Recent agent-based recommendation frameworks aim to simulate user behaviors by incorporating memory mechanisms and prompting strategies, but they struggle with hallucinating non-existent items and full-catalog ranking. Besides, a largely underexplored opportunity lies in leveraging LLMs' commonsense reasoning to capture user intent through substitute and complement relationships between items, which are usually implicit in datasets and difficult for traditional ID-based recommenders to capture. In this work, we propose a novel LLM-agent framework, AgentDR, which bridges LLM reasoning with scalable recommendation tools. Our approach delegates full-ranking tasks to traditional models while utilizing LLMs to (i) integrate multiple recommendation outputs based on personalized tool suitability and (ii) reason over substitute and complement relationships grounded in user history. This design mitigates hallucination, scales to large catalogs, and enhances recommendation relevance through relational reasoning. Through extensive experiments on three public grocery datasets, we show that our framework achieves superior full-ranking performance, yielding on average a twofold improvement over its underlying tools. We also introduce a new LLM-based evaluation metric that jointly measures semantic alignment and ranking correctness.

Distributed graph neural network (GNN) training facilitates learning on massive graphs that surpass the storage and computational capabilities of a single machine. Traditional distributed frameworks strive for performance parity with centralized training by maximally recovering cross-instance node dependencies, relying either on inter-instance communication or periodic fallback to centralized training. However, these processes create overhead and constrain the scalability of the framework. In this work, we propose a streamlined framework for distributed GNN training that eliminates these costly operations, yielding improved scalability, convergence speed, and performance over state-of-the-art approaches. Our framework (1) comprises independent trainers that asynchronously learn local models from locally available parts of the training graph and (2) synchronizes these local models only through periodic (time-based) model aggregation. Contrary to prevailing belief, our theoretical analysis shows that it is not essential to maximize the recovery of cross-instance node dependencies to achieve performance parity with centralized training. Instead, our framework leverages randomized assignment of nodes or super-nodes (i.e., collections of original nodes) to partition the training graph in order to enhance data uniformity and minimize discrepancies in gradient and loss function across instances. Experiments on social and e-commerce networks with up to 1.3 billion edges show that our proposed framework achieves state-of-the-art performance and 2.31 \(\times\) speedup compared to the fastest baseline despite using less training data.

Graphs are a fundamental data structure used to represent relationships in domains as diverse as the social sciences, bioinformatics, cybersecurity, the Internet, and more. One of the central observations in network science is that real-world graphs are globally sparse, yet contain numerous "pockets" of high edge density. A fundamental task in graph mining is to discover these dense subgraphs. Most common formulations of the problem involve finding a single (or a few) "optimally" dense subsets. But in most real applications, one does not care for the optimality. Instead, we want to find a large collection of dense subsets that covers a significant fraction of the input graph. We give a mathematical formulation of this problem, using a new definition of regularly triangle-rich (RTR) families. These families capture the notion of dense subgraphs that contain many triangles and have degrees comparable to the subgraph size. We design a provable algorithm, RTRExtractor, that can discover RTR families that approximately cover any RTR set. The algorithm is efficient and is inspired by recent results that use triangle counts for community testing and clustering. We show that RTRExtractor has excellent behavior on a large variety of real-world datasets. It is able to process graphs with hundreds of millions of edges within minutes. Across many datasets, RTRExtractor achieves high coverage using high edge density datasets. For example, the output covers a quarter of the vertices with subgraphs of edge density more than (say) 0.5, for datasets with 10M+ edges. We show an example of how the output of RTRExtractor correlates with meaningful sets of similar vertices in a citation network, demonstrating the utility of RTRExtractor for unsupervised graph discovery tasks.

Large language model (LLM) agents have demonstrated impressive capabilities in utilizing external tools and knowledge to boost accuracy and reduce hallucinations. However, developing prompting techniques that enable LLM agents to effectively use these tools and knowledge remains a heuristic and labor-intensive task. Here, we introduce AvaTaR, a novel and automated framework that optimizes an LLM agent to effectively leverage provided tools, improving performance on a given task. During optimization, we design a comparator module to iteratively deliver insightful and comprehensive prompts to the LLM agent by contrastively reasoning between positive and negative examples sampled from training data. We demon- strate AvaTaR on four complex multimodal retrieval datasets featuring textual, visual, and relational information, and three general question-answering (QA) datasets. We find AvaTaR consistently outperforms state-of-the-art approaches across all seven tasks, exhibiting strong generalization ability when applied to novel cases and achieving an average relative improvement of 14% on the Hit@1 metric for the retrieval datasets and 13% for the QA datasets. Code and dataset are available at https://github.com/zou-group/avatar.

Answering real-world complex queries, such as complex product search, often requires accurate retrieval from semi-structured knowledge bases that involve blend of unstructured (e.g., textual descriptions of products) and structured (e.g., entity relations of products) information. However, many previous works studied textual and relational retrieval tasks as separate topics. To address the gap, we develop STARK, a large-scale Semi-structure retrieval benchmark on Textual and Relational Knowledge Bases. Our benchmark covers three domains: product search, academic paper search, and queries in precision medicine. We design a novel pipeline to synthesize realistic user queries that integrate diverse relational information and complex textual properties, together with their ground-truth answers (items). We conduct rigorous human evaluation to validate the quality of our synthesized queries. We further enhance the benchmark with high-quality human-generated queries to provide an authentic reference. STARK serves as a comprehensive testbed for evaluating the performance of retrieval systems driven by large language models (LLMs). Our experiments suggest that STARK presents significant challenges to the current retrieval and LLM systems, highlighting the need for more capable semi-structured retrieval systems.

Developing text mining approaches to mine aspects from customer reviews has been well-studied due to its importance in understanding customer needs and product attributes. In contrast, it remains unclear how to predict the future emerging aspects of a new product that currently has little review information. This task, which we named product aspect forecasting, is critical for recommending new products, but also challenging because of the missing reviews. Here, we propose ForeSeer, a novel textual mining and product embedding approach progressively trained on temporal product graphs for this novel product aspect forecasting task. ForeSeer transfers reviews from similar products on a large product graph and exploits these reviews to predict aspects that might emerge in future reviews. A key novelty of our method is to jointly provide review, product, and aspect embeddings that are both time-sensitive and less affected by extremely imbalanced aspect frequencies. We evaluated ForeSeer on a real-world product review system containing 11,536,382 reviews and 11,000 products over 3 years. We observe that ForeSeer substantially outperformed existing approaches with at least 49.1% AUPRC improvement under the real setting where aspect associations are not given. ForeSeer further improves future link prediction on the product graph and the review aspect association prediction. Collectively, Foreseer offers a novel framework for review forecasting by effectively integrating review text, product network, and temporal information, opening up new avenues for online shopping recommendation and e-commerce applications.

Modeling customer shopping intentions is a crucial task for e-commerce, as it directly impacts user experience and engagement. Thus, accurately understanding customer preferences is essential for providing personalized recommendations. Session-based recommendation, which utilizes customer session data to predict their next interaction, has become increasingly popular. However, existing session datasets have limitations in terms of item attributes, user diversity, and dataset scale. As a result, they cannot comprehensively capture the spectrum of user behaviors and preferences.To bridge this gap, we present the Amazon Multilingual Multi-locale Shopping Session Dataset, namely Amazon-M2. It is the first multilingual dataset consisting of millions of user sessions from six different locales, where the major languages of products are English, German, Japanese, French, Italian, and Spanish.Remarkably, the dataset can help us enhance personalization and understanding of user preferences, which can benefit various existing tasks as well as enable new tasks. To test the potential of the dataset, we introduce three tasks in this work:(1) next-product recommendation, (2) next-product recommendation with domain shifts, and (3) next-product title generation.With the above tasks, we benchmark a range of algorithms on our proposed dataset, drawing new insights for further research and practice. In addition, based on the proposed dataset and tasks, we hosted a competition in the KDD CUP 2023 https://www.aicrowd.com/challenges/amazon-kdd-cup-23-multilingual-recommendation-challenge and have attracted thousands of users and submissions. The winning solutions and the associated workshop can be accessed at our website~https://kddcup23.github.io/.

At E-Commerce stores such as Amazon, eBay, and Taobao, the shopping items and the query words that customers use to search for the items form a bipartite graph that captures search behavior. Such a query-item graph can be used to forecast search trends or improve search results. For example, generating query-item associations, which is equivalent to predicting links in the bipartite graph, can yield recommendations that can customize and improve the user search experience. Although the bipartite shopping graphs are straightforward to model search behavior, they suffer from two challenges: 1) The majority of items are sporadically searched and hence have noisy/sparse query associations, leading to a long-tail distribution. 2) Infrequent queries are more likely to link to popular items, leading to another hurdle known as disassortative mixing.

Zijie Huang, Zheng Li, Haoming Jiang, Tianyu Cao, Hanqing Lu, Bing Yin, Karthik Subbian, Yizhou Sun, Wei Wang. Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2022.

The large-scale nature of product catalog and the changing demands of customer queries makes product search a challenging problem. The customer queries are ambiguous and implicit. They may be looking for an exact match of their query, or a functional equivalent (i.e., substitute), or an accessory to go with it (i.e., complement). It is important to distinguish these three categories from merely classifying an item for a customer query as relevant or not. This information can help direct the customer and improve search applications to understand the customer mission. In this paper, we formulate search relevance as a multi-class classification problem and propose a graph-based solution to classify a given query-item pair as exact, substitute, complement, or irrelevant (ESCI). The customer engagement (clicks, add-to-cart, and purchases) between query and items serve as a crucial information for this problem. However, existing approaches rely purely on the textual information (such as BERT) and do not sufficiently focus on the structural relationships. Another challenge in including the structural information is the sparsity of such data in some regions. We propose Structure-Aware multilingual LAnguage Model (SALAM), that utilizes a language model along with a graph neural network, to extract region-specific semantics as well as relational information for the classification of query-product pairs. Our model is first pre-trained on a large region-agnostic dataset and behavioral graph data and then fine-tuned on region-specific versions to address the sparsity. We show in our experiments that SALAM significantly outperforms the current matching frameworks on the ESCI classification task in several regions. We also demonstrate the effectiveness of using a two-phased training setup (i.e., pre-training and fine-tuning) in capturing region-specific information. Also, we provide various challenges and solutions for using the model in an industrial setting and outline its contribution to the e-commerce engine.

Recent studies have revealed important properties that are unique to graph datasets such as hierarchies and global structures. This has driven research into hyperbolic space due to their ability to effectively encode the inherent hierarchy present in graph datasets. However, a major bottleneck here is the obscurity of hyperbolic geometry and a better comprehension of its gyrovector operations. In this tutorial, we aim to introduce researchers and practitioners in the data mining community to the hyperbolic equivariants of the Euclidean operations that are necessary to tackle their application to neural networks. We describe the popular hyperbolic variants of GNN architectures and explain their implementation, in contrast to the Euclidean counterparts. Also, we motivate our tutorial through critical analysis of existing applications in the areas of graph mining, knowledge graph reasoning, search, NLP, and computer vision.

Embeddings, low-dimensional vector representation of objects, are fundamental in building modern machine learning systems. In industrial settings, there is usually an embedding team that trains an embedding model to solve intended tasks (e.g., product recommendation). The produced embeddings are then widely consumed by consumer teams to solve their unintended tasks (e.g., fraud detection). However, as the embedding model gets updated and retrained to improve performance on the intended task, the newly-generated embeddings are no longer compatible with the existing consumer models. This means that historical versions of the embeddings can never be retired or all consumer teams have to retrain their models to make them compatible with the latest version of the embeddings, both of which are extremely costly in practice.

Graph Neural Networks (GNNs) have emerged as powerful tools to encode graph-structured data. Due to their broad applications, there is an increasing need to develop tools to explain how GNNs make decisions given graph-structured data. Existing learning-based GNN explanation approaches are task-specific in training and hence suffer from crucial drawbacks. Specifically, they are incapable of producing explanations for a multitask prediction model with a single explainer. They are also unable to provide explanations in cases where the GNN is trained in a self-supervised manner, and the resulting representations are used in future downstream tasks. To address these limitations, we propose a Task-Agnostic GNN Explainer (TAGE) that is independent of downstream models and trained under self-supervision with no knowledge of downstream tasks. TAGE enables the explanation of GNN embedding models with unseen downstream tasks and allows efficient explanation of multitask models. Our extensive experiments show that TAGE can significantly speed up the explanation efficiency by using the same model to explain predictions for multiple downstream tasks while achieving explanation quality as good as or even better than current state-of-the-art GNN explanation approaches.

Product search is a fundamentally challenging problem due to the large-size of product catalogues and the complexity of extracting semantic information from products. In addition to this, the black-box nature of most search systems also hamper a smooth customer experience. Current approaches in this area utilize lexical and semantic product information to match user queries against products. However, these models lack (i) a hierarchical query representation, (ii) a mechanism to detect and capture inter-entity relationships within a query, and (iii) a query composition method specific to e-commerce domain. To address these challenges, in this paper, we propose an AtteNTive Hyperbolic Entity Model (ANTHEM), a novel attention-based product search framework that models query entities as two-vector hyperboloids, learns inter-entity intersections and utilizes attention to unionize individual entities and inter-entity intersections to predict product matches from the search space. ANTHEM utilizes the first and second vector of hyperboloids to determine the query's semantic position and to tune its surrounding search volume, respectively. The attention networks capture the significance of intra-entity and inter-entity intersections to the final query space. Additionally, we provide a mechanism to comprehend ANTHEM and understand the significance of query entities towards the final resultant products. We evaluate the performance of our model on real data collected from popular e-commerce sites. Our experimental study on the offline data demonstrates compelling evidence of ANTHEM's superior performance over state-of-the-art product search methods with an improvement of more than 10% on various metrics. We also demonstrate the quality of ANTHEM's query encoder using a query matching task.

Human labeling is time-consuming and costly. This problem is further exacerbated in extremely imbalanced class label scenarios, such as detecting fraudsters in online websites. Active learning selects the most relevant example for human labelers to improve the model performance at a lower cost. However, existing methods for active learning for graph data often assumes that both data and label distributions are balanced. These assumptions fail in extreme rare-class classification scenarios, such as classifying abusive reviews in an e-commerce website.

Short text is ambiguous and often relies predominantly on the domain and context at hand in order to attain semantic relevance. Existing classification models perform poorly on short text due to data sparsity and inadequate context. Auxiliary context, which can often provide sufficient background regarding the domain, is typically available in several application scenarios. While some of the existing works aim to leverage real-world knowledge to enhance short-text representations, they fail to place appropriate emphasis on the auxiliary context. Such models do not harness the full potential of the available context in auxiliary sources. To address this challenge, we reformulate short-text classification as a dual channel self-supervised learning problem (that leverages auxiliary context) with a generation network and a corresponding prediction model. We propose a self-supervised framework, Pseudo-Auxiliary Context generation network for Short-text Modeling (PACS) , to comprehensively leverage auxiliary context and it is jointly learned with a prediction network in an end-to-end manner. Our PACS model consists of two sub-networks: a Context Generation Network (CGN) that models the auxiliary context’s distribution and a Prediction Network (PN) to map the short-text features and auxiliary context distribution to the final class label. Our experimental results on diverse datasets demonstrate that PACS outperforms formidable state-of-the-art baselines. We also demonstrate the performance of our model on cold-start scenarios (where contextual information is non-existent) during prediction. Furthermore, we perform interpretability and ablation studies to analyze various representational features captured by our model and the individual contribution of its modules to the overall performance of PACS, respectively.

Due to intense competition and lack of real estate on the front page of large e-commerce platforms, sellers are sometimes motivated to garner non-genuine signals (clicks, add-to-carts, purchases) on their products, to make them appear more appealing to customers. This hurts customers' trust on the platform, and also hurts genuine sellers who sell their items without looking to game the system. While it is important to find the sellers and the buyers who are colluding to garner these non-genuine signals, doing so is highly nontrivial. Firstly, the set of bad actors in the system is a very small fraction of all the buyers/sellers on the platform. Secondly, bad actors ``hide" with the good ones, making them hard to detect. In this paper, we develop CONGCN, a context aware heterogeneous graph convolutional network to detect bad actors on a large heterogeneous graph. While our method is motivated by abuse detection in e-commerce, the method is applicable to other areas such as computational biology and finance, where large heterogeneous graphs are pervasive, and the amount of labeled data is very limited. We train CONGCN via novel sampling methods, and context aware message passing in a semi-supervised fashion to predict dishonest buyers and sellers in e-commerce. Extensive experiments show that our method is effective, beating several baselines; generalizable to an inductive setting and highly scalable

Abusive behavior in online retail websites and communities threatens the experience of regular community members. Such behavior often takes place within a complex, dynamic, and large-scale network of users interacting with items. Detecting abuse is challenging due to the scarcity of labeled abuse instances and complexity of combining temporal and network patterns while operating at a massive scale. Previous approaches to dynamic graph modeling either do not scale, do not effectively generalize from a few labeled instances, or compromise performance for scalability. Here we present BiDyn, a general method to detect abusive behavior in dynamic bipartite networks at scale, while generalizing from limited training labels. BiDyn develops an efficient hybrid RNN-GNN architecture trained via a novel stacked ensemble training scheme. We also propose a novel pre-training framework for dynamic graphs that helps to achieve superior performance at scale. Our approach outperforms recent large-scale dynamic graph baselines in an abuse classification task by up to 14% AUROC while requiring 10x less memory per training batch in both open and proprietary datasets.

Logical reasoning over Knowledge Graphs (KGs) is a fundamental technique that can provide an efficient querying mechanism over large and incomplete databases. Current approaches employ spatial geometries such as boxes to learn query representations that encompass the answer entities and model the logical operations of projection and intersection. However, their geometry is restrictive and leads to non-smooth strict boundaries, which further results in ambiguous answer entities. Furthermore, previous works propose transformation tricks to handle unions which results in non-closure and, thus, cannot be chained in a stream. In this paper, we propose a Probabilistic Entity Representation Model (PERM) to encode entities as a Multivariate Gaussian density with mean and covariance parameters to capture its semantic position and smooth decision boundary, respectively. Additionally, we also define the closed logical operations of projection, intersection, and union that can be aggregated using an end-to-end objective function. On the logical query reasoning problem, we demonstrate that the proposed PERM significantly outperforms the state-of-the-art methods on various public benchmark KG datasets on standard evaluation metrics. We also evaluate PERM’s competence on a COVID-19 drug-repurposing case study and show that our proposed work is able to recommend drugs with substantially better F1 than current methods. Finally, we demonstrate the working of our PERM’s query answering process through a low-dimensional visualization of the Gaussian representations.

Knowledge Graphs (KGs) are ubiquitous structures for information storage in several real-world applications such as web search, e-commerce, social networks, and biology. Querying KGs remains a foundational and challenging problem due to their size and complexity. Promising approaches to tackle this problem include embedding the KG units (e.g., entities and relations) in a Euclidean space such that the query embedding contains the information relevant to its results. These approaches, however, fail to capture the hierarchical nature and semantic information of the entities present in the graph. Additionally, most of these approaches only utilize multi-hop queries (that can be modeled by simple translation operations) to learn embeddings and ignore more complex operations such as intersection, and union of simpler queries. To tackle such complex operations, in this paper, we formulate KG representation learning as a self-supervised logical query reasoning problem that utilizes translation, intersection and union queries over KGs. We propose Hyperboloid Embeddings (HypE), a novel self-supervised dynamic reasoning framework, that utilizes positive first-order existential queries on a KG to learn representations of its entities and relations as hyperboloids in a Poincaré ball. HypE models the positive first-order queries as geometrical translation, intersection, and union. For the problem of KG reasoning in real-world datasets, the proposed HypE model significantly outperforms the state-of-the art results. We also apply HypE to an anomaly detection task on a popular e-commerce website product taxonomy as well as hierarchically organized web articles and demonstrate significant performance improvements compared to existing baseline methods. Finally, we also visualize the learned HypE embeddings in a Poincaré ball to clearly interpret and comprehend the representation space.

Showing items that do not match search query intent degrades customer experience in ecommerce.These mismatches result from counterfactual biases of the ranking algorithms toward noisy behavioral signals such as clicks and purchases in the search logs.Mitigating the problem requires a large labeled dataset, which is expensive and time-consuming to obtain.In this paper, we develop a deep, endto-end model that learns to effectively classify mismatches and to generate hard mismatched examples to improve the classifier.We train the model end-to-end by introducing a latent variable into the cross-entropy loss that alternates between using the real and generated samples.This not only makes the classifier more robust but also boosts the overall ranking performance.Our model achieves a relative gain compared to baselines by over 26% in F-score, and over 17% in Area Under PR curve.On live search traffic, our model gains significant improvement in multiple countries.

Gradient Boosted Decision Trees (GBDTs) are widely used for building ranking and relevance models in search and recommendation. Considerations such as latency and interpretability dictate the use of as few features as possible to train these models. Feature selection in GBDT models typically involves heuristically ranking the features by importance and selecting the top few, or by per- forming a full backward feature elimination routine. On-the-fly feature selection methods proposed previously scale suboptimally with the number of features, which can be daunt- ing in high dimensional settings. We develop a scalable forward feature selection variant for GBDT, via a novel group testing procedure that works well in high dimensions, and enjoys favorable theoretical performance and computational guarantees. We show via ex- tensive experiments on both public and proprietary datasets that the proposed method offers significant speedups in training time, while being as competitive as existing GBDT methods in terms of model performance metrics. We also extend the method to the multitask setting, allowing the practitioner to select common features across tasks, as well as selecting task-specific features.

Product search forms an indispensable component of any e-commerce service, and helps customers find products of their interest from a large catalog on these websites. When products that are irrelevant to the search query are surfaced, it leads to a poor customer experience, thus reducing user trust and increasing the likelihood of churn. While identifying and removing such results from product search is crucial, doing so is a burdensome task that requires large amounts of human annotated data to train accurate models. This problem is exacerbated when products are cross-listed across countries that speak multiple languages, and customers specify queries in multiple languages and from different cultural contexts. In this work, we propose a novel multi-lingual multi-task learning framework, to jointly train product search models on multiple languages, with limited amount of training data from each language. By aligning the query and product representations from different languages into a language-independent vector space of queries and products, respectively, the proposed model improves the performance over baseline search models in any given language. We evaluate the performance of our model on real data collected from a leading e-commerce service. Our experimental evaluation demonstrates up to 23% relative improvement in the classification F1-score compared to the state-of-the-art baseline models.

E-commerce search engines are the primary means by which customers shop for products online. Each customer query contains multiple facets such as product type, color, brand, etc. A successful search engine retrieves products that are relevant to the query along each of these attributes. However, due to lexical (erroneous title, description, etc.) and behavioral irregularities (clicks or purchases of products that do not belong to the same facet as the query), some mismatched products are often included in search results. These irregularities can be detected using simple binary classifiers like gradient boosted decision trees or logistic regression. Typically, these binary classifiers use strong independence assumptions between the results and ignore structural relationships available in the data, such as the connections between products and queries. In this paper, we use the connections that exist between products and query to identify a special kind of structure we refer to as a micrograph. Further, we make use of Statistical Relational Learning (SRL) to incorporate these micrographs in the data and pose the problem as a structured prediction problem. We refer to this approach as structured mismatch classification (\SMC). In addition, we show that naive addition of structure does not improve the performance of the model and hence introduce a variation of \SMC, strong \SMC~(\SSMC), which improves over the baseline by passing information from high-confidence predictions to lower confidence predictions. In our empirical evaluation we show that our proposed approach outperforms the baseline classification methods by up to 12% in precision. Furthermore, we use quasi-Newton methods to make our method viable for real-time inference in a search engine and show that our approach is up to 150 times faster than existing ADMM-based solvers.

In this paper we advocate for a hyperparametric approach to learn diffusion in the independent cascade (IC) model. The sample complexity of this model is a function of the number of edges in the network and consequently learning becomes infeasible when the network is large. We study a natural restriction of the hypothesis class using additional information available in order to dramatically reduce the sample complexity of the learning process. In particular we assume that diffusion probabilities can be described as a function of a global hyperparameter and features of the individuals in the network. One of the main challenges with this approach is that training a model reduces to optimizing a non-convex objective. Despite this obstacle, we can shrink the best-known sample complexity bound for learning IC by a factor of |E|/d where |E| is the number of edges in the graph and d is the dimension of the hyperparameter. We show that under mild assumptions about the distribution generating the samples one can provably train a model with low generalization error. Finally, we use large-scale diffusion data from Facebook to show that a hyperparametric model using approximately 20 features per node achieves remarkably high accuracy.

Detecting large reshare cascades is an important problem in online social networks. There are a variety of attempts to model this problem, from using time series analysis methods to stochastic processes. Most of these approaches heavily depend on the underlying network features and use network information to detect the virality of cascades. In most cases, however, getting such detailed network information can be hard or even impossible.

Recommender systems have become increasingly popular in recent years because of the broader popularity of many web-enabled electronic commerce applications. However, most recommender systems today are designed in the context of an offline setting. The online setting is, however, much more challenging because the existing methods do not work very effectively for very large-scale systems. In many applications, it is desirable to provide real-time recommendations in large-scale scenarios. The main problem in applying streaming algorithms for recommendations is that the in-core storage space for memory-resident operations is quite limited. In this paper, we present a probabilistic neighborhood-based algorithm for performing recommendations in real-time. We present experimental results, which show the effectiveness of our approach in comparison to state-of-the-art methods.

Influence analysis is an important problem in social network analysis due to its impact on viral marketing and targeted advertisements. Most of the existing influence analysis methods determine the influencers in a static network with an influence propagation model based on pre-defined edge propagation probabilities. However, none of these models can be queried to find influencers in both context and time-sensitive fashion from a streaming social data. In this paper, we propose an approach to maintain real-time influence scores of users in a social stream using a topic and time-sensitive approach, while the network and topic is constantly evolving over time. We show that our approach is efficient in terms of online maintenance and effective in terms various types of real-time context- and time-sensitive queries. We evaluate our results on both social and collaborative network data sets.

The problem of discovering information flow trends in social networks has become increasingly relevant due to the increasing amount of content in online social networks, and its relevance as a tool for research into the content trends analysis in the network. An important part of this analysis is to determine the key patterns of flow in the underlying network. Almost all the work in this area has focused on fixed models of the network structure, and edge-based transmission between nodes. In this article, we propose a fully content-centered model of flow analysis in networks, in which the analysis is based on actual content transmissions in the underlying social stream, rather than a static model of transmission on the edges. First, we introduce the problem of influence analysis in the context of information flow in networks. We then propose a novel algorithm InFlowMine to discover the information flow patterns in the network and demonstrate the effectiveness of the discovered information flows using an influence mining application. This application illustrates the flexibility and effectiveness of our information flow model to find topic- or network-specific influencers, or their combinations. We empirically show that our information flow mining approach is effective and efficient than the existing methods on a number of different measures.

The problem of node classification in networks is an important one in a wide variety of social networking domains. In many real applications such as product recommendations, the class of interest may be very rare. In such scenarios, it is often very difficult to learn the most relevant node classification characteristics, both because of the paucity of training data, and because of poor connectivity among rare class nodes in the network structure. Node classification methods crucially dependent upon structural homophily, and a lack of connectivity among rare class nodes can create significant challenges. However, many such social networks are content-rich, and the content-rich nature of such networks can be leveraged to compensate for the lack of structural connectivity among rare class nodes. While content-centric and semi-supervised methods have been used earlier in the context of paucity of labeled data, the rare class scenario has not been investigated in this context. In fact, we are not aware of any known classification method which is tailored towards rare class detection in networks. This paper will present a spectral approach for rare-class detection, which uses a distance-preserving transform, in order to combine the structural information in the network with the available content. We will show the advantage of this approach over traditional methods for collective classification.

Link prediction is an important problem in online social and collaboration networks, for recommending friends and future collaborators. Most of the existing approaches for link prediction are focused on building unsupervised or supervised classification models based on the availability of accepts and rejects of the past recommendations. Several of these methods are feature-based and they construct a large number of network-level features to make the prediction more effective. A more flexible approach is to allow the model to learn the required features from the network for a specific task, rather than explicit feature engineering. In addition, most of the social and collaboration relationships do not happen instantly and rather build slowly over time through several low cost interactions, such as email and chat. The existing approaches often ignore the availability of such auxiliary networks to make link prediction more robust and effective. The main focus of this work is to build a robust and effective classifier for link prediction using multiple auxiliary networks. We develop a supervised random walk model, that does not require any explicit feature construction, and can be personalized to each user based on the past accept and reject behavior. Our approach consistently outperforms several popular baselines in terms of precision and recall in multiple real-life data sets. Also, our approach is robust to noise and sparsity in auxiliary networks, while several popular baselines, specifically feature-based ones, are inconsistent in their performance under such conditions.

Recommendation methods have mainly dealt with the problem of recommending new items to the user while user visitation behavior to the familiar items (items which have been consumed before) are little understood. In this paper, we analyze user activity streams and show that user's temporal consumption of familiar items is driven by boredom. Specifically, users move on to a different item when bored and return to the same item when their interest is restored. To model this behavior we include two latent psychological states of preference for items - sensitization and boredom. In the sensitization state the user is highly engaged with the item, while in the boredom state the user is disinterested. We model this behavior using a Hidden Semi-Markov Model for the gaps between user consumption activities. We show that our model performs much better than the state-of-the-art temporal recommendation models at predicting the revisit time to the item. Moreover, we attribute two main reasons for this: (1) recommending items that are not in the bored state for the user, (2) recommending items where user has restored her interests.

Understanding the dynamics of reciprocation is of great interest in sociology and computational social science. The recent growth of Massively Multi-player Online Games (MMOGs) has provided unprecedented access to large-scale data which enables us to study such complex human behavior in a more systematic manner. In this paper, we consider three different networks in the EverQuest2 game: chat, trade, and trust. The chat network has the highest level of reciprocation (33%) because there are essentially no barriers to it. The trade network has a lower rate of reciprocation (27%) because it has the obvious barrier of requiring goods or money for exchange; morever, there is no clear benefit to returning a trade link except in terms of social connections. The trust network has the lowest reciprocation (14%) because this equates to sharing certain within-game assets such as weapons, and so there is a high barrier for such connections In general, we observe that reciprocation rate is inversely related to the barrier level in these networks. We also note that reciprocation has connections across the heterogeneous networks. Our experiments indicate that players make use of the medium-barrier reciprocations to strengthen a relationship. We hypothesize that lower-barrier interactions are an important component to predicting higher-barrier ones. We verify our hypothesis using predictive models for trust reciprocations with features from trade interactions. Incorporating the number of trades (both before and after the initial trust link) boosts our ability to predict if the trust will be reciprocated up to 11% with respect to the AUC. More generally, we see strong correlations across the different networks and emphasize that network dynamics, such as reciprocation, cannot be studied in isolation on just a single type of connection.

The existing methods for finding influencers use the process of information diffusion to discover the nodes with maximum information spread. These models capture only the process of information diffusion and not the actual social value of collaborations in the network. We have proposed a method for finding influencers using the idea that people generate more value for their work by collaborating with peers of high influence. The social value generated through such collaborations denotes the notion of individual social capital. We hypothesize and show that players with high social capital are often key influencers in the network. We propose a value-allocation model to compute the social capital and allocate the fair share of this capital to each individual involved in the collaboration. We show that our allocation satisfies several axioms of fairness and falls in the same class as the Myerson's allocation function. We implement our allocation rule using an efficient algorithm SoCap and show that our algorithm outperforms the baselines in several real-life data sets. Specifically, in DBLP network, our algorithm outperforms PageRank, PMIA and Weighted Degree baselines up to 8% in terms of precision, recall and F1-measure.

The problem of discovering information flow trends and influencers in social networks has become increasingly relevant both because of the increasing amount of content available from online networks in the form of social streams, and because of its relevance as a tool for content trends analysis. An important part of this analysis is to determine the key patterns of flow and corresponding influencers in the underlying network. Almost all the work on influence analysis has focused on fixed models of the network structure, and edge-based transmission between nodes. In this paper, we propose a fully content-centered model of flow analysis in social network streams, in which the analysis is based on actual content transmissions in the network, rather than a static model of transmission on the edges. First, we introduce the problem of information flow mining in social streams, and then propose a novel algorithm InFlowMine to discover the information flow patterns in the network. We then leverage this approach to determine the key influencers in the network. Our approach is flexible, since it can also determine topic-specific influencers. We experimentally show the effectiveness and efficiency of our model.

The problem of community detection is a challenging one because of the presence of hubs and noisy links, which tend to create highly imbalanced graph clusters. Often, these resulting clusters are not very intuitive and difficult to interpret. With the growing availability of network information, there is a significant amount of prior knowledge available about the communities in social, communication and several other networks. These community labels may be noisy and very limited, though they do help in community detection. In this paper, we explore the use of such noisy labeled information for finding high quality communities. We will present an adaptive density-based clustering which allows flexible incorporation of prior knowledge in to the community detection process. We use a random walk framework to compute the node densities and the level of supervision regulates the node densities and the quality of resulting density based clusters. Our framework is general enough to produce both overlapping and non-overlapping clusters. We empirically show that even with a tiny amount of supervision, our approach can produce superior communities compared to popular baselines.

There are several Global Climate Models (GCM) reported by various countries to the Intergovernmental Panel on Climate Change (IPCC). Due to the varied nature of the GCM model assumptions, the future projections of the GCMs show high variability which makes it difficult to come up with confident projections into the future. Climate scientists combine these multiple GCM models to minimize the variability and the prediction error. Most of these model combinations are specifically for a location, or at a global scale. They do not consider regional or local smoothing (including the IPCC model). In this paper, we address this problem of combining multiple GCM model outputs with spatial smoothing as an important desired criterion. The problem formulation takes the form of multiple least squares regression for each geographic location with graph Laplacian based smoothing amongst the neighboring locations. Unlike the existing Laplacian regression frameworks, our formulation has both inner and outer products of the coefficient matrix, and has Sylvester equations as its special case. We discuss a few approaches to solve the problem, including a closed-form by solving a large linear system, as well as gradient descent methods which turn out to be more efficient. We establish the superiority of our approach in terms of model accuracy and smoothing compared to several popular baselines on real GCM climate datasets.

This paper aims to solve the problem of improving the ranking of answer candidates for factoid based questions in a state-of-the-art Question Answering system. We first provide an extensive comparison of 5 ranking algorithms on two datasets -- from the Jeopardy quiz show and a medical domain. We then show the effectiveness of a cascading approach, where the ranking produced by one ranker is used as input to the next stage. The cascading approach shows sizeable gains on both datasets. We finally evaluate several rank aggregation techniques to combine these algorithms, and find that Supervised Kemeny aggregation is a robust technique that always beats the baseline ranking approach used by Watson for the Jeopardy competition. We further corroborate our results on TREC Question Answering datasets.

Matrix factorization from a small number of observed entries has recently garnered much attention as the key ingredient of successful recommendation systems. One unresolved problem in this area is how to adapt current methods to handle changing user preferences over time. Recent proposals to address this issue are heuristic in nature and do not fully exploit the time-dependent structure of the problem. As a principled and general temporal formulation, we propose a dynamical state space model of matrix factorization. Our proposal builds upon probabilistic matrix factorization, a Bayesian model with Gaussian priors. We utilize results in state tracking, i.e. the Kalman filter, to provide accurate recommendations in the presence of both process and measurement noise. We show how system parameters can be learned via expectation-maximization and provide comparisons to current published techniques.

Social networks generate a large amount of text content over time because of continuous interaction between participants. The mining of such social streams is more challenging than traditional text streams, because of the presence of both text content and implicit network structure within the stream. The problem of event detection is also closely related to clustering, because the events can only be inferred from aggregate trend changes in the stream. In this paper, we will study the two related problems of clustering and event detection in social streams. We will study both the supervised and unsupervised case for the event detection problem. We present experimental results illustrating the effectiveness of incorporating network structure in event discovery over purely content-based methods.

A customer reported problem (or trouble ticket) in software maintenance is typically solved by one or more maintenance engineers. The decision of allocating the ticket to one or more engineers is generally taken by the lead, based on customer delivery deadlines and a guided complexity assessment from each maintenance engineer. The key challenge in such a scenario is two folds, un-truthful (hiked up) elicitation of ticket complexity by each engineer to the lead and the decision of allocating the ticket to a group of engineers who will solve the ticket with in customer deadline. The decision of allocation should ensure individual and coalitional rationality along with coalitional stability. In this paper we use game theory to examine the issue of truthful elicitation of ticket complexities by engineers for solving ticket as a group given a specific customer delivery deadline. We formulate this problem as strategic form game and propose two mechanisms, (1) division of labor (DOL) and (2) extended second price (ESP). In the proposed mechanisms we show that truth telling by each engineer constitutes a dominant strategy Nash equilibrium of the underlying game. Also we analyze the existence of individual rationality (IR) and coalitional rationality (CR) properties to motivate voluntary and group participation. We use Core, solution concept from co-operative game theory to analyze the stability of the proposed group based on the allocation and payments.