module and Medical Support (MS) module. The DDI module learns safer and more effective drug representations from the drug-drug interactions. To capture the potential causal relationship between DDI and medication use, the MD module considers the representations of patients and drugs as context, DDI and patients’ similarity as treatment, and medication use as outcome to construct counterfactual links for the representation learning. Furthermore, the MS module provides drug candidates to doctors with explanations. Experiments on the chronic data collected from the Hong Kong Chronic Disease Study Project and a public diagnostic data MIMIC-III demonstrate that DSSDDI can be a reliable reference for doctors in terms of safety and efficiency of clinical diagnosis, with significant improvements compared to baseline methods. Source code of the proposed DSSDDI is publicly available at https://github.com/TianBian95/DSSDDI.

Recently, "pre-training and fine-tuning'' has been adopted as a standard workflow for many graph tasks since it can take general graph knowledge to relieve the lack of graph annotations from each application. However, graph tasks with node level, edge level, and graph level are far diversified, making the pre-training pretext often incompatible with these multiple tasks. This gap may even cause a "negative transfer'' to the specific application, leading to poor results. Inspired by the prompt learning in natural language processing (NLP), which has presented significant effectiveness in leveraging prior knowledge for various NLP tasks, we study the prompting topic for graphs with the motivation of filling the gap between pre-trained models and various graph tasks. In this paper, we propose a novel multi-task prompting method for graph models. Specifically, we first unify the format of graph prompts and language prompts with the prompt token, token structure, and inserting pattern. In this way, the prompting idea from NLP can be seamlessly introduced to the graph area. Then, to further narrow the gap between various graph tasks and state-of-the-art pre-training strategies, we further study the task space of various graph applications and reformulate downstream problems to the graph-level task. Afterward, we introduce meta-learning to efficiently learn a better initialization for the multi-task prompt of graphs so that our prompting framework can be more reliable and general for different tasks. We conduct extensive experiments, results from which demonstrate the superiority of our method.

A graph models the connections among objects. One important graph analytical task is clustering which partitions a data graph into clusters with dense innercluster connections. A line of clustering maximizes a function called modularity. Modularity-based clustering is widely adopted on dyadic graphs due to its scalability and clustering quality which depends highly on its selection of a random graph model. The random graph model decides not only which clustering is preferred - modularity measures the quality of a clustering based on its alignment to the edges of a random graph, but also the cost of computing such an alignment. Existing random hypergraph models either measure the hyperedge-cluster alignment in an All-Or-Nothing (AON) manner, losing important group-wise information, or introduce expensive alignment computation, refraining the clustering from scaling up. This paper proposes a new random hypergraph model called Hyperedge Expansion Model (HEM), a non-AON hypergraph modularity function called Partial Innerclusteredge modularity (PI) based on HEM, a clustering algorithm called Partial Innerclusteredge Clustering (PIC) that optimizes PI, and novel computation optimizations. PIC is a scalable modularity-based hypergraph clustering that can effectively capture the non-AON hyperedge-cluster relation. Our experiments show that PIC outperforms eight state-of-the-art methods on real-world hypergraphs in terms of both clustering quality and scalability and is up to five orders of magnitude faster than the baseline methods.

Graph edit distance (GED) computation is a fundamental NP-hard problem in graph theory. Given a graph pair ( G 1 , G 2 ), GED is defined as the minimum number of primitive operations converting G 1 to G 2 . Early studies focus on search-based inexact algorithms such as A*-beam search, and greedy algorithms using bipartite matching due to its NP-hardness. They can obtain a sub-optimal solution by constructing an edit path (the sequence of operations that converts G 1 to G 2 ). Recent studies convert the GED between a given graph pair ( G 1 , G 2 ) into a similarity score in the range (0, 1) by a well designed function. Then machine learning models (mostly based on graph neural networks) are applied to predict the similarity score. They achieve a much higher numerical precision than the sub-optimal solutions found by classical algorithms. However, a major limitation is that these machine learning models cannot generate an edit path. They treat the GED computation as a pure regression task to bypass its intrinsic complexity, but ignore the essential task of converting G 1 to G 2 . This severely limits the interpretability and usability of the solution. In this paper, we propose a novel deep learning framework that solves the GED problem in a two-step manner: 1) The proposed graph neural network GEDGNN is in charge of predicting the GED value and a matching matrix; and 2) A post-processing algorithm based on k -best matching is used to derive k possible node matchings from the matching matrix generated by GEDGNN. The best matching will finally lead to a high-quality edit path. Extensive experiments are conducted on three real graph data sets and synthetic power-law graphs to demonstrate the effectiveness of our framework. Compared to the best result of existing GNN-based models, the mean absolute error (MAE) on GED value prediction decreases by 4.9% ~ 74.3%. Compared to the state-of-the-art searching algorithm Noah, the MAE on GED value based on edit path reduces by 53.6% ~ 88.1%.

The recent years have witnessed the growing popularity of indoor location-based services (LBS) in practice and research. Among others, indoor shortest path query (ISPQ) is of fundamental importance for indoor LBS. However, existing works on ISPQ ignore indoor temporal variations, e.g., the open and close times associated with entities like doors and rooms. In this paper, we define a new type of query called Indoor Temporal-variation aware Shortest Path Query (ITSPQ). It returns the valid shortest path based on the up-to-date indoor topology at the query time. A set of techniques is designed to answer ITSPQ efficiently. We design a graph structure (IT-Graph) that captures indoor temporal variations. To process ITSPQ using IT-Graph, we design two algorithms that check a doors accessibility synchronously and asynchronously. Furthermore, we propose a novel index structure (IT-Index) that extends the state-of-the-art index significantly by storing dynamic door-to-door distances in a compact distance cube associated with tree nodes. When processing ITSPQ using IT-Index, we make use of the distance cube to avoid time-consuming indoor distance computation on-the-fly. We evaluate the proposed techniques using extensive experiments on synthetic and real data. The results show that our IT-Index based method is the most efficient for processing ITSPQ at a modest cost of index memory consumption.

Modern sociology has profoundly uncovered many convincing social criteria for behavioral analysis. Unfortunately, many of them are too subjective to be measured and very challenging to be presented in online social networks (OSNs) for the large data volume and complicated environments to be explored. On the other hand, data mining techniques can better find data patterns but many of them leave behind unnatural understanding to humans. Although there are some works trying to integrate social observations for specific tasks, they are still hard to be applied to more general cases. In this paper, we propose a fundamental methodology to support the further fusion of data mining techniques and sociological behavioral criteria. Our highlights are three-fold: First, we propose an effective hypergraph awareness and a fast line graph construction framework. The hypergraph can more profoundly indicate the interactions between individuals and their environments because each edge in the hypergraph (a.k.a hyperedge) contains more than two nodes, which is perfect to describe social. A line graph treats each social environment as a super node with the underlying influence between different environments. In this way, we go beyond traditional pair-wise relations and explore richer patterns under various sociological criteria; Second, we propose a novel hypergraph-based neural network to learn social influence flowing from users to users, users to environments, environment to users, and environments to environments. The neural network can be learned via a task-free method, making our model very flexible to support various data mining tasks and sociological analysis; Third, we propose both qualitative and quantitive solutions to effectively evaluate the most common sociological criteria like social conformity, social equivalence, environmental evolving and social polarization. Our extensive experiments show that our framework can better support both data mining tasks for online user behaviors and sociological analysis.

A real-world graph often has frequently interacting nodes on less frequently updated edges. Each interaction activates an existing edge and changes the activeness of the edge. In such an activation network, nodes that are cohesively connected by active edges form a cluster in both structural and temporal senses. For activation networks, incrementally maintaining a structure for an efficient clustering query processing is thus important. This raises problems on maintaining the edge activeness, combining the structural cohesiveness and activeness for clustering, and designing indexes for online clustering queries. This paper considers the time-decay scheme in modelling the activeness and proposes a suite of techniques with great effort made on simplification and innovation for efficiency, effectiveness and scalability. The query time is only related to the query results as opposed to the graph. The index size is linear up to a logarithmic factor. Extensive experiments verify the quality of the clustering results and moreover, the update time is up to six orders of magnitude faster than the baseline.

User Satisfaction Estimation (USE) is an important yet challenging task in goal-oriented conversational systems. Whether the user is satisfied with the system largely depends on the fulfillment of the user’s needs, which can be implicitly reflected by users’ dialogue acts. However, existing studies often neglect the sequential transitions of dialogue act or rely heavily on annotated dialogue act labels when utilizing dialogue acts to facilitate USE. In this paper, we propose a novel framework, namely USDA, to incorporate the sequential dynamics of dialogue acts for predicting user satisfaction, by jointly learning User Satisfaction Estimation and Dialogue Act Recognition tasks. In specific, we first employ a Hierarchical Transformer to encode the whole dialogue context, with two task-adaptive pre-training strategies to be a second-phase in-domain pre-training for enhancing the dialogue modeling ability. In terms of the availability of dialogue act labels, we further develop two variants of USDA to capture the dialogue act information in either supervised or unsupervised manners. Finally, USDA leverages the sequential transitions of both content and act features in the dialogue to predict the user satisfaction. Experimental results on four benchmark goal-oriented dialogue datasets across different applications show that the proposed method substantially and consistently outperforms existing methods on USE, and validate the important role of dialogue act sequences in USE.

Given one or more query vertices, Community Search (CS) aims to find densely intra-connected and loosely inter-connected structures containing query vertices. Attributed Community Search (ACS), a related problem, is more challenging since it finds communities with both cohesive structures and homogeneous vertex attributes. However, most methods for the CS task rely on inflexible pre-defined structures and studies for ACS treat each attribute independently. Moreover, the most popular ACS strategies decompose ACS into two separate sub-problems, i.e., the CS task and subsequent attribute filtering task. However, in real-world graphs, the community structure and the vertex attributes are closely correlated to each other. This correlation is vital for the ACS problem. In this vein, we argue that the separation strategy cannot fully capture the correlation between structure and attributes simultaneously and it would compromise the final performance. In this paper, we propose Graph Neural Network (GNN) models for both CS and ACS problems, i.e., Query Driven-GNN (QD-GNN) and Attributed Query Driven-GNN (AQD-GNN). In QD-GNN, we combine the local query-dependent structure and global graph embedding. In order to extend QD-GNN to handle attributes, we model vertex attributes as a bipartite graph and capture the relation between attributes by constructing GNNs on this bipartite graph. With a Feature Fusion operator, AQD-GNN processes the structure and attribute simultaneously and predicts communities according to each attributed query. Experiments on real-world graphs with ground-truth communities demonstrate that the proposed models outperform existing CS and ACS algorithms in terms of both efficiency and effectiveness. More recently, an interactive setting for CS is proposed that allows users to adjust the predicted communities. We further verify our approaches under the interactive setting and extend to the attributed context. Our method achieves 2.37% and 6.29% improvements in F1-score than the state-of-the-art model without attributes and with attributes respectively.

Big data analytics often requires processing complex join queries in parallel in distributed systems such as Hadoop, Spark, Flink. The previous works consider that the main bottleneck of processing complex join queries is the communication cost incurred by shuffling of intermediate results, and propose a way to cut down such shuffling cost to zero by a one-round multiway join algorithm. The one-round multi-way join algorithm is built on a one-round communication optimal algorithm for data shuffling over servers and a worst-case optimal computation algorithm for sequential join evaluation on each server. The previous works focus on optimizing the communication bottleneck, while neglecting the fact that the query could be computationally intensive. With the communication cost being well optimized, the computation cost may become a bottleneck. To reduce the computation bottleneck, a way is to trade computation with communication via pre-computing some partial results, but it can make communication or pre-computing becomes the bottleneck. With one of the three costs being considered at a time, the combined lowest cost may not be achieved. Thus the question left unanswered is how much should be traded such that the combined cost of computation, communication, and pre-computing is minimal.In this work, we study the problem of co-optimize communication, pre-computing, and computation cost in one-round multiway join evaluation. We propose a multi-way join approach ADJ (Adaptive Distributed Join) for complex join which finds one optimal query plan to process by exploring cost-effective partial results in terms of the trade-off between pre-computing, communication, and computation.We analyze the input relations for a given join query and find one optimal over a set of query plans in some specific form, with high-quality cost estimation by sampling. Our extensive experiments confirm that ADJ outperforms the existing multi-way join methods by up to orders of magnitude.

The graph isomorphism is to determine whether two graphs are isomorphic. A closely related problem is graph automorphism (symmetry) detection, where an isomorphism between two graphs is a bijection between their vertex sets that preserves adjacency, and an automorphism is an isomorphism from a graph to itself. By graph automorphism, we deal with symmetric subgraph matching (SSM), which is to find all subgraphs in a graph G that are symmetric to a given subgraph q in G. To test two graphs for isomorphism, canonical labeling has been studied to relabel a graph in such a way that isomorphic graphs are identical after relabeling. Efficient canonical labeling algorithms are designed by individualization-refinement. They enumerate all permutations using a search tree, and select the minimum one as the canonical labeling. These algorithms face difficulties in handling massive graphs, and the search trees used are for pruning purposes which cannot answer symmetric subgraphs matching. In this paper, we design a new efficient canonical labeling algorithm DviCL based on the observation that we can use the k-th minimum permutation as the canonical labeling. Different from previous algorithms, we take a divide-and-conquer approach to partition a graph G. By partitioning G, an AutoTree is constructed, which preserves symmetric structures as well as the automorphism group of G. The canonical labeling for a tree node can be obtained by composing those of its child nodes, and the canonical labeling for the root is the one for G. Such AutoTree can also be effectively used to answer the automorphism group and symmetric subgraphs. We conducted extensive performance studies using 22 large graphs, and confirmed that DviCL is much more efficient and robust than the state-of-the-art.

We present Mask-GVAE, a variational generative model for blind denoising large discrete graphs, in which ”blind denoising” means we don’t require any supervision from clean graphs. We focus on recovering graph structures via deleting irrelevant edges and adding missing edges, which has many applications in real-world scenarios, for example, enhancing the quality of connections in a co-authorship network. Mask-GVAE makes use of the robustness in low eigenvectors of graph Laplacian against random noise and decomposes the input graph into several stable clusters. It then harnesses the huge computations by decoding probabilistic smoothed subgraphs in a variational manner. On a wide variety of benchmarks, Mask-GVAE outperforms competing approaches by a significant margin on PSNR and WL similarity.

Indoor shortest path query (ISPQ) is of fundamental importance for indoor location-based services (LBS). However, existing ISPQs ignore indoor temporal variations, e.g., the open and close times associated with entities like doors and rooms. In this paper, we define a new type of query called Indoor Temporal-variation aware Shortest Path Query (ITSPQ). It returns the valid shortest path based on the up-to-date indoor topology at the query time. A set of techniques is designed to answer ITSPQ efficiently. We design a graph structure (IT-Graph) that captures indoor temporal variations. To process ITSPQ using IT-Graph, we design two algorithms that check a door's accessibility synchronously and asynchronously, respectively. We experimentally evaluate the proposed techniques using synthetic data. The results show that our methods are efficient.

Many real data come in the form of non-grid objects, i.e. graphs, from social networks to molecules. Adaptation of deep learning from grid-alike data (e.g. images) to graphs has recently received unprecedented attention from both machine learning and data mining communities, leading to a new cross-domain field---Deep Graph Learning (DGL). Instead of painstaking feature engineering, DGL aims to learn informative representations of graphs in an end-to-end manner. It has exhibited remarkable success in various tasks, such as node/graph classification, link prediction, etc.

It has been demonstrated that adversarial graphs, i.e., graphs with imperceptible perturbations added, can cause deep graph models to fail on node/graph classification tasks. In this paper, we extend adversarial graphs to the problem of community detection which is much more difficult. We focus on black-box attack and aim to hide targeted individuals from the detection of deep graph community detection models, which has many applications in real-world scenarios, for example, protecting personal privacy in social networks and understanding camouflage patterns in transaction networks. We propose an iterative learning framework that takes turns to update two modules: one working as the constrained graph generator and the other as the surrogate community detection model. We also find that the adversarial graphs generated by our method can be transferred to other learning based community detection models.

Searching similar graphs in graph databases for a query graph has attracted extensive attention recently. Existing works on graph similarity queries are threshold based approaches which return graphs with distances to the query smaller than a given threshold. However, in many applications the number of answer graphs for the same threshold can vary significantly for different queries. In this paper, we study the problem of finding top-k most similar graphs for a query under the distance measure based on maximum common subgraph (MCS). Since computing MCS is NP-hard, we devise a novel framework to prune unqualified graphs based on the lower bounds of graph distance, and accordingly derive four lower bounds with different tightness and computational cost for pruning. To further reduce the number of MCS computations, we also propose an improved framework based on both lower and upper bounds, and derive three new upper bounds. To support efficient pruning, we design three indexes with different tradeoffs between pruning power and construction cost. To accelerate the index construction, we explore bound relaxation techniques, based on which approximate indexes can be efficiently built. We conducted extensive performance studies on real-life graph datasets to validate the effectiveness and efficiency of our approaches.

As a fundamental NP-hard problem in graph theory, the maximum independent set (MIS) has attracted a lot of efforts to improve the time efficiency. However, most graphs in real scenarios are usually changing over time. But the previous studies take the stationary graphs as input, the computation of MIS in dynamic graphs receives little attention. Since computing the exact MIS is intractable, we compute the high-quality (large-size) independent set for dynamic graphs in this paper, where 4 graph updating operations are allowed: adding or deleting a vertex or an edge. Based on two state-of-the-art reduction rules that are designed for static graphs, we propose a novel scheme, i.e., dependency graph based independent set computation, which can support computing the high-quality independent set on the basis of the previous result rather than calculating from scratch. Moreover, a dynamic searching strategy is devised to improve time efficiency. In order to make it more useful in practical applications, we devise an effective yet efficient method to deal with the batch update. To confirm the effectiveness and efficiency of the proposed methods, we conduct extensive experiments over both real and synthetic datasets.

In this paper we use a time-evolving graph which consists of a sequence of graph snapshots over time to model many real-world networks. We study the path classification problem in a time-evolving graph, which has many applications in real-world scenarios, for example, predicting path failure in a telecommunication network and predicting path congestion in a traffic network in the near future. In order to capture the temporal dependency and graph structure dynamics, we design a novel deep neural network named Long Short-Term Memory R-GCN (LRGCN). LRGCN considers temporal dependency between time-adjacent graph snapshots as a special relation with memory, and uses relational GCN to jointly process both intra-time and inter-time relations. We also propose a new path representation method named self-attentive path embedding (SAPE), to embed paths of arbitrary length into fixed-length vectors. Through experiments on a real-world telecommunication network and a traffic network in California, we demonstrate the superiority of LRGCN to other competing methods in path failure prediction, and prove the effectiveness of SAPE on path representation.

Node classification and graph classification are two graph learning problems that predict the class label of a node and the class label of a graph respectively. A node of a graph usually represents a real-world entity, e.g., a user in a social network, or a protein in a protein-protein interaction network. In this work, we consider a more challenging but practically useful setting, in which a node itself is a graph instance. This leads to a hierarchical graph perspective which arises in many domains such as social network, biological network and document collection. For example, in a social network, a group of people with shared interests forms a user group, whereas a number of user groups are interconnected via interactions or common members. We study the node classification problem in the hierarchical graph where a “node” is a graph instance, e.g., a user group in the above example. As labels are usually limited in real-world data, we design two novel semi-supervised solutions named SEmi-supervised grAph cLassification via Cautious/Active Iteration (or SEAL-C/AI in short). SEAL-C/AI adopt an iterative framework that takes turns to build or update two classifiers, one working at the graph instance level and the other at the hierarchical graph level. To simplify the representation of the hierarchical graph, we propose a novel supervised, self-attentive graph embedding method called SAGE, which embeds graph instances of arbitrary size into fixed-length vectors. Through experiments on synthetic data and Tencent QQ group data, we demonstrate that SEAL-C/AI not only outperform competing methods by a significant margin in terms of accuracy/Macro-F1, but also generate meaningful interpretations of the learned representations.

Given a query vertex in a graph, the task of community search is to find all meaningful communities containing the query vertex in an online manner. In this demonstration, we propose a novel query processing system for searching and visualizing communities in graphs, called VizCS. It exhibits three key innovative features. First, VizCS adopts several community models and supports community search on dynamic graphs where nodes/edges undergo frequently insertions/deletions. Second, VizCS offers a user-friendly visual interface to formulate queries and a real-time response query processing engine. Last but not least, VizCS generates a community exploration wall by offering interactive community visualization, which facilitates users to in-depth understanding of the data. Furthermore, VizCS becomes a community search platform that can visualize and compare different community results by various state-of-the-art algorithms and user-uploaded approaches.

Most existing algorithms computing the maximum independent set (MIS) or independent set (IS) are designed for handling static graphs, which may not be practicable as many networks are dynamically evolving over time. In this paper, we study the MIS/IS problem in evolving graphs by considering graph update operations: vertex/edge addition and vertex/edge deletion. Instead of computing the MIS/IS of the updated graph from scratch, we propose a baseline algorithm that finds the MIS/IS at time t i+1 based on the MIS/IS at time ti. Due to the hardness of computing an exact MIS, we develop an efficient constant-time algorithm LSTwo to return a high-quality (large-size) independent set. Then we design a lazy search algorithm which produces higher-quality independent sets. To improve the time efficiency further, we devise the conditional besieging and k-petal based methods to reduce the search space. Extensive experimental studies over large-scale graphs confirm the effectiveness and efficiency of our proposed algorithms.

With the increase of elderly population, Alzheimer's Disease (AD), as the most common cause of dementia among the elderly, is affecting more and more senior people. It is crucial for a patient to receive accurate and timely diagnosis of AD. Current diagnosis relies on doctors' experience and clinical test, which, unfortunately, may not be performed until noticeable AD symptoms are developed. In this work, we present our novel solution named time-aware TICC and CNN (TATC), for predicting AD from actigraphy data. TATC is a multivariate time series classification method using a neural attention-based deep learning approach. It not only performs accurate prediction of AD risk, but also generates meaningful interpretation of daily behavior pattern of subjects. TATC provides an automatic, low-cost solution for continuously monitoring the change of physical activity of subjects in daily living environment. We believe the future deployment of TATC can benefit both doctors and patients in early detection of potential AD risk.

The gap between unstructured natural language and structured data makes it challenging to build a system that supports using natural language to query large knowledge graphs. Many existing methods construct a structured query for the input question based on a syntactic parser. Once the input question is parsed incorrectly, a false structured query will be generated, which may result in false or incomplete answers. The problem gets worse especially for complex questions. In this paper, we propose a novel systematic method to understand natural language questions by using a large number of binary templates rather than semantic parsers. As sufficient templates are critical in the procedure, we present a low-cost approach that can build a huge number of templates automatically. To reduce the search space, we carefully devise an index to facilitate the online template decomposition. Moreover, we design effective strategies to perform the two-level disambiguations (i.e., entity-level ambiguity and structure-level ambiguity) by considering the query semantics. Extensive experiments over several benchmarks demonstrate that our proposed approach is effective as it significantly outperforms state-of-the-art methods in terms of both precision and recall.

Approximate nearest neighbor (ANN) search in high-dimensional space plays an essential role in many multimedia applications. Recently, product quantization (PQ) based methods for ANN search have attracted enormous attention in the community of computer vision, due to its good balance between accuracy and space requirement. PQ based methods embed a high-dimensional vector into a short binary code (called PQ code), and the squared Euclidean distance is estimated by asymmetric quantizer distance (AQD) with pretty high precision. Thus, ANN search in the original space can be converted to similarity search on AQD using the PQ approach. All existing PQ methods are in-memory solutions, which may not handle massive data if they cannot fit entirely in memory. In this paper, we propose an I/O-efficient PQ based solution for ANN search. We design an index called PQB+-forest to support efficient similarity search on AQD. PQB+-forest first creates a number of partitions of the PQ codes by a coarse quantizer and then builds a B+-tree, called PQB+-tree, for each partition. The search process is greatly expedited by focusing on a few selected partitions that are closest to the query, as well as by the pruning power of PQB+-trees. According to the experiments conducted on two large-scale data sets containing up to 1 billion vectors, our method outperforms its competitors, including the state-of-the-art PQ method and the state-of-the-art LSH methods for ANN search.

In recent years, modeling the relation between two graphs has received unprecedented attention from researchers due to its wide applications in many areas, such as social analysis and bioinformatics. The nature of relations between two graphs can be divided into two categories: the vertex relation and the link relation. Many studies focus on modeling the vertex relation between graphs and try to find the vertex correspondence between two graphs. However, the link relation between graphs has not been fully studied. Specifically, we model the cross-graph link relation as cross-graph dependence, which reflects the dependence of a vertex in one graph on a vertex in the other graph. A generic problem, called Graph Dependence Minimization (GDM), is defined as: given two graphs with cross-graph dependence, how to select a subset of vertexes from one graph and copy them to the other, so as to minimize the cross-graph dependence. Many real applications can benefit from the solution to GDM. Examples include reducing the cross-language links in online encyclopedias, optimizing the cross-platform communication cost between different cloud services, and so on. This problem is trivial if we can select as many vertexes as we want to copy. But what if we can only choose a limited number of vertexes to copy so as to make the two graphs as independent as possible? We formulate GDM with a budget constraint into a combinatorial optimization problem, which is proven to be NP-hard. We propose two algorithms to solve GDM. Firstly, we prove the submodularity of the objective function of GDM and adopt the size-constrained submodular minimization (SSM) algorithm to solve it. Since the SSM-based algorithm cannot scale to large graphs, we design a heuristic algorithm with a provable approximation guarantee. We prove that the error achieved by the heuristic algorithm is bounded by an additive factor which is proportional to the square of the given budget. Extensive experiments on both synthetic and real-world graphs show that the proposed algorithms consistently outperform the well-studied graph centrality measure based solutions. Furthermore, we conduct a case study on the Wikipedia graphs with millions of vertexes and links to demonstrate the potential of GDM to solve real-world problems.

The ever-increasing knowledge graphs impose an urgent demand of providing effective and easy-to-use query techniques for end users. Structured query languages, such as SPARQL, offer a powerful expression ability to query RDF datasets. However, they are difficult to use. Keywords are simple but have a very limited expression ability. Natural language question (NLQ) is promising on querying knowledge graphs. A huge challenge is how to understand the question clearly so as to translate the unstructured question into a structured query. In this paper, we present a data + oracle approach to answer NLQs over knowledge graphs. We let users verify the ambiguities during the query understanding. To reduce the interaction cost, we formalize an interaction problem and design an efficient strategy to solve the problem. We also propose a query prefetch technique by exploiting the latency in the interactions with users. Extensive experiments over the QALD dataset demonstrate that our proposed approach is effective as it outperforms state-of-the-art methods in terms of both precision and recall.

With the advent of cellular network technologies, mobile Internet access becomes the norm in everyday life. In the meantime, the complaints made by subscribers about unsatisfactory cellular network access also become increasingly frequent. From a network operator's perspective, achieving accurate and timely cellular network diagnosis about the causes of the complaints is critical for both improving subscriber-perceived experience and maintaining network robustness. We present the Intelligent Customer Care Assistant (ICCA), a distributed fault classification system that exploits a data-driven approach to perform large-scale cellular network diagnosis. ICCA takes massive network data as input, and realizes both offline model training and online feature computation to distinguish between user and network faults in real time. ICCA is currently deployed in a metropolitan LTE network in China that is serving around 50 million subscribers. We show via evaluation that ICCA achieves high classification accuracy (85.3%) and fast query response time (less than 2.3 seconds). We also report our experiences learned from the deployment.

Extracting keyphrases from documents automatically is an important and interesting task since keyphrases provide a quick summarization for documents. Although lots of efforts have been made on keyphrase extraction, most of the existing methods (the co-occurrence-based methods and the statistic-based methods) do not take semantics into full consideration. The co-occurrence-based methods heavily depend on the co-occurrence relations between two words in the input document, which may ignore many semantic relations. The statistic-based methods exploit the external text corpus to enrich the document, which introduce more unrelated relations inevitably. In this paper, we propose a novel approach to extract keyphrases using knowledge graphs, based on which we could detect the latent relations of two keyterms (i.e., noun words and named entities) without introducing many noises. Extensive experiments over real data show that our method outperforms the state-of-the-art methods including the graph-based co-occurrence methods and statistic-based clustering methods.

Subgraph matching finds a set I of all occurrences of a pattern graph in a target graph. It has a wide range of applications while suffers an expensive computation. This efficiency issue has been studied extensively. All existing approaches, however, turn a blind eye to the output crisis , that is, when the system has to materialize I as a preprocessing/intermediate/final result or an index, the cost of the export of I dominates the overall cost, which could be prohibitive even for a small pattern graph. This paper studies subgraph matching via two problems. 1) Is there an ideal compression of I ? 2) Will the compression of I reversely boost the computation of I ? For the problem 1), we propose a technique called VCBC to compress I to code ( I ) which serves effectively the same as I. For problem 2), we propose a subgraph matching computation framework CBF which computes code( I )instead of I to bring down the output cost. CBF further reduces the overall cost by reducing the intermediate results. Extensive experiments show that the compression ratio of VCBC can be up to 10 5 which also significantly lowers the output cost of CBF. Extensive experiments show the superior performance of CBF over existing approaches.

Matching dependencies (MDs) have recently been proposed to make data dependencies tolerant to various information representations, and found useful in data quality applications such as record matching. Instead of the strict equality function used in traditional dependency syntax (e.g., functional dependencies), MDs specify constraints based on similarity and identification. However, in practice, MDs may still be too strict and applicable only in a subset of tuples in a relation. Thereby, we study the conditional matching dependencies (CMDs), which bind matching dependencies only in a certain part of a table, i.e., MDs conditionally applicable in a subset of tuples. Compared to MDs, CMDs have more expressive power that enables them to satisfy wider application needs. In this article, we study several important theoretical and practical issues of CMDs, including irreducible CMDs with respect to the implication, discovery of CMDs from data, reliable CMDs agreed most by a relation, approximate CMDs almost satisfied in a relation, and finally applications of CMDs in record matching and missing value repairing. Through an extensive experimental evaluation in real data sets, we demonstrate the efficiency of proposed CMDs discovery algorithms and effectiveness of CMDs in real applications.

This paper addresses the classical triangle listing problem, which aims at enumerating all the tuples of three vertices connected with each other by edges. This problem has been intensively studied in internal and external memory, but it is still an urgent challenge in distributed environment where multiple machines across the network can be utilized to achieve good performance and scalability. As one of the de facto computing methodologies in distributed environment, MapReduce has been used in some of existing triangle listing algorithms. However, these algorithms usually need to shuffle a huge amount of intermediate data, which seriously hinders the scalability on large scale graphs. In this paper, we propose a new triangle listing algorithm in MapReduce, FTL, which utilizes a light weight data structure to substantially reduce the intermediate data transferred during the shuffle stage, and also is equipped with multiple-round techniques to ease the burden on memory and network bandwidth when dealing with graphs at billion scale. We prove that the size of the intermediate data can be well bounded near to the number of triangles in the graph. To further reduce the shuffle size in each round, we also devise a compact data structure to store the intermediate data, which can save space up to 2/3. The extensive experimental results show that our algorithms outperform existing competitors by several times on large real world graphs.

With the soaring popularity of online social media like Twitter, analyzing short text has emerged as an increasingly important task which is challenging to classical topic models, as topic sparsity exists in short text. Topic sparsity refers to the observation that individual document usually concentrates on several salient topics, which may be rare in entire corpus. Understanding this sparse topical structure of short text has been recognized as the key ingredient for mining user-generated Web content and social medium, which are featured in the form of extremely short posts and discussions. However, the existing sparsity-enhanced topic models all assume over-complicated generative process, which severely limits their scalability and makes them unable to automatically infer the number of topics from data.

Information diffusion through various types of networks, such as social networks and media networks, is a very common phenomenon on the Internet nowadays. In many scenarios, we can track only the time when the information reaches a node. However, the source infecting this node is usually unobserved. Inferring the underlying diffusion network based on cascade data (observed sequence of infected nodes with timestamp) without additional information is an essential and challenging task in information diffusion. Many studies have focused on constructing complex models to infer the underlying diffusion network in a parametric way. However, the diffusion process in the real world is very complex and hard to be captured by a parametric model. Even worse, inferring the parameters of a complex model is impractical under a large data volume.

Data imputation aims at filling in missing attribute values in databases. Existing imputation approaches to nonquantitive string data can be roughly put into two categories: (1) inferring-based approaches [2], and (2) retrieving-based approaches [1]. Specifically, the inferring-based approaches find substitutes or estimations for the missing ones from the complete part of the data set. However, they typically fall short in filling in unique missing attribute values which do not exist in the complete part of the data set [1]. The retrieving-based approaches resort to external resources for help by formulating proper web search queries to retrieve web pages containing the missing values from the Web, and then extracting the missing values from the retrieved web pages [1]. This webbased retrieving approach reaches a high imputation precision and recall, but on the other hand, issues a large number of web search queries, which brings a large overhead [1].

Large graphs are getting increasingly popular and even indispensable in many applications, for example, in social media data, large networks, and knowledge bases. Efficient graph analytics thus becomes an important subject of study. To increase efficiency and scalability, in-memory computation and parallelism have been explored extensively to speed up various graph analytical workloads. In many graph analytical engines (e.g., Pregel, Neo4j, GraphLab), parallelism is achieved via one of the three concurrency control models, namely, bulk synchronization processing (BSP), asynchronous processing, and synchronous processing. Among them, synchronous processing has the potential to achieve the best performance due to fine-grained parallelism, while ensuring the correctness and the convergence of the computation, if an effective concurrency control scheme is used. This paper explores the topological properties of the underlying graph to design and implement a highly effective concurrency control scheme for efficient synchronous processing in an in-memory graph analytical engine. Our design uses a novel hybrid approach that combines 2PL (two-phase locking) with OCC (optimistic concurrency control), for high degree and low degree vertices in a graph respectively. Our results show that the proposed hybrid synchronous scheduler has significantly outperformed other synchronous schedulers in existing graph analytical engines, as well as BSP and asynchronous schedulers.

Due to the large number of requirements for city express services in recent years, the current city express system is found to be unsatisfactory for both the service providers and customers. In this paper, we are the first to systematically study the large-scale dynamic city express problem. We aim to increase both the effectiveness and the efficiency of the scheduling algorithm. The challenges of the problem stem from the highly dynamic environment, the NP-completeness with respect to the number of requests, and real-time demands for the scheduling result. We introduce a basic algorithm to assign a request to a courier on a first-come, first-served basis. To improve the effectiveness of the basic algorithm, we adopt a batch assignment strategy that computes the pickup-delivery routes for a group of requests received in a short period rather than dealing with each request individually. To improve the efficiency of the algorithm, we further design a two-level priority queue structure to reduce redundant shortest distance calculation and repeated candidate generation. We develop a simulation system and conduct extensive performance studies on the real road network of Beijing city. The experimental results demonstrate the high effectiveness and efficiency of our algorithms. Remarkably, our system can achieve much better service quality and largely reduce the operation cost of a city express company simultaneously.

City express services are in great demand in recent years. However, the current city express system is found to be unsatisfactory for both the service providers and customers. In this paper, we are the first to systematically study the large-scale dynamic city express problem. We aim to increase both the effectiveness and the efficiency of the scheduling algorithm. To improve the effectiveness, we adopt a batch assignment strategy that computes the pickup-delivery routes for a group of requests received in a short period rather than dealing with each request individually. To improve the efficiency, we design a two-level priority queue structure to reduce redundant shortest distance calculation and repeated candidate generation. We develop a simulation system and conduct extensive performance studies in the real road network of Beijing city. The experimental results demonstrate the high effectiveness and efficiency of our algorithm.

In nowadays social networks, a huge volume of content containing rich information, such as reviews, ratings, microblogs, etc., is being generated, consumed and diffused by users all the time. Given the temporal information, we can obtain the event cascade which indicates the time sequence of the arrival of information to users. Many models have been proposed to explain how information diffuses. However, most existing models cannot give a clear explanation why every specific event happens in the event cascade. Such explanation is essential for us to have a deeper understanding of information diffusion as well as a better prediction of future event cascade.

Recently, there has been significant interest in the study of the community search problem in social and information networks: given one or more query nodes, find densely connected communities containing the query nodes. However, most existing studies do not address the "free rider" issue, that is, nodes far away from query nodes and irrelevant to them are included in the detected community. Some state-of-the-art models have attempted to address this issue, but not only are their formulated problems NP-hard, they do not admit any approximations without restrictive assumptions, which may not always hold in practice. In this paper, given an undirected graph G and a set of query nodes Q , we study community search using the k -truss based community model. We formulate our problem of finding a closest truss community (CTC), as finding a connected k -truss subgraph with the largest k that contains Q , and has the minimum diameter among such subgraphs. We prove this problem is NP-hard. Furthermore, it is NP-hard to approximate the problem within a factor (2-ε), for any ε > 0. However, we develop a greedy algorithmic framework, which first finds a CTC containing Q , and then iteratively removes the furthest nodes from Q , from the graph. The method achieves 2-approximation to the optimal solution. To further improve the efficiency, we make use of a compact truss index and develop efficient algorithms for k -truss identification and maintenance as nodes get eliminated. In addition, using bulk deletion optimization and local exploration strategies, we propose two more efficient algorithms. One of them trades some approximation quality for efficiency while the other is a very efficient heuristic. Extensive experiments on 6 real-world networks show the effectiveness and efficiency of our community model and search algorithms.

Given two sets of vertices in a graph, it is often of a great interest to find out how these vertices are connected, especially to identify the vertices of high prominence defined on the topological structure. In this work, we formally define a V ertex S et B onding query (shorted as VSB), which returns a minimum set of vertices with the maximum importance w.r.t total betweenness and shortest path reachability in connecting two sets of input vertices. We find that such a kind of query is representative and could be widely applied in many real world scenarios, e.g., logistic planning, social community bonding and etc. Challenges are that many of such applications are constructed on graphs that are too large to fit in single server, and the VSB query evaluation turns to be NP-hard. To cope with the scalability issue and return the near optimal result in almost real time, we propose a generic solution framework on a shared nothing distributed environment. With the development of two novel techniques, guided graph exploration and betweenness ranking on exploration, we are able to efficiently evaluate queries for error bounded results with bounded space cost. We demonstrate the effectiveness of our solution with extensive experiments over both real and synthetic large graphs on the Google's Cloud platform. Comparing to the exploration only baseline method, our method achieves several times of speedup.

The problem of mobile sequential recommendation is to suggest a route connecting a set of pick-up points for a taxi driver so that he/she is more likely to get passengers with less travel cost. Essentially, a key challenge of this problem is its high computational complexity. In this paper, we propose a novel dynamic programming based method to solve the mobile sequential recommendation problem consisting of two separate stages: an offline pre-processing stage and an online search stage. The offline stage pre-computes potential candidate sequences from a set of pick-up points. A backward incremental sequence generation algorithm is proposed based on the identified iterative property of the cost function. Simultaneously, an incremental pruning policy is adopted in the process of sequence generation to reduce the search space of the potential sequences effectively. In addition, a batch pruning algorithm is further applied to the generated potential sequences to remove some non-optimal sequences of a given length. Since the pruning effectiveness keeps growing with the increase of the sequence length, at the online stage, our method can efficiently find the optimal driving route for an unloaded taxi in the remaining candidate sequences. Moreover, our method can handle the problem of optimal route search with a maximum cruising distance or a destination constraint. Experimental results on real and synthetic data sets show that both the pruning ability and the efficiency of our method surpass the state-of-the-art methods. Our techniques can therefore be effectively employed to address the problem of mobile sequential recommendation with many pick-up points in real-world applications.

Data imputation aims at filling in missing attribute values in databases. Most existing imputation methods to string attribute values are inferring-based approaches, which usually fail to reach a high imputation recall by just inferring missing values from the complete part of the data set. Recently, some retrieving-based methods are proposed to retrieve missing values from external resources such as the World Wide Web, which tend to reach a much higher imputation recall, but inevitably bring a large overhead by issuing a large number of search queries. In this paper, we investigate the interaction between the inferring-based methods and the retrieving-based methods. We show that retrieving a small number of selected missing values can greatly improve the imputation recall of the inferring-based methods. With this intuition, we propose an inTeractive Retrieving-Inferring data imPutation approach (TRIP), which performs retrieving and inferring alternately in filling in missing attribute values in a data set. To ensure the high recall at the minimum cost, TRIP faces a challenge of selecting the least number of missing values for retrieving to maximize the number of inferable values. Our proposed solution is able to identify an optimal retrieving-inferring scheduling scheme in deterministic data imputation, and the optimality of the generated scheme is theoretically analyzed with proofs. We also analyze with an example that the optimal scheme is not feasible to be achieved in τ-constrained stochastic data imputation (τ-SDI), but still, our proposed solution identifies an expected-optimal scheme in τ-SDI. Extensive experiments on four data collections show that TRIP retrieves on average 20 percent missing values and achieves the same high recall that was reached by the retrieving-based approach.

Recovering matrices from incomplete and corrupted observations is a fundamental problem with many applications in various areas of science and engineering. In theory, under certain conditions, this problem can be solved via a natural convex relaxation. However, all current provable algorithms suffer from superlinear per-iteration cost, which severely limits their applicability to large scale problems. In this paper, we propose a robust principal component analysis (RPCA) plus matrix completion framework to recover low-rank and sparse matrices from missing and grossly corrupted observations. Under the unified framework, we first present a convex robust matrix completion model to replace the linear projection operator constraint by a simple equality one. To further improve the efficiency of our convex model, we also develop a scalable structured factorization model, which can yield an orthogonal dictionary and a robust data representation simultaneously. Then, we develop two alternating direction augmented Lagrangian (ADAL) algorithms to efficiently solve the proposed problems. Finally, we discuss the convergence analysis of our algorithms. Experimental results verified both the efficiency and effectiveness of our methods compared with the state-of-the-art algorithms.

We investigate latent aspect mining problem that aims at automatically discovering aspect information from a collection of review texts in a domain in an unsupervised manner. One goal is to discover a set of aspects which are previously unknown for the domain, and predict the user's ratings on each aspect for each review. Another goal is to detect key terms for each aspect. Existing works on predicting aspect ratings fail to handle the aspect sparsity problem in the review texts leading to unreliable prediction. We propose a new generative model to tackle the latent aspect mining problem in an unsupervised manner. By considering the user and item side information of review texts, we introduce two latent variables, namely, user intrinsic aspect interest and item intrinsic aspect quality facilitating better modeling of aspect generation leading to improvement on the accuracy and reliability of predicted aspect ratings. Furthermore, we provide an analytical investigation on the Maximum A Posterior (MAP) optimization problem used in our proposed model and develop a new block coordinate gradient descent algorithm to efficiently solve the optimization with closed-form updating formulas. We also study its convergence analysis. Experimental results on the two real-world product review corpora demonstrate that our proposed model outperforms existing state-of-the-art models.

We introduce and formulate two types of random-walk domination problems in graphs motivated by a number of applications in practice (e.g., item-placement problem in online social networks, Ads-placement problem in advertisement networks, and resource-placement problem in P2P networks). Specifically, given a graph G, the goal of the first type of random-walk domination problem is to target k nodes such that the total hitting time of an L-length random walk starting from the remaining nodes to the targeted nodes is minimized. The second type of random-walk domination problem is to find k nodes to maximize the expected number of nodes that hit any one targeted node through an L-length random walk. We prove that these problems are two special instances of the submodular set function maximization with cardinality constraint problem. To solve them effectively, we propose a dynamic-programming (DP) based greedy algorithm which is with near-optimal performance guarantee. The DP-based greedy algorithm, however, is not very efficient due to the expensive marginal gain evaluation. To further speed up the algorithm, we propose an approximate greedy algorithm with linear time complexity w.r.t. the graph size and also with near-optimal performance guarantee. The approximate greedy algorithm is based on carefully designed random walk sampling and sample-materialization techniques. Extensive experiments demonstrate the effectiveness, efficiency and scalability of the proposed algorithms.

Recovering low-rank and sparse matrices from partial, incomplete or corrupted observations is an important problem in many areas of science and engineering. In this paper, we propose a scalable robust bilateral factorization (RBF) method to recover both structured matrices from missing and grossly corrupted data such as robust matrix completion (RMC), or incomplete and grossly corrupted measurements such as compressive principal component pursuit (CPCP). With the unified framework, we first present two robust trace norm regularized bilateral factorization models for RMC and CPCP problems, which can achieve an orthogonal dictionary and a robust data representation, simultaneously. Then, we apply the alternating direction method of multipliers to efficiently solve the RMC problems. Finally, we provide the convergence analysis of our algorithm, and extend it to address general CPCP problems. Experimental results verified both the efficiency and effectiveness of our RBF method compared with the state-of-the-art methods.

Most existing low-n-rank minimization algorithms for tensor completion suffer from high computational cost due to involving multiple singular value decompositions (SVDs) at each iteration. To address this issue, we propose a novel factor matrix trace norm minimization method for tensor completion problems. Based on the CANDECOMP/PARAFAC (CP) decomposition, we first formulate a factor matrix rank minimization model by deducing the relation between the rank of each factor matrix and the mode-n rank of a tensor. Then, we introduce a tractable relaxation of our rank function, which leads to a convex combination problem of much smaller scale matrix nuclear norm minimization. Finally, we develop an efficient alternating direction method of multipliers (ADMM) scheme to solve the proposed problem. Experimental results on both synthetic and real-world data validate the effectiveness of our approach. Moreover, our method is significantly faster than the state-of-the-art approaches and scales well to handle large datasets.

Community detection which discovers densely connected structures in a network has been studied a lot. In this paper, we study online community search which is practically useful but less studied in the literature. Given a query vertex in a graph, the problem is to find meaningful communities that the vertex belongs to in an online manner. We propose a novel community model based on the k-truss concept, which brings nice structural and computational properties. We design a compact and elegant index structure which supports the efficient search of k-truss communities with a linear cost with respect to the community size. In addition, we investigate the k-truss community search problem in a dynamic graph setting with frequent insertions and deletions of graph vertices and edges. Extensive experiments on large real-world networks demonstrate the effectiveness and efficiency of our community model and search algorithms.

MapReduce has become one of the most popular parallel computing paradigms in cloud, due to its high scalability, reliability, and fault-tolerance achieved for a large variety of applications in big data processing. In the literature, there are MapReduce Class MRC and Minimal MapReduce Class MMC to define the memory consumption, communication cost, CPU cost, and number of MapReduce rounds for an algorithm to execute in MapReduce. However, neither of them is designed for big graph processing in MapReduce, since the constraints in MMC can be hardly achieved simultaneously on graphs and the conditions in MRC may induce scalability problems when processing big graph data. In this paper, we study scalable big graph processing in MapReduce. We introduce a Scalable Graph processing Class SGC by relaxing some constraints in MMC to make it suitable for scalable graph processing. We define two graph join operators in SGC, namely, EN join and NE join, using which a wide range of graph algorithms can be designed, including PageRank, breadth first search, graph keyword search, Connected Component (CC) computation, and Minimum Spanning Forest (MSF) computation. Remarkably, to the best of our knowledge, for the two fundamental graph problems CC and MSF computation, this is the first work that can achieve O(log(n)) MapReduce rounds with $O(n+m)$ total communication cost in each round and constant memory consumption on each machine, where $n$ and $m$ are the number of nodes and edges in the graph respectively. We conducted extensive performance studies using two web-scale graphs Twitter and Friendster with different graph characteristics. The experimental results demonstrate that our algorithms can achieve high scalability in big graph processing.

Topic modeling has been proved to be an effective method for exploratory text mining. It is a common assumption of most topic models that a document is generated from a mixture of topics. In real-world scenarios, individual documents usually concentrate on several salient topics instead of covering a wide variety of topics. A real topic also adopts a narrow range of terms instead of a wide coverage of the vocabulary. Understanding this sparsity of information is especially important for analyzing user-generated Web content and social media, which are featured as extremely short posts and condensed discussions. In this paper, we propose a dual-sparse topic model that addresses the sparsity in both the topic mixtures and the word usage. By applying a "Spike and Slab" prior to decouple the sparsity and smoothness of the document-topic and topic-word distributions, we allow individual documents to select a few focused topics and a topic to select focused terms, respectively. Experiments on different genres of large corpora demonstrate that the dual-sparse topic model outperforms both classical topic models and existing sparsity-enhanced topic models. This improvement is especially notable on collections of short documents.

Recommendation systems have been widely used in E-commerce sites, social networks, etc. One of the core tasks in recommendation systems is to predict the users' ratings on items. Although many models and algorithms have been proposed, how to make accurate prediction for new users with extremely few rating records still remains a big challenge, which is called the cold start problem. Many existing methods utilize additional information, such as social graphs, to cope with the cold start problem. However, the side information may not always be available. In contrast to such methods, we propose a more general solution to address the cold start problem based on the observed user rating records only. Specifically we define a random walk on a bipartite graph of users and items to simulate the preference propagation among users, in order to alleviate the data sparsity problem for cold start users. Then we propose a Monte Carlo algorithm to estimate the similarity between different users. This algorithm takes a precomputation approach, and thus can efficiently compute the user similarity given any new user for rating prediction. In addition, our algorithm can easily handle dynamic updates and can be parallelized naturally, which are crucial for large recommendation systems. Theoretical analysis is presented to demonstrate the efficiency and effectiveness of our algorithm, and extensive experiments also confirm our theoretical findings.

Matching keys, specifying what attributes to compare and how to compare them for identifying the same real-world entities, are found to be useful in applications like record matching, blocking and windowing [7]. Owing to the complex redundant semantics among matching keys, capturing a proper set of matching keys is highly non-trivial. Analogous to minimal/candidate keys w.r.t. functional dependencies, relative candidate keys (RCKs [7], with a minimal number of compared attributes, see a more formal definition in Section 2) can clear up redundant semantics w.r.t. "what attributes to compare". However, we note that redundancy issues may still exist among rcks on the same attributes about "how to compare them". In this paper, we propose to find a concise set of matching keys, which has less redundancy and can still meet the requirements on coverage and validity. Specifically, we study approximation algorithms to efficiently discover a near optimal set. To ensure the quality of matching keys, the returned results are guaranteed to be RCKs (minimal on compared attributes), and most importantly, minimal w.r.t. distance restrictions (i.e., redundancy free w.r.t. "how to compare the attributes"). The experimental evaluation demonstrates that our concise RCK set is more effective than the existing rck choosing method. Moreover, the proposed pruning methods show up to 2 orders of magnitude improvement w.r.t. time costs on concise RCK set discovery.

A broad class of data, ranging from similarity networks, workflow networks to protein networks, can be modeled as graphs with data values as vertex labels. The vertex labels (data values) are often dirty for various reasons such as typos or erroneous reporting of results in scientific experiments. Neighborhood constraints , specifying label pairs that are allowed to appear on adjacent vertexes in the graph, are employed to detect and repair erroneous vertex labels. In this paper, we study the problem of repairing vertex labels to make graphs satisfy neighborhood constraints. Unfortunately, the relabeling problem is proved to be NP hard, which motivates us to devise approximation methods for repairing, and identify interesting special cases (star and clique constraints) that can be efficiently solved. We propose several approximate repairing algorithms including greedy heuristics, contraction method and a hybrid approach. The performances of algorithms are also analyzed for the special case. Our extensive experimental evaluation, on both synthetic and real data, demonstrates the effectiveness of eliminating frauds in several types of application networks. Remarkably, the hybrid method performs well in practice, i.e., guarantees termination, while achieving high effectiveness at the same time.

Graph clustering, also known as community detection, is a long-standing problem in data mining. In recent years, with the proliferation of rich attribute information available for objects in real-world graphs, how to leverage not only structural but also attribute information for clustering attributed graphs becomes a new challenge. Most existing works took a distance-based approach. They proposed various distance measures to fuse structural and attribute information and then applied standard techniques for graph clustering based on these distance measures. In this article, we take an alternative view and propose a novel Bayesian framework for attributed graph clustering. Our framework provides a general and principled solution to modeling both the structural and the attribute aspects of a graph. It avoids the artificial design of a distance measure in existing methods and, furthermore, can seamlessly handle graphs with different types of edges and vertex attributes. We develop an efficient variational method for graph clustering under this framework and derive two concrete algorithms for clustering unweighted and weighted attributed graphs. Experimental results on large real-world datasets show that our algorithms significantly outperform the state-of-the-art distance-based method, in terms of both effectiveness and efficiency.

Shortest distance query is a fundamental operation in large-scale networks. Many existing methods in the literature take a landmark embedding approach, which selects a set of graph nodes as landmarks and computes the shortest distances from each landmark to all nodes as an embedding. To answer a shortest distance query, the precomputed distances from the landmarks to the two query nodes are used to compute an approximate shortest distance based on the triangle inequality. In this paper, we analyze the factors that affect the accuracy of distance estimation in landmark embedding. In particular, we find that a globally selected, query-independent landmark set may introduce a large relative error, especially for nearby query nodes. To address this issue, we propose a query-dependent local landmark scheme, which identifies a local landmark close to both query nodes and provides more accurate distance estimation than the traditional global landmark approach. We propose efficient local landmark indexing and retrieval techniques, which achieve low offline indexing complexity and online query complexity. Two optimization techniques on graph compression and graph online search are also proposed, with the goal of further reducing index size and improving query accuracy. Furthermore, the challenge of immense graphs whose index may not fit in the memory leads us to store the embedding in relational database, so that a query of the local landmark scheme can be expressed with relational operators. Effective indexing and query optimization mechanisms are designed in this context. Our experimental results on large-scale social networks and road networks demonstrate that the local landmark scheme reduces the shortest distance estimation error significantly when compared with global landmark embedding and the state-of-the-art sketch-based embedding.

The importance of introducing distance constraints to data dependencies, such as differential dependencies (DDs), has recently been recognized. The differential dependencies are tolerant to small variations, which enable them to apply to wide data quality checking applications, such as detecting data violations. However, the determination of distance thresholds for the differential dependencies is non-trivial. It often relies on a truth data instance which embeds the distance constraints. To find useful distance threshold patterns from data, there are several guidelines of statistical measures to specify, e.g., support, confidence and dependent quality. Unfortunately, given a data instance, users might not have any knowledge about the data distribution, thus it is very challenging to set the right parameters. In this paper, we study the determination of distance thresholds for differential dependencies, in a parameter-free style. Specifically, we compute an expected utility based on the statistical measures from the data. According to our analysis as well as experimental verification, distance threshold patterns with higher expected utility could offer better use in real applications, such as violation detection. We then develop efficient algorithms to determine the distance thresholds having the maximum expected utility. Finally, our extensive experimental evaluation demonstrates the effectiveness and efficiency of the proposed methods.

Supported by the technical advances and the commercial success of GPS-enabled mobile devices, geo-tagged photos have drawn plenteous attention in research community. The explosive growth of geo-tagged photos enables many large-scale applications, such as location-based photo browsing, landmark recognition, etc. Meanwhile, as the number of geo-tagged photos continues to climb, new challenges are brought to various applications. The existence of massive near-duplicate geo-tagged photos jeopardizes the effective presentation for the above applications. A new dimension in the search and presentation of geo-tagged photos is urgently demanded. In this paper, we devise a location visualization framework to efficiently retrieve and present diverse views captured within a local proximity. Novel photos, in terms of capture locations and visual content, are identified and returned in response to a query location for diverse visualization. For real-time response and good scalability, a new Hybrid Index structure which integrates R-tree and Geographic Grid is proposed to quickly identify the Maximal Near-duplicate Photo Groups (MNPG) in the query proximity. The most novel photos from different groups are then returned to generate diverse views on the location. Extensive experiments on synthetic and real-life photo datasets prove the novelty and efficiency of our methods.

Location-based social networks have been gaining increasing popularity in recent years. To increase users’ engagement with location-based services, it is important to provide attractive features, one of which is geo-targeted ads and coupons. To make ads and coupon delivery more effective, it is essential to predict the location that is most likely to be visited by a user at the next step. However, an inherent challenge in location prediction is a huge prediction space, with millions of distinct check-in locations as prediction target. In this paper we exploit the check-in category information to model the underlying user movement pattern. We propose a framework which uses a mixed hidden Markov model to predict the category of user activity at the next step and then predict the most likely location given the estimated category distribution. The advantages of modeling the category level include a significantly reduced prediction space and a precise expression of the semantic meaning of user activities. Extensive experimental results show that, with the predicted category distribution, the number of location candidates for prediction is 5.45 times smaller, while the prediction accuracy is 13.21% higher.

Different from a large body of research on social networks that has focused almost exclusively on positive relationships, we study signed social networks with both positive and negative links. Specifically, we focus on how to reliably and effectively predict the signs of links in a newly formed signed social network (called a target network). Since usually only a very small amount of edge sign information is available in such newly formed networks, this small quantity is not adequate to train a good classifier. To address this challenge, we need assistance from an existing, mature signed network (called a source network) which has abundant edge sign information. We adopt the transfer learning approach to leverage the edge sign information from the source network, which may have a different yet related joint distribution of the edge instances and their class labels.

Social contagion depicts a process of information (e.g., fads, opinions, news) diffusion in the online social networks. A recent study reports that in a social contagion process the probability of contagion is tightly controlled by the number of connected components in an individual's neighborhood. Such a number is termed structural diversity of an individual and it is shown to be a key predictor in the social contagion process. Based on this, a fundamental issue in a social network is to find top-kusers with the highest structural diversities. In this paper, we, for the first time, study the top-kstructural diversity search problem in a large network. Specifically, we develop an effective upper bound of structural diversity for pruning the search space. The upper bound can be incrementally refined in the search process. Based on such upper bound, we propose an efficient framework for top-kstructural diversity search. To further speed up the structural diversity evaluation in the search process, several carefully devised heuristic search strategies are proposed. Extensive experimental studies are conducted in 13 real-world large networks, and the results demonstrate the efficiency and effectiveness of the proposed methods.

It is quite common for networks emerging nowadays to have labels or textual contents on the nodes. On such networks, we study the problem of top-knearest keyword (k-NK) search. In a networkGmodeled as an undirected graph, each node is attached with zero or more keywords, and each edge is assigned with a weight measuring its length. Given a query nodeqinGand a keyword λ, a k-NK query seeksknodes which contain λ and are nearest toq. k-NK is not only useful as a stand-alone query but also as a building block for tackling complex graph pattern matching problems. The key to an accurate k-NK result is a precise shortest distance estimation in a graph. Based on the latest distance oracle technique, we build a shortest path tree for a distance oracle and use the tree distance as a more accurate estimation. With such representation, the original k-NK query on a graph can be reduced to answering the query on a set of trees and then assembling the results obtained from the trees. We propose two efficient algorithms to report the exact k-NK result on a tree. One is query time optimized for a scenario when a small number of result nodes are of interest to users. The other handles k-NK queries for an arbitrarily largekefficiently. In obtaining a k-NK result on a graph from that on trees, a global storage technique is proposed to further reduce the index size and the query time. Extensive experimental results conform with our theoretical findings, and demonstrate the effectiveness and efficiency of our k-NK algorithms on large real graphs.

Hash-based methods achieve fast similarity search by representing high-dimensional data with compact binary codes. However, both generating binary codes and encoding unseen data effectively and efficiently remain very challenging tasks. In this article, we focus on these tasks to implement approximate similarity search by proposing a novel hash based method named sparse hashing (SH for short). To generate interpretable (or semantically meaningful) binary codes, the proposed SH first converts original data into low-dimensional data through a novel nonnegative sparse coding method. SH then converts the low-dimensional data into Hamming space (i.e., binary encoding low-dimensional data) by a new binarization rule. After this, training data are represented by generated binary codes. To efficiently and effectively encode unseen data, SH learns hash functions by taking a-priori knowledge into account, such as implicit group effect of the features in training data, and the correlations between original space and the learned Hamming space. SH is able to perform fast approximate similarity search by efficient bit XOR operations in the memory of a modern PC with short binary code representations. Experimental results show that the proposed SH significantly outperforms state-of-the-art techniques.

Measuring robustness of complex networks is a fundamental task for analyzing the structure and function of complex networks. In this paper, we study the network robustness under the maximal vertex coverage (MVC) attack, where the attacker aims to delete as many edges of the network as possible by attacking a small fraction of nodes. First, we present two robustness metrics of complex networks based on MVC attack. We then propose an efficient randomized greedy algorithm with near-optimal performance guarantee for computing the proposed metrics. Finally, we conduct extensive experiments on 20 real datasets. The results show that P2P and co-authorship networks are extremely robust under the MVC attack while both the online social networks and the Email communication networks exhibit vulnerability under the MVC attack. In addition, the results demonstrate the efficiency and effectiveness of our proposed algorithms for computing the corresponding robustness metrics.

A graph models complex structural relationships among objects, and has been prevalently used in a wide range of applications. Building an automated graph classification model becomes very important for predicting unknown graphs or understanding complex structures between different classes. The graph classification framework being widely used consists of two steps, namely, feature selection and classification. The key issue is how to select important subgraph features from a graph database with a large number of graphs including positive graphs and negative graphs. Given the features selected, a generic classification approach can be used to build a classification model. In this paper, we focus on feature selection. We identify two main issues with the most widely used feature selection approach which is based on a discriminative score to select frequent subgraph features, and introduce a new diversified discriminative score to select features that have a higher diversity. We analyze the properties of the newly proposed diversified discriminative score, and conducted extensive performance studies to demonstrate that such a diversified discriminative score makes positive/negative graphs separable and leads to a higher classification accuracy.

Querying similar graphs in graph databases has been widely studied in graph query processing in recent years. Existing works mainly focus on subgraph similarity search and supergraph similarity search. In this paper, we study the problem of finding top-k graphs in a graph database that are most similar to a query graph. This problem has many applications, such as image retrieval and chemical compound structure search. Regarding the similarity measure, feature based and kernel based similarity measures have been used in the literature. But such measures are rough and may lose the connectivity information among substructures. In this paper, we introduce a new similarity measure based on the maximum common subgraph (MCS) of two graphs. We show that this measure can better capture the common and different structures of two graphs. Since computing the MCS of two graphs is NP-hard, we propose an algorithm to answer the top-k graph similarity query using two distance lower bounds with different computational costs, in order to reduce the number of MCS computations. We further introduce an indexing technique, which can better make use of the triangle property of similarities among graphs in the database to get tighter lower bounds. Three different indexing methods are proposed with different tradeoffs between pruning power and construction cost. We conducted extensive performance studies on large real datasets to evaluate the performance of our approaches.

Shortest distance query between two nodes is a fundamental operation in large-scale networks. Most existing methods in the literature take a landmark embedding approach, which selects a set of graph nodes as landmarks and computes the shortest distances from each landmark to all nodes as an embedding. To handle a shortest distance query between two nodes, the precomputed distances from the landmarks to the query nodes are used to compute an approximate shortest distance based on the triangle inequality. In this paper, we analyze the factors that affect the accuracy of the distance estimation in the landmark embedding approach. In particular we find that a globally selected, query-independent landmark set plus the triangulation based distance estimation introduces a large relative error, especially for nearby query nodes. To address this issue, we propose a query-dependent local landmark scheme, which identifies a local landmark close to the specific query nodes and provides a more accurate distance estimation than the traditional global landmark approach. Specifically, a local landmark is defined as the least common ancestor of the two query nodes in the shortest path tree rooted at a global landmark. We propose efficient local landmark indexing and retrieval techniques, which are crucial to achieve low offline indexing complexity and online query complexity. Two optimization techniques on graph compression and graph online search are also proposed, with the goal to further reduce index size and improve query accuracy. Our experimental results on large-scale social networks and road networks demonstrate that the local landmark scheme reduces the shortest distance estimation error significantly when compared with global landmark embedding.

The importance of introducing distance constraints to data dependencies, such as differential dependencies (DDs) [28], has recently been recognized. The metric distance constraints are tolerant to small variations, which enable them apply to wide data quality checking applications, such as detecting data violations. However, the determination of distance thresholds for the metric distance constraints is non-trivial. It often relies on a truth data instance which embeds the distance constraints. To find useful distance threshold patterns from data, there are several guidelines of statistical measures to specify, e.g., support, confidence and dependent quality. Unfortunately, given a data instance, users might not have any knowledge about the data distribution, thus it is very challenging to set the right parameters. In this paper, we study the determination of distance thresholds for metric distance constraints, in a parameter-free style. Specifically, we compute an expected utility based on the statistical measures from the data. According to our analysis as well as experimental verification, distance threshold patterns with higher expected utility could offer better usage in real applications, such as violation detection. We then develop efficient algorithms to determine the distance thresholds having the maximum expected utility. Finally, our extensive experimental evaluation demonstrates the effectiveness and efficiency of the proposed methods.

With the growth of the Internet and E-commerce, bipartite rating networks are ubiquitous. In such bipartite rating networks, there exist two types of entities: the users and the objects, where users give ratings to objects. A fundamental problem in such networks is how to rank the objects by user's ratings. Although it has been extensively studied in the past decade, the existing algorithms either cannot guarantee convergence, or are not robust to the spammers. In this paper, we propose six new reputation-based algorithms, where the users' reputation is determined by the aggregated difference between the users' ratings and the corresponding objects' rankings. We prove that all of our algorithms converge into a unique fixed point. The time and space complexity of our algorithms are linear w.r.t. the size of the graph, thus they can be scalable to large datasets. Moreover, our algorithms are robust to the spamming users. We evaluate our algorithms using three real datasets. The experimental results confirm the effectiveness, efficiency, and robustness of our algorithms.

Graph clustering, also known as community detection, is a long-standing problem in data mining. However, with the proliferation of rich attribute information available for objects in real-world graphs, how to leverage structural and attribute information for clustering attributed graphs becomes a new challenge. Most existing works take a distance-based approach. They proposed various distance measures to combine structural and attribute information. In this paper, we consider an alternative view and propose a model-based approach to attributed graph clustering. We develop a Bayesian probabilistic model for attributed graphs. The model provides a principled and natural framework for capturing both structural and attribute aspects of a graph, while avoiding the artificial design of a distance measure. Clustering with the proposed model can be transformed into a probabilistic inference problem, for which we devise an efficient variational algorithm. Experimental results on large real-world datasets demonstrate that our method significantly outperforms the state-of-art distance-based attributed graph clustering method.

Traditional information retrieval models do not necessarily provide users with optimal search experience because the top ranked documents may contain excessively redundant information. Therefore, satisfying search results should be not only relevant to the query but also diversified to cover different subtopics of the query. In this paper, the authors propose a novel pattern-based framework to diversify search results, where each pattern is a set of semantically related terms covering the same subtopic. They first apply a maximal frequent pattern mining algorithm to extract the patterns from retrieval results of the query. The authors then propose to model a subtopic with either a single pattern or a group of similar patterns. A profile-based clustering method is adapted to group similar patterns based on their context information. The search results are then diversified using the extracted subtopics. Experimental results show that the proposed pattern-based methods are effective to diversify the search results.

We study the problem of answering k -hop reachability queries in a directed graph, i.e., whether there exists a directed path of length k , from a source query vertex to a target query vertex in the input graph. The problem of k -hop reachability is a general problem of the classic reachability (where k = ∞). Existing indexes for processing classic reachability queries, as well as for processing shortest path queries, are not applicable or not efficient for processing k -hop reachability queries. We propose an index for processing k -hop reachability queries, which is simple in design and efficient to construct. Our experimental results on a wide range of real datasets show that our index is more efficient than the state-of-the-art indexes even for processing classic reachability queries, for which these indexes are primarily designed. We also show that our index is efficient in answering k -hop reachability queries.

A lot of data are in sequential formats. In this study, we are interested in sequential data that goes in pairs. There are many interesting datasets in this format coming from various domains including parallel textual corpora, duplicate bug reports, and other pairs of related sequences of events. Our goal is to mine a set of closed discriminative dyadic sequential patterns from a database of sequence pairs each belonging to one of the two classes +ve and -ve. These dyadic sequential patterns characterize the discriminating facets contrasting the two classes. They are potentially good features to be used for the classification of dyadic sequential data. They can be used to characterize and flag correct and incorrect translations from parallel textual corpora, automate the manual and time consuming duplicate bug report detection process, etc. We provide a solution of this new problem by proposing new search space traversal strategy, projected database structure, pruning properties, and novel mining algorithms. To demonstrate the scalability and utility of our solution, we have experimented with both synthetic and real datasets. Experiment results show that our solution is scalable. Mined patterns are also able to improve the accuracy of one possible downstream application, namely the detection of duplicate bug reports using pattern-based classification.

Name ambiguity has long been viewed as a challenging problem in many applications, such as scientific literature management, people search, and social network analysis. When we search a person name in these systems, many documents (e.g., papers, web pages) containing that person's name may be returned. It is hard to determine which documents are about the person we care about. Although much research has been conducted, the problem remains largely unsolved, especially with the rapid growth of the people information available on the Web. In this paper, we try to study this problem from a new perspective and propose an ADANA method for disambiguating person names via active user interactions. In ADANA, we first introduce a pairwise factor graph (PFG) model for person name disambiguation. The model is flexible and can be easily extended by incorporating various features. Based on the PFG model, we propose an active name disambiguation algorithm, aiming to improve the disambiguation performance by maximizing the utility of the user's correction. Experimental results on three different genres of data sets show that with only a few user corrections, the error rate of name disambiguation can be reduced to 3.1%. A real system has been developed based on the proposed method and is available online.

Social networks, sensor networks, biological networks, and many other information networks can be modeled as a large graph. Graph vertices represent entities, and graph edges represent their relationships or interactions. In many large graphs, there is usually one or more attributes associated with every graph vertex to describe its properties. In many application domains, graph clustering techniques are very useful for detecting densely connected groups in a large graph as well as for understanding and visualizing a large graph. The goal of graph clustering is to partition vertices in a large graph into different clusters based on various criteria such as vertex connectivity or neighborhood similarity. Many existing graph clustering methods mainly focus on the topological structure for clustering, but largely ignore the vertex properties, which are often heterogenous. In this article, we propose a novel graph clustering algorithm, SA-Cluster , which achieves a good balance between structural and attribute similarities through a unified distance measure. Our method partitions a large graph associated with attributes into k clusters so that each cluster contains a densely connected subgraph with homogeneous attribute values. An effective method is proposed to automatically learn the degree of contributions of structural similarity and attribute similarity. Theoretical analysis is provided to show that SA-Cluster is converging quickly through iterative cluster refinement. Some optimization techniques on matrix computation are proposed to further improve the efficiency of SA-Cluster on large graphs. Extensive experimental results demonstrate the effectiveness of SA-Cluster through comparisons with the state-of-the-art graph clustering and summarization methods.

Classification has been used for modeling many kinds of data sets, including sets of items, text documents, graphs, and networks. However, there is a lack of study on a new kind of data, trajectories on road networks. Modeling such data is useful with the emerging GPS and RFID technologies and is important for effective transportation and traffic planning. In this work, we study methods for classifying trajectories on road networks. By analyzing the behavior of trajectories on road networks, we observe that, in addition to the locations where vehicles have visited, the order of these visited locations is crucial for improving classification accuracy. Based on our analysis, we contend that (frequent) sequential patterns are good feature candidates since they preserve this order information. Furthermore, when mining sequential patterns, we propose to confine the length of sequential patterns to ensure high efficiency. Compared with closed sequential patterns, these partial (i.e., length-confined) sequential patterns allow us to significantly improve efficiency almost without losing accuracy. In this paper, we present a framework for frequent pattern-based classification for trajectories on road networks. Our comparative study over a broad range of classification approaches demonstrates that our method significantly improves accuracy over other methods in some synthetic and real trajectory data.

In recent years, many networks have become available for analysis, including social networks, sensor networks, biological networks, etc. Graph clustering has shown its effectiveness in analyzing and visualizing large networks. The goal of graph clustering is to partition vertices in a large graph into clusters based on various criteria such as vertex connectivity or neighborhood similarity. Many existing graph clustering methods mainly focus on the topological structures, but largely ignore the vertex properties which are often heterogeneous. Recently, a new graph clustering algorithm, SA-Cluster, has been proposed which combines structural and attribute similarities through a unified distance measure. SA-Cluster performs matrix multiplication to calculate the random walk distances between graph vertices. As the edge weights are iteratively adjusted to balance the importance between structural and attribute similarities, matrix multiplication is repeated in each iteration of the clustering process to recalculate the random walk distances which are affected by the edge weight update. In order to improve the efficiency and scalability of SA-Cluster, in this paper, we propose an efficient algorithm Inc-Cluster to incrementally update the random walk distances given the edge weight increments. Complexity analysis is provided to estimate how much runtime cost Inc-Cluster can save. Experimental results demonstrate that Inc-Cluster achieves significant speedup over SA-Cluster on large graphs, while achieving exactly the same clustering quality in terms of intra-cluster structural cohesiveness and attribute value homogeneity.

Massive radio frequency identification (RFID) data sets are expected to become commonplace in supply chain management systems. Warehousing and mining this data is an essential problem with great potential benefits for inventory management, object tracking, and product procurement processes. Since RFID tags can be used to identify each individual item, enormous amounts of location-tracking data are generated. With such data, object movements can be modeled by movement graphs, where nodes correspond to locations and edges record the history of item transitions between locations. In this study, we develop a movement graph model as a compact representation of RFID data sets. Since spatiotemporal as well as item information can be associated with the objects in such a model, the movement graph can be huge, complex, and multidimensional in nature. We show that such a graph can be better organized around gateway nodes, which serve as bridges connecting different regions of the movement graph. A graph-based object movement cube can be constructed by merging and collapsing nodes and edges according to an application-oriented topological structure. Moreover, we propose an efficient cubing algorithm that performs simultaneous aggregation of both spatiotemporal and item dimensions on a partitioned movement graph, guided by such a topological structure.

Recently, video search reranking has been an effective mechanism to improve the initial text-based ranking list by incorporating visual consistency among the result videos. While existing methods attempt to rerank all the individual result videos, they suffer from several drawbacks. In this article, we propose a new video reranking paradigm called cluster-based video reranking (CVR). The idea is to first construct a video near-duplicate graph representing the visual similarity relationship among videos, followed by identifying the near-duplicate clusters from the video near-duplicate graph, then ranking the obtained near-duplicate clusters based on cluster properties and intercluster links, and finally for each ranked cluster, a representative video is selected and returned. Compared to existing methods, the new CVR ranks clusters and exhibits several advantages, including superior reranking by utilizing more reliable cluster properties, fast reranking on a small number of clusters, diverse and representative results. Particularly, we formulate the near-duplicate cluster identification as a novel maximally cohesive subgraph mining problem. By leveraging the designed cluster scoring properties indicating the cluster's importance and quality, random walk is applied over the near-duplicate cluster graph to rank clusters. An extensive evaluation study proves the novelty and superiority of our proposals over existing methods.

As information networks become ubiquitous, extracting knowledge from information networks has become an important task. Both ranking and clustering can provide overall views on information network data, and each has been a hot topic by itself. However, ranking objects globally without considering which clusters they belong to often leads to dumb results, e.g., ranking database and computer architecture conferences together may not make much sense. Similarly, clustering a huge number of objects (e.g., thousands of authors) in one huge cluster without distinction is dull as well.

Software is a ubiquitous component of our daily life. We often depend on the correct working of software systems. Due to the difficulty and complexity of software systems, bugs and anomalies are prevalent. Bugs have caused billions of dollars loss, in addition to privacy and security threats. In this work, we address software reliability issues by proposing a novel method to classify software behaviors based on past history or runs. With the technique, it is possible to generalize past known errors and mistakes to capture failures and anomalies. Our technique first mines a set of discriminative features capturing repetitive series of events from program execution traces. It then performs feature selection to select the best features for classification. These features are then used to train a classifier to detect failures. Experiments and case studies on traces of several benchmark software systems and a real-life concurrency bug from MySQL server show the utility of the technique in capturing failures and anomalies. On average, our pattern-based classification technique outperforms the baseline approach by 24.68% in accuracy.

Graph classification is an increasingly important step in numerous application domains, such as function prediction of molecules and proteins, computerised scene analysis, and anomaly detection in program flows. Among the various approaches proposed in the literature, graph classification based on frequent subgraphs is a popular branch: Graphs are represented as (usually binary) vectors, with components indicating whether a graph contains a particular subgraph that is frequent across the dataset. On large graphs, however, one faces the enormous problem that the number of these frequent subgraphs may grow exponentially with the size of the graphs, but only few of them possess enough discriminative power to make them useful for graph classification. Efficient and discriminative feature selection among frequent subgraphs is hence a key challenge for graph mining. In this article, we propose an approach to feature selection on frequent subgraphs, called CORK, that combines two central advantages. First, it optimizes a submodular quality criterion, which means that we can yield a near-optimal solution using greedy feature selection. Second, our submodular quality function criterion can be integrated into gSpan, the state-of-the-art tool for frequent subgraph mining, and help to prune the search space for discriminative frequent subgraphs even during frequent subgraph mining.

The goal of graph clustering is to partition vertices in a large graph into different clusters based on various criteria such as vertex connectivity or neighborhood similarity. Graph clustering techniques are very useful for detecting densely connected groups in a large graph. Many existing graph clustering methods mainly focus on the topological structure for clustering, but largely ignore the vertex properties which are often heterogenous. In this paper, we propose a novel graph clustering algorithm, SA-Cluster , based on both structural and attribute similarities through a unified distance measure. Our method partitions a large graph associated with attributes into k clusters so that each cluster contains a densely connected subgraph with homogeneous attribute values. An effective method is proposed to automatically learn the degree of contributions of structural similarity and attribute similarity. Theoretical analysis is provided to show that SA-Cluster is converging. Extensive experimental results demonstrate the effectiveness of SA-Cluster through comparison with the state-of-the-art graph clustering and summarization methods.

The application of frequent patterns in classification has demonstrated its power in recent studies. It often adopts a two-step approach: frequent pattern (or classification rule) mining followed by feature selection (or rule ranking). However, this two-step process could be computationally expensive, especially when the problem scale is large or the minimum support is low. It was observed that frequent pattern mining usually produces a huge number of "patterns" that could not only slow down the mining process but also make feature selection hard to complete. In this paper, we propose a direct discriminative pattern mining approach, DDPMine, to tackle the efficiency issue arising from the two-step approach. DDPMine performs a branch-and-bound search for directly mining discriminative patterns without generating the complete pattern set. Instead of selecting best patterns in a batch, we introduce a "feature-centered" mining approach that generates discriminative patterns sequentially on a progressively shrinking FP-tree by incrementally eliminating training instances. The instance elimination effectively reduces the problem size iteratively and expedites the mining process. Empirical results show that DDPMine achieves orders of magnitude speedup without any downgrade of classification accuracy. It outperforms the state-of-the-art associative classification methods in terms of both accuracy and efficiency.

Frequent patterns provide solutions to datasets that do not have well-structured feature vectors. However, frequent pattern mining is non-trivial since the number of unique patterns is exponential but many are non-discriminative and correlated. Currently, frequent pattern mining is performed in two sequential steps: enumerating a set of frequent patterns, followed by feature selection. Although many methods have been proposed in the past few years on how to perform each separate step efficiently, there is still limited success in eventually finding highly compact and discriminative patterns. The culprit is due to the inherent nature of this widely adopted two-step approach. This paper discusses these problems and proposes a new and different method. It builds a decision tree that partitions the data onto different nodes. Then at each node, it directly discovers a discriminative pattern to further divide its examples into purer subsets. Since the number of examples towards leaf level is relatively small, the new approach is able to examine patterns with extremely low global support that could not be enumerated on the whole dataset by the two-step method. The discovered feature vectors are more accurate on some of the most difficult graph as well as frequent itemset problems than most recently proposed algorithms but the total size is typically 50% or more smaller. Importantly, the minimum support of some discriminative patterns can be extremely low (e.g. 0.03%). In order to enumerate these low support patterns, state-of-the-art frequent pattern algorithm either cannot finish due to huge memory consumption or have to enumerate 101 to 103 times more patterns before they can even be found. Software and datasets are available by contacting the author.

Recently, frequent sequential pattern mining algorithms have been widely used in software engineering field to mine various source code or specification patterns.In practice, software evolves from one version to another in its life span.The effort of mining frequent sequential patterns across multiple versions of a software can be substantially reduced by efficient incremental mining.This problem is challenging in this domain since the databases are usually updated in all kinds of manners including insertion, various modifications as well as removal of sequences.Also, different mining tools may have various mining constraints, such as low minimum support.None of the existing work can be applied effectively due to various limitations of such work.For example, our recent work, IncSpan, failed solving the problem because it could neither handle low minimum support nor removal of sequences from database.In this paper, we propose a novel, comprehensive incremental mining algorithm for frequent sequential pattern, CISpan (Comprehensive Incremental Sequential Pattern mining).CISpan supports both closed and complete incremental frequent sequence mining, with all kinds of updates to the database.Compared to IncSpan, CISpan tolerates a wide range for minimum support threshold (as low as 2).Our performance study shows that in addition to handling more test cases on which IncSpan fails, CISpan outperforms IncSpan in all test cases which IncSpan could handle, including various sequence length, number of sequences, modification ratio, etc., with an average of 3.4 times speedup.We also tested CISpan's performance on databases transformed from 20 consecutive versions of Linux Kernel source code.On average, CISpan outperforms the non-incremental CloSpan by 42 times.

Online bibliographic databases, such as DBLP in computer science and PubMed in medical sciences, contain abundant information about research publications in different fields. Each such database forms a gigantic information network (hence called BibNet), connecting in complex ways research papers, authors, conferences/journals, and possibly citation information as well, and provides a fertile land for information network analysis. Our BibNetMiner is designed for sophisticated information network mining on such bibliographic databases. In this demo, we will take the DBLP database as an example, demonstrate several attractive functions of BibNetMiner, including clustering, ranking and profiling of conferences and authors based on the research subfields. A user-friendly, visualization-enhanced interface will be provided to facilitate interactive exploration of a bibliographic database. This project will serve as an example to demonstrate the power of links in information network mining. Since the dataset is large and the network is heterogeneous, such a study will benefit the research on the analysis of massive heterogeneous information networks.

With ever-increasing amounts of graph data from disparate sources, there has been a strong need for exploiting significant graph patterns with user-specified objective functions. Most objective functions are not antimonotonic, which could fail all of frequency-centric graph mining algorithms. In this paper, we give the first comprehensive study on general mining method aiming to find most significant patterns directly. Our new mining framework, called LEAP (Descending Leap Mine), is developed to exploit the correlation between structural similarity and significance similarity in a way that the most significant pattern could be identified quickly by searching dissimilar graph patterns. Two novel concepts, structural leap search and frequency descending mining, are proposed to support leap search in graph pattern space. Our new mining method revealed that the widely adopted branch-and-bound search in data mining literature is indeed not the best, thus sketching a new picture on scalable graph pattern discovery. Empirical results show that LEAP achieves orders of magnitude speedup in comparison with the state-of-the-art method. Furthermore, graph classifiers built on mined patterns outperform the up-to-date graph kernel method in terms of efficiency and accuracy, demonstrating the high promise of such patterns.

The application of frequent patterns in classification appeared in sporadic studies and achieved initial success in the classification of relational data, text documents and graphs. In this paper, we conduct a systematic exploration of frequent pattern-based classification, and provide solid reasons supporting this methodology. It was well known that feature combinations (patterns) could capture more underlying semantics than single features. However, inclusion of infrequent patterns may not significantly improve the accuracy due to their limited predictive power. By building a connection between pattern frequency and discriminative measures such as information gain and Fisher score, we develop a strategy to set minimum support in frequent pattern mining for generating useful patterns. Based on this strategy, coupled with a proposed feature selection algorithm, discriminative frequent patterns can be generated for building high quality classifiers. We demonstrate that the frequent pattern-based classification framework can achieve good scalability and high accuracy in classifying large datasets. Empirical studies indicate that significant improvement in classification accuracy is achieved (up to 12% in UCI datasets) using the so-selected discriminative frequent patterns.

Extensive research for frequent-pattern mining in the past decade has brought forth a number of pattern mining algorithms that are both effective and efficient. However, the existing frequent-pattern mining algorithms encounter challenges at mining rather large patterns, called colossal frequent patterns, in the presence of an explosive number of frequent patterns. Colossal patterns are critical to many applications, especially in domains like bioinformatics. In this study, we investigate a novel mining approach called pattern-fusion to efficiently find a good approximation to the colossal patterns. With Pattern-Fusion, a colossal pattern is discovered by fusing its small core patterns in one step, whereas the incremental pattern-growth mining strategies, such as those adopted in Apriori and FP-growth, have to examine a large number of mid-sized ones. This property distinguishes pattern-fusion from all the existing frequent pattern mining approaches and draws a new mining methodology. Our empirical studies show that, in cases where current mining algorithms cannot proceed, pattern-fusion is able to mine a result set which is a close enough approximation to the complete set of the colossal patterns, under a quality evaluation model proposed in this paper.

Using frequent patterns to analyze data has been one of the fundamental approaches in many data mining applications. Research in frequent pattern mining has so far mostly focused on developing efficient algorithms to discover various kinds of frequent patterns, but little attention has been paid to the important next step—interpreting the discovered frequent patterns. Although the compression and summarization of frequent patterns has been studied in some recent work, the proposed techniques there can only annotate a frequent pattern with nonsemantical information (e.g., support), which provides only limited help for a user to understand the patterns. In this article, we study the novel problem of generating semantic annotations for frequent patterns. The goal is to discover the hidden meanings of a frequent pattern by annotating it with in-depth, concise, and structured information. We propose a general approach to generate such an annotation for a frequent pattern by constructing its context model, selecting informative context indicators, and extracting representative transactions and semantically similar patterns. This general approach can well incorporate the user's prior knowledge, and has potentially many applications, such as generating a dictionary-like description for a pattern, finding synonym patterns, discovering semantic relations, and summarizing semantic classes of a set of frequent patterns. Experiments on different datasets show that our approach is effective in generating semantic pattern annotations.

Recent studies have proposed methods to discover approximate frequent itemsets in the presence of random noise. By relaxing the rigid requirement of exact frequent pattern mining, some interesting patterns, which would previously be fragmented by exact pattern mining methods due to the random noise or measurement error, are successfully recovered. Unfortunately, a large number of "uninteresting" candidates are explored as well during the mining process, as a result of the relaxed pattern mining methodology. This severely slows down the mining process. Even worse, it is hard for an end user to distinguish the recovered interesting patterns from these uninteresting ones. In this paper, we propose an efficient algorithm AC-Close to recover the approximate closed itemsets from "core patterns". By focusing on the so-called core patterns, integrated with a top-down mining and several effective pruning strategies, the algorithm narrows down the search space to those potentially interesting ones. Experimental results show that AC-Close substantially outperforms the previously proposed method in terms of efficiency, while delivers a similar set of interesting recovered patterns.

As a fundamental data mining task, frequent pattern mining has widespread applications in many different domains. Research in frequent pattern mining has so far mostly focused on developing efficient algorithms to discover various kinds of frequent patterns, but little attention has been paid to the important nextstep - interpreting the discovered frequent patterns. Although some recent work has studied the compression and summarization of frequent patterns, the proposed techniques can only annotate a frequent pattern with non-semantical information (e.g. support), which provides only limited help for a user to understand the patterns.In this paper, we propose the novel problem of generating semantic annotations for frequent patterns. The goal is to annotate a frequent pattern with in-depth, concise, and structured information that can better indicate the hidden meanings of the pattern. We propose a general approach to generate such anannotation for a frequent pattern by constructing its context model, selecting informative context indicators, and extracting representative transactions and semantically similar patterns. This general approach has potentially many applications such as generating a dictionary-like description for a pattern, finding synonym patterns, discovering semantic relations, and summarizing semantic classes of a set of frequent patterns. Experiments on different datasets show that our approach is effective in generating semantic pattern annotations.

Observed in many applications, there is a potential need of extracting a small set of frequent patterns having not only high significance but also low redundancy. The significance is usually defined by the context of applications. Previous studies have been concentrating on how to compute top-k significant patterns or how to remove redundancy among patterns separately. There is limited work on finding those top-k patterns which demonstrate high-significance and low-redundancy simultaneously.In this paper, we study the problem of extracting redundancy-aware top-k patterns from a large collection of frequent patterns. We first examine the evaluation functions for measuring the combined significance of a pattern set and propose the MMS (Maximal Marginal Significance) as the problem formulation. The problem is known as NP-hard. We further present a greedy algorithm which approximates the optimal solution with performance bound O(log k) (with conditions on redundancy), where k is the number of reported patterns. The direct usage of redundancy-aware top-k patterns is illustrated through two real applications: disk block prefetch and document theme extraction. Our method can also be applied to processing redundancy-aware top-k queries in traditional database.

Frequent-pattern mining has been studied extensively on scalable methods for mining various kinds of patterns including itemsets, sequences, and graphs. However, the bottleneck of frequent-pattern mining is not at the efficiency but at the interpretability, due to the huge number of patterns generated by the mining process.In this paper, we examine how to summarize a collection of itemset patterns using only K representatives, a small number of patterns that a user can handle easily. The K representatives should not only cover most of the frequent patterns but also approximate their supports. A generative model is built to extract and profile these representatives, under which the supports of the patterns can be easily recovered without consulting the original dataset. Based on the restoration error, we propose a quality measure function to determine the optimal value of parameter K. Polynomial time algorithms are developed together with several optimization heuristics for efficiency improvement. Empirical studies indicate that we can obtain compact summarization in real datasets.

In this paper, we study the issues related to the design and construction of high-performance sequence index structures in large sequence databases. To build effective indices, a novel method, called SeqIndex, is proposed, in which the selection of indices is based on the analysis of discriminative, frequent sequential patterns mined from large sequence databases. Such an analysis leads to the construction of compact and effective indexing structures. Furthermore, we eliminate the requirement of setting an optimal support threshold beforehand, which is difficult for users to provide in practice. The discriminative, frequent pattern based indexing method is proven very effective based on our performance study.

Many real life sequence databases grow incrementally. It is undesirable to mine sequential patterns from scratch each time when a small set of sequences grow, or when some new sequences are added into the database. Incremental algorithm should be developed for sequential pattern mining so that mining can be adapted to incremental database updates. However, it is nontrivial to mine sequential patterns incrementally, especially when the existing sequences grow incrementally because such growth may lead to the generation of many new patterns due to the interactions of the growing subsequences with the original ones. In this study, we develop an efficient algorithm, IncSpan, for incremental mining of sequential patterns, by exploring some interesting properties. Our performance study shows that IncSpan outperforms some previously proposed incremental algorithms as well as a non-incremental one with a wide margin.