Many critical web applications, from e-commerce price prediction to user engagement forecasting, rely on regression models trained on tabular data. These models often face a dual challenge: the inherent imbalance in continuous target values and, more critically, the unpredictable distribution shifts that occur when the model is deployed online. While data imbalance in classification is well-studied, its intersection with regression tasks in dynamic, real-world settings is underexplored. Existing methods for imbalanced regression often assume that the test data distribution is known and stable, an assumption that rarely holds true for live web systems and can lead to significant performance degradation. To address this gap, we propose a novel framework featuring two key innovations: (i) a Region-Aware Mixture of Experts that leverages a Gaussian Mixture Model to identify distinct data sub-populations. This allows us to synthesize targeted training data and train specialized experts, each tailored to a specific data region. (ii) a Test-Time Self-Supervised Expert Aggregation mechanism. This is the core of our adaptation strategy, dynamically adjusting the weights of each expert based on the features of incoming test instances. This enables our model to adapt on-the-fly to varying test distributions without costly retraining. We evaluated our method on four real-world tabular regression datasets: house pricing, bike sharing, and age prediction. These tasks are representative of real-world scenarios that inherently involve both target imbalance and dynamic distribution shifts (e.g., temporal or market-driven changes). The results demonstrate that our approach significantly outperforms existing imbalanced regression methods, especially under these shifts, achieving an average MAE improvement of 7.1%.

Multivariate time series data, collected across various fields such as manufacturing and wearable technology, exhibit states at multiple levels of granularity, from coarse-grained system behaviors to fine-grained, detailed events. Effectively segmenting and integrating states across these different granularities is crucial for tasks like predictive maintenance and performance optimization. However, existing time series segmentation methods face two key challenges: (1) the inability to handle multiple levels of granularity within a unified model, and (2) limited adaptability to new, evolving patterns in dynamic environments. To address these challenges, we propose PromptTSS, a novel framework for time series segmentation with multi-granularity states. PromptTSS uses a unified model with a prompting mechanism that leverages label and boundary information to guide segmentation, capturing both coarse- and fine-grained patterns while adapting dynamically to unseen patterns. Experiments show PromptTSS improves accuracy by 24.49% in multi-granularity segmentation, 17.88% in single-granularity segmentation, and up to 599.24% in transfer learning, demonstrating its adaptability to hierarchical states and evolving time series dynamics. Our code is available at https://github.com/blacksnail789521/PromptTSS.

Tailoring structured financial reports from companies' earnings releases is crucial for understanding financial performance and has been widely adopted in real-world analytics. However, existing summarization methods often generate broad, high-level summaries, which may lack the precision and detail required for financial reports that typically focus on specific, structured sections. While Large Language Models (LLMs) hold promise, generating reports adhering to predefined multi-section templates remains challenging. This paper investigates two LLM-based approaches popular in industry for generating templated financial reports: an agentic information retrieval (IR) framework and a decomposed IR approach, namely AgenticIR and DecomposedIR. The AgenticIR utilizes collaborative agents prompted with the full template. In contrast, the DecomposedIR approach applies a prompt chaining workflow to break down the template and reframe each section as a query answered by the LLM using the earnings release. To quantitatively assess the generated reports, we evaluated both methods in two scenarios: one using a financial dataset without direct human references, and another with a weather-domain dataset featuring expert-written reports. Experimental results show that while AgenticIR may excel in orchestrating tasks and generating concise reports through agent collaboration, DecomposedIR statistically significantly outperforms AgenticIR approach in providing broader and more detailed coverage in both scenarios, offering reflection on the utilization of the agentic framework in real-world applications.

Multivariate time-series forecasting is vital in various domains, e.g., economic planning and weather prediction. Deep train-from-scratch models have exhibited effective performance yet require large amounts of data, which limits real-world applicability. Recently, researchers have leveraged the representation learning transferability of pre-trained Large Language Models (LLMs) to handle limited non-linguistic datasets effectively. However, incorporating LLMs with time-series data presents challenges of limited adaptation due to different compositions between time-series and linguistic data, and the inability to process multi-scale temporal information. To tackle these challenges, we propose LLM4TS, a framework for time-series forecasting with pre-trained LLMs. LLM4TS consists of a two-stage fine-tuning strategy: the time-series alignment stage to align LLMs with the nuances of time-series data and the forecasting fine-tuning stage for downstream time-series forecasting tasks. Furthermore, our framework features a novel two-level aggregation method that integrates multi-scale temporal data within pre-trained LLMs, enhancing their ability to interpret time-specific information. In experiments across seven time-series forecasting datasets, LLM4TS is superior to existing state-of-the-art methods compared with trained-from-scratch models in full-shot scenarios and also achieves the highest rank in few-shot scenarios. In addition, evaluations compared with different unsupervised representation learning approaches highlight LLM4TS’s effectiveness with representation learning in forecasting tasks. Ablation studies further validate each component’s contribution to LLM4TS and underscore the essential role of utilizing LLM’s pre-trained weights for optimal performance. The code is available at https://github.com/blacksnail789521/LLM4TS .

Electronic health records (EHRs) are multimodal by nature, consisting of structured tabular features like lab tests and unstructured clinical notes. In real-life clinical practice, doctors use complementary multimodal EHR data sources to get a clearer picture of patients' health and support clinical decision-making. However, most EHR predictive models do not reflect these procedures, as they either focus on a single modality or overlook the inter-modality interactions/redundancy. In this work, we propose MEDFuse, a Multimodal EHR Data Fusion framework that incorporates masked lab-test modeling and large language models (LLMs) to effectively integrate structured and unstructured medical data. MEDFuse leverages multimodal embeddings extracted from two sources: LLMs fine-tuned on free clinical text and masked tabular transformers trained on structured lab test results. We design a disentangled transformer module, optimized by a mutual information loss to 1) decouple modality-specific and modality-shared information and 2) extract useful joint representation from the noise and redundancy present in clinical notes. Through comprehensive validation on the public MIMIC-III dataset and the in-house FEMH dataset, MEDFuse demonstrates great potential in advancing clinical predictions, achieving over 90% F1 score in the 10-disease multi-label classification task.

Multivariate time-series data in numerous real-world applications (e.g., healthcare and industry) are informative but challenging due to the lack of labels and high dimensionality. Recent studies in self-supervised learning have shown their potential in learning rich representations without relying on labels, yet they fall short in learning disentangled embeddings and addressing issues of inductive bias (e.g., transformation-invariance). To tackle these challenges, we propose TimeDRL, a generic multivariate time-series representation learning frame-work with disentangled dual-level embeddings. TimeDRL is characterized by three novel features: (i) disentangled derivation of timestamp-level and instance-level embeddings from patched time-series data using a [CLS] token strategy; (ii) utilization of timestamp-predictive and instance-contrastive tasks for disentangled representation learning, with the former optimizing timestamp-level embeddings with predictive loss, and the latter optimizing instance-level embeddings with contrastive loss; and (iii) avoidance of augmentation methods to eliminate inductive biases, such as transformation-invariance from cropping and masking. Comprehensive experiments on 6 time-series forecasting datasets and 5 time-series classification datasets have shown that TimeDRL consistently surpasses existing representation learning approaches, achieving an average improvement of forecasting by 58.02% in MSE and classification by 1.48% in accuracy. Further-more, extensive ablation studies confirmed the relative contribution of each component in TimeDRL's architecture, and semi-supervised learning evaluations demonstrated its effectiveness in real-world scenarios, even with limited labeled data. The code is available at https://github.com/blacksnail789521/TimeDRL.

The significant increase in credit card transactions can be attributed to the rapid growth of online shopping and digital payments, particularly during the COVID-19 pandemic. To safeguard cardholders, e-commerce companies, and financial institutions, the implementation of an effective and real-time fraud detection method using modern artificial intelligence techniques is imperative. However, the development of machine-learning-based approaches for fraud detection faces challenges such as inadequate transaction representation, noise labels, and data imbalance. Additionally, practical considerations like dynamic thresholds, concept drift, and verification latency need to be appropriately addressed. In this study, we designed a fraud detection method that accurately extracts a series of spatial and temporal representative features to precisely describe credit card transactions. Furthermore, several auxiliary self-supervised objectives were developed to model cardholders’ behavior sequences. By employing intelligent sampling strategies, potential noise labels were eliminated, thereby reducing the level of data imbalance. The developed method encompasses various innovative functions that cater to practical usage requirements. We applied this method to two real-world datasets, and the results indicated a higher F1 score compared to the most commonly used online fraud detection methods.

Personalized medication recommendations aim to suggest a set of medications based on the clinical conditions of a patient. Not only should the patient’s diagnosis, procedure, and medication history be considered, but drug-drug interactions (DDIs) must also be taken into account to prevent adverse drug reactions. Although recent studies on medication recommendation have considered DDIs and patient history, personalized disease progression and prescription have not been explicitly modeled. In this work, we proposed FastRx, a Fastformer-based medication recommendation model to capture longitudinality in patient history, in combination with Graph Convolutional Networks (GCNs) to handle DDIs and co-prescribed medications in Electronic Health Records (EHRs). Our extensive experiments on the MIMIC-III dataset demonstrated superior performance of the proposed FastRx over existing state-of-the-art models for medication recommendation. The source code and data used in the experiments are available at https://github.com/pnmthaoct/FastRx.

In recent years, there has been a significant surge in commercial demand for citation graph-based tasks, such as patent analysis, social network analysis, and recommendation systems. Graph Neural Networks (GNNs) are widely used for these tasks due to their remarkable performance in capturing topological graph information. However, GNNs’ output results are highly dependent on the composition of local neighbors within the topological structure. To address this issue, we identify two types of neighbors in a citation graph: explicit neighbors based on the topological structure and implicit neighbors based on node features. Our primary motivation is to clearly define and visualize these neighbors, emphasizing their importance in enhancing graph neural network performance. We propose a Correlation-aware Network (CNet) to re-organize the citation graph and learn more valuable informative representations by leveraging these implicit and explicit neighbors. Our approach aims to improve graph data augmentation and classification performance, with the majority of our focus on stating the importance of using these neighbors, while also introducing a new graph data augmentation method. We compare CNet with state-of-the-art (SOTA) GNNs and other graph data augmentation approaches acting on GNNs. Extensive experiments demonstrate that CNet effectively extracts more valuable informative representations from the citation graph, significantly outperforming baselines. The code is available on public GitHub. 1

The marketplace system connecting demands and supplies has been explored to develop unbiased decision-making in valuing properties. Real estate appraisal serves as one of the high-cost property valuation tasks for financial institutions since it requires domain experts to appraise the estimation based on the corresponding knowledge and the judgment of the market. Existing automated valuation models reducing the subjectivity of domain experts require a large number of transactions for effective evaluation, which is predominantly limited to not only the labeling efforts of transactions but also the generalizability of new developing and rural areas. To learn representations from unlabeled real estate sets, existing self-supervised learning (SSL) for tabular data neglects various important features, and fails to incorporate domain knowledge. In this paper, we propose DoRA, a Domain-based self-supervised learning framework for low-resource Real estate Appraisal. DoRA is pre-trained with an intra-sample geographic prediction as the pretext task based on the metadata of the real estate for equipping the real estate representations with prior domain knowledge. Furthermore, inter-sample contrastive learning is employed to generalize the representations to be robust for limited transactions of downstream tasks. Our benchmark results on three property types of real-world transactions show that DoRA significantly outperforms the SSL baselines for tabular data, the graph-based methods, and the supervised approaches in the few-shot scenarios by at least 7.6% for MAPE, 11.59% for MAE, and 3.34% for HR10%. We expect DoRA to be useful to other financial practitioners with similar marketplace applications who need general models for properties that are newly built and have limited records. The source code is available at https://github.com/wwweiwei/DoRA.

With the recent progress in sports analytics, deep learning approaches have demonstrated the effectiveness of mining insights into players' tactics for improving performance quality and fan engagement. This is attributed to the availability of public ground-truth datasets. While there are a few available datasets for turn-based sports for action detection, these datasets severely lack structured source data and stroke-level records since these require high-cost labeling efforts from domain experts and are hard to detect using automatic techniques. Consequently, the development of artificial intelligence approaches is significantly hindered when existing models are applied to more challenging structured turn-based sequences. In this paper, we present ShuttleSet, the largest publicly-available badminton singles dataset with annotated stroke-level records. It contains 104 sets, 3,685 rallies, and 36,492 strokes in 44 matches between 2018 and 2021 with 27 top-ranking men's singles and women's singles players. ShuttleSet is manually annotated with a computer-aided labeling tool to increase the labeling efficiency and effectiveness of selecting the shot type with a choice of 18 distinct classes, the corresponding hitting locations, and the locations of both players at each stroke. In the experiments, we provide multiple benchmarks (i.e., stroke influence, stroke forecasting, and movement forecasting) with baselines to illustrate the practicability of using ShuttleSet for turn-based analytics, which is expected to stimulate both academic and sports communities. Over the past two years, a visualization platform has been deployed to illustrate the variability of analysis cases from ShuttleSet for coaches to delve into players' tactical preferences with human-interactive interfaces, which was also used by national badminton teams during multiple international high-ranking matches.

Identifying significant shots in a rally is important for evaluating players’ performance in badminton matches. While there are several studies that have quantified player performance in other sports, analyzing badminton data has remained untouched. In this article, we introduce a badminton language to fully describe the process of the shot, and we propose a deep-learning model composed of a novel short-term extractor and a long-term encoder for capturing a shot-by-shot sequence in a badminton rally by framing the problem as predicting a rally result. Our model incorporates an attention mechanism to enable the transparency between the action sequence and the rally result, which is essential for badminton experts to gain interpretable predictions. Experimental evaluation based on a real-world dataset demonstrates that our proposed model outperforms the strong baselines. We also conducted case studies to show the ability to enhance players’ decision-making confidence and to provide advanced insights for coaching, which benefits the badminton analysis community and bridges the gap between the field of badminton and computer science.

We study the forecasting problem for traffic with dynamic, possibly periodical, and joint spatial-temporal dependency between regions. Given the aggregated inflow and outflow traffic of regions in a city from time slots 0 to$t - 1$, we predict the traffic at time$t$for any region. Prior arts in the area often considered the spatial and temporal dependencies in a decoupled manner, or were rather computationally intensive in training with a large number of hyper-parameters which needed tuning. We propose ST-TIS, a novel, lightweight and accurateSpatial-TemporalTransformer withinformation fusion and regionsampling for traffic forecasting. ST-TIS extends the canonical Transformer with information fusion and region sampling. The information fusion module captures the complex spatial-temporal dependency between regions. The region sampling module is to improve the efficiency and prediction accuracy, cutting the computation complexity for dependency learning from$O(n^{2})$to$O(n\sqrt{n})$, where$n$is the number of regions. With far fewer parameters than state-of-the-art deep learning models, ST-TIS's offline training is significantly faster in terms of tuning and computation (with a reduction of up to$90\%$on training time and network parameters). Notwithstanding such training efficiency, extensive experiments show that ST-TIS is substantially more accurate in online prediction than state-of-the-art approaches (with an average improvement of$9.5\%$on RMSE, and$12.4\%$on MAPE compared to STDN and DSAN).

Docked bike systems have been widely deployed in many cities around the world. To the service provider, predicting the demand and supply of bikes at any station is crucial to offering the best service quality. The docked bike prediction problem is highly challenging because of the complicated joint spatial-temporal (ST) dependency as bikes are picked up and dropped off, the so-called “flows”, between stations. Prior works often considered the spatial and temporal dependencies separately using sequential network models, and based on locality assumptions. Without sufficiently capturing the joint spatial and temporal features, these approaches are not optimal for attaining the best prediction accuracy. We propose STGNN-DJD, a novel data-driven Spatial-Temporal Graph Neural Network to solve the bike demand and supply prediction problem by unifiedly embedding the Dynamic and Joint ST Dependency in two novel ST graphs. Given station locations and historical rental data on bike flow over the past time slots 0 to$t-1$, we seek to predict online the bike demand and supply at any station at time$t$. To extract joint spatial-temporal dependency, STGNN-DJD employs a graph generator to construct, at the beginning of time$t$, two graphs that embed the flow relationships between stations at various time slots (flow-convoluted graph) and dynamic demand-supply pattern correlation between stations (pattern correlation graph), respectively. Given the two spatial-temporal graphs, STGNN-DJD subsequently employs a graph neural network with novel flow-based and attention-based aggregators to generate embedding of each station for docked bike prediction. We have conducted extensive experiments on two large bike-sharing datasets. Our re-sults confirm the effectiveness of STGNN-DJD as compared with other state-of-the-art approaches, with significant improvement on RMSE and MAE (by 20%-50%). We also provide a case study on dynamic dependencies between stations and demonstrate that the locality assumption does not always hold for a docked bike system.

The customer base is the most valuable resource to E-commerce companies. A comprehensive understanding of customers’ preferences and behavior is crucial to developing good marketing strategies, in order to achieve optimal customer lifetime values (CLVs). For example, by exploring customer behavior patterns, given a marketing plan with a limited budget, a set of potential customers is able to be identified to maximize profit. In other words, personalized campaigns at the right time and in the right place can be treated as the last stage of consumption. Moreover, effective future purchase estimation and recommendation help guide the customer to the up-selling stage. The proposed willing-to-pay prediction model (W2P) exploits the transaction data to predict customer payment behavior based on a probabilistic graphical model, which provides semantic explanation of the estimated results and deals with the sparsity of payment data from each customer. Existing work in this domain ranks the customers by their probabilities of purchase in different conditions. However, the customer with the highest purchase probability does not necessarily spend the most. Therefore, we propose a CLV maximization algorithm based on the prediction results. In addition, we improve the model by behavioral segmentation wherein we group the customers by payment behaviors to reduce the size of the offline models and enhance the accuracy for low-frequency customers. The experiment results show that our model outperforms the state-of-the-art methods in purchase behavior prediction.

With the rapid advances and the penetration of the Internet of Things and sensors, a massive amount of trajectory data, given by discrete locations at certain timestamps, have been extracted or collected. Knowing the similarity between trajectories is fundamental to understanding their spatial-temporal correlation, with direct and far-reaching applications in contact tracing, companion detection, personalized marketing, etc. In this work, we consider the general and realistic sensing scenario that the locations of the trajectories may be noisy, and that these trajectories are sporadically sampled with randomness and asynchrony from the underlying continuous paths. Most of the prior work on trajectory similarity has not sufficiently considered the temporal dimension, or the issues of location noise and sporadic sampling, while others have limitations of strong assumptions such as a fixed known speed of users or the availability of a large amount of training data.We propose a novel and effective spatial-temporal measure termed STS (Spatial-Temporal Similarity) to evaluate the spatial-temporal overlap between any two trajectories. In order to account for the location noise and sporadic sampling, STS models each location in a trajectory as an observable outcome drawn from a probability distribution. With that, it efficiently reduces the need for training data by estimating a personalized spatial-temporal probability distribution of the object position from its own trajectory. Based on that, it subsequently computes the co-location probability and hence derives the similarity of any two trajectories. We have conducted extensive experiments to evaluate STS using real large-scale indoor (mall) and outdoor (taxi) datasets. Our results show that STS is substantially more accurate and robust than the state-of-the-art approaches, with an improvement of 63% on precision and 85% on mean rank.

Identifying significant shots in a rally is important for evaluating players’ performance in badminton matches. While there are several studies that have quantified player performance in other sports, analyzing badminton data is remained untouched. In this paper, we introduce a badminton language to fully describe the process of the shot and propose a deep learning model composed of a novel short-term extractor and a long-term encoder for capturing a shot-by-shot sequence in a badminton rally by framing the problem as predicting a rally result. Our model incorporates an attention mechanism to enable the transparency of the action sequence to the rally result, which is essential for badminton experts to gain interpretable predictions. Experimental evaluation based on a real-world dataset demonstrates that our proposed model outperforms the strong baselines. The source code is publicly available at https://github.com/wywyWang/Shot-Influence.

Mining data collected from industrial manufacturing process plays an important role for intelligent manufacturing in Industry 4.0. In this paper, we propose a deep convolutional model for predicting wafer fabrication quality in an intelligent integrated-circuit manufacturing application. The wafer fabrication quality prediction is motivated by the need for improving product line efficiency and reducing manufacturing cost by detecting potential defective work-in-process (WIP) wafers. This work considers the following two crucial data characteristics for wafer fabrication. First, our model is designed to learn spatial correlation between quality measurements on WIP wafers and fabrication results through an encoder-decoder neural network. Second, we leverage the fact that different products share the same raw manufacturing process to enable the knowledge transferring between prediction models of different products. Performance evaluation on real data sets is conducted to validate the strengths of our model on quality prediction, model interpretability, and feasibility of transferring knowledge.

In various web applications like targeted advertising and recommender systems, the available categorical features (e.g., product type) are often of great importance but sparse. As a widely adopted solution, models based on Factorization Machines (FMs) are capable of modelling high-order interactions among features for effective sparse predictive analytics. As the volume of web-scale data grows exponentially over time, sparse predictive analytics inevitably involves dynamic and sequential features. However, existing FM-based models assume no temporal orders in the data, and are unable to capture the sequential dependencies or patterns within the dynamic features, impeding the performance and adaptivity of these methods. Hence, in this paper, we propose a novel Sequence-Aware Factorization Machine (SeqFM) for temporal predictive analytics, which models feature interactions by fully investigating the effect of sequential dependencies. As static features (e.g., user gender) and dynamic features (e.g., user interacted items) express different semantics, we innovatively devise a multi-view self-attention scheme that separately models the effect of static features, dynamic features and the mutual interactions between static and dynamic features in three different views. In SeqFM, we further map the learned representations of feature interactions to the desired output with a shared residual network. To showcase the versatility and generalizability of SeqFM, we test SeqFM in three popular application scenarios for FM-based models, namely ranking, classification and regression tasks. Extensive experimental results on six large-scale datasets demonstrate the superior effectiveness and efficiency of SeqFM.

An alternative to current mainstream preprocessing methods is proposed: Value Selection (VS). Unlike the existing methods such as feature selection that removes features and instance selection that eliminates instances, value selection eliminates the values (with respect to each feature) in the dataset with two purposes: reducing the model size and preserving its accuracy. Two probabilistic methods based on information theory's metric are proposed: PVS and P+VS. Extensive experiments on the benchmark datasets with various sizes are elaborated. Those results are compared with the existing preprocessing methods such as feature selection, feature transformation, and instance selection methods. Experiment results show that value selection can achieve the balance between accuracy and model size reduction.

Network embedding has been proven effective to learn low-dimensional vector representations for network vertices, and recently received a tremendous amount of research attention. However, most of existing methods for network embedding merely focus on preserving the first and second order proximities between nodes, and the important properties of node centrality are neglected. Various centrality measures such as Degree, Closeness, Betweenness, Eigenvector and PageRank centralities have been designed to measure the importance of individual nodes. In this paper, we focus on a novel yet unsolved problem that aims to learn low-dimensional continuous nodes representations that not only preserve the network structure, but also keep the centrality information. We propose a generalizable model, namely GraphCSC, that utilizes both linkage information and centrality information to learn low-dimensional vector representations for network vertices. The learned embeddings by GraphCSC are able to preserve different centrality information of nodes. In addition, we further propose GraphCSC-M, a more comprehensive model that can preserve different centrality information simultaneously through learning multiple centrality-specific embeddings, and a novel attentive multi-view learning approach is developed to compress multiple embeddings of one node into a compact vector representation. Extensive experiments have been conducted to demonstrate that our model is able to preserve different centrality information of nodes, and achieves better performance on several benchmark tasks compared with recent state-of-the-art network embedding methods.

In this paper, we investigate the problem of mining multivariate time series data generated from sensors mounted on manufacturing stations for early product quality prediction. In addition to accurate quality prediction, another crucial requirement for industrial production scenarios is model interpretability, i.e., to understand the significance of an individual time series with respect to the final quality. Aiming at the goals, this paper proposes a multi-task learning model with an encoder-decoder architecture augmented by the matrix factorization technique and the attention mechanism. Our model design brings two major advantages. First, by jointly considering the input multivariate time series reconstruction task and the quality prediction in a multi-task learning model, the performance of the quality prediction task is boosted. Second, by incorporating the matrix factorization technique, we enable the proposed model to pay/learn attentions on the component of the multivariate time series rather than on the time axis. With the attention on components, the correlation between a sensor reading and a final quality measure can be quantized to improve the model interpretability. Comprehensive performance evaluation on real data sets is conducted. The experimental results validate that strengths of the proposed model on quality prediction and model interpretability.

Public transportation is beating heart of a city. Understanding how citizens utilize public transportation can be used to optimize many applications such as traffic planning, crowd flow prediction, and location-based marketing. However, obtaining how citizens used transportation is not a trivial task. It is almost not possible to ask citizens to report their exact location and their transportation mode; moreover, there are usually various public transportation that move along the similar paths. These increase challenges to identify people's transport modes. To address these issues, this paper proposes Public Transport General Framework (PTGF) to identify people's transport modes by their cellular data in both offline and online manners. Regarding the offline phase, given historical cellular data of people and urban transportation networks, PTGF derives cellular data into trajectories, to match each trajectory to public transportation networks to find the most possible transport modes for sub-trajectories of a trajectory. In the online phase, given streaming trajectories, PTGF identifies the transport modes of each location by an LSTM which are trained by historical trajectories with transport modes annotated in the offline phase. Extensive experiments are conducted by using both synthetic and real datasets. The experimental results show that the accuracy of PTGF in offline phase around 80% and that in online phase F1-score around 0.7, which could prove that the effectiveness of the proposed framework PTGF.

The improvement of collision avoidance for ship navigation in encounter situation is an important topic in maritime traffic safety. Most research on maritime collision avoidance has focused on planning a safe path for a ship to keep away from the approaching ship under the requirements of the International Regulations for Preventing Collision at Sea (COLREGs). However, the specific anti-collision actions are actually carried out by the navigators' own experience according to the local encounter situation.

With the advances in the development of mobile payments, a huge amount of payment data are collected by banks. User payment data offer a good dataset to depict customer behavior patterns. A comprehensive understanding of customers' purchase behavior is crucial to developing good marketing strategies, which may trigger much greater purchase amounts. For example, by exploring customer behavior patterns, given a target store, a set of potential customers is able to be identified.

In Real-Time Bidding (RTB) advertising, estimating the winning price is an important task in evaluating the bid cost of bid requests in Demand-Side Platforms (DSPs). The prior works utilize censored linear regression for winning price estimation by considering both winning and losing bid records. In the traditional regression models, the winning price of each bid request is based on Gaussian distribution. However, the property of Gaussian distribution is not suitable for the winning price of each bid request, and it is hard to link the physical meaning of Gaussian distribution and the winning price. Therefore, in this paper, based on our observation and analysis, the winning price of each bid request is modeled by a unique gamma distribution with respect to its features. Then we propose a gamma-based censored linear regression with regularization for winning price estimation. To derive the parameters of our proposed complicated model based on bid records, our approach is to divide this hard problem into two sub-problems, which are easier to solve. In practice, we also provide four heuristic initial parameter settings that are able to greatly reduce the computation cost when deriving the parameters. The experimental results demonstrate that our approach is highly effective for estimating the winning price compared with the state-of-the-art approaches in three real datasets.

Public transportation is essential in people's daily life and it is crucial to understand how people move around the city. Some prior works have exploited GPS, Wi-Fi or bluetooth to collect data, in which extra sensors or devices were needed. Other works utilized data from smart card systems. However, some public transportation systems have their own smart card system and the smart card data cannot include all kinds of transportation modes, which makes it unsuitable for our study.Nowadays, each user has his/her own mobile phones and from the cellular data of mobile phone service providers, it is possible to know the uses' transportation mode and the fine-grained crowd flows. As such, given a set of cellular data, we propose a system for public transportation mode detection, crowd density estimation, and crowd flow estimation. Note that we only have cellular data, no extra sensor data collected from users' mobile phones. In this paper, we refer to some external data sources (e.g., the bus routing networks) to identify transportation modes. Users' cellular data sometimes have uncertainty about user location information. Thus, we propose two approaches for different transportation mode detection considering the cell tower properties, spatial and temporal factors. We demonstrate our system using the data from Chunghwa Telecom, which is the largest telecommunication company in Taiwan, to show the usefulness of our system.

Collision-free is one of the major safety concerns for maritime traffic management. To analyze the collision data and understand the cause of the collision can contribute the improvement of the maritime traffic safety and management. However, the real collisions is not always available to analyze. Based on a massive AIS trajectory data collected, we focus on mining the ships' movement behaviors those may bring a possible collision if they do not take any avoidance, called Maritime Traffic Conflict. Even though the maritime traffic conflict is a non-accident incident, the movement behaviors of maritime traffic conflict may have the similar behaviors of navigational collision for analysis. Thus, we propose ConflictFinder to provide a framework for maritime traffic conflict mining. Different from existing methods those focus on detecting the conflicts between two ships in a restricted water way, we discover the conflicts occurred by multi-ships in open sea. For analysis of maritime traffic conflicts, a prototype of ConflictFinder is implemented which helps with gaining a better understanding of traffic conflicts discovered and can be applied to the improvement of maritime traffic safety evaluation and management.

In mobile communication services, users can communicate with each other over different telecommunication carriers. For telecom operators, how to acquire and retain users is a significant and practical task. Note that telecom operators only have their own customer profiles. For the users from other telecom operators, their information is sparse. Thus, given a set of communication logs, the main theme of our work is to identify the potential users who will possibly join the target services in the near future. Since only a limited amount of information is available, one challenging issue is how to extract features from the communication logs. In this article, we propose a Communication-Based Feature Generation (CBFG) framework that extracts features and builds models to infer the potential users. Explicitly, we construct a heterogeneous information network from the communication logs of users. Then, we extract the explicit features, which refer to those calling features of users, from the potential users’ interaction behaviors in the heterogeneous information network. Moreover, from the calling behaviors of users, one could extract the possible community structures of users. Based on the community structures, we further extract the implicit features of users. In light of both explicit and implicit features, we propose an information-gain-based method to select the effective features. According to the features selected, we utilize three popular classifiers (i.e., AdaBoost, Random Forest, and SVM) to build models to target the potential users. In addition, we have designed a sampling approach to extract training data for classifiers. To evaluate our methods, we have conducted experiments on a real dataset. The results of our experiments show that the features extracted by our proposed method can be effective for targeting the potential users.

With the popularity of social media (e.g., Facebook and Flicker), users can easily share their check-in records and photos during their trips. In view of the huge number of user historical mobility records in social media, we aim to discover travel experiences to facilitate trip planning. When planning a trip, users always have specific preferences regarding their trips. Instead of restricting users to limited query options such as locations, activities, or time periods, we consider arbitrary text descriptions as keywords about personalized requirements. Moreover, a diverse and representative set of recommended travel routes is needed. Prior works have elaborated on mining and ranking existing routes from check-in data. To meet the need for automatic trip organization, we claim that more features of Places of Interest (POIs) should be extracted. Therefore, in this paper, we propose an efficient Keyword-aware Representative Travel Route framework that uses knowledge extraction from users' historical mobility records and social interactions. Explicitly, we have designed a keyword extraction module to classify the POI-related tags, for effective matching with query keywords. We have further designed a route reconstruction algorithm to construct route candidates that fulfill the requirements. To provide befitting query results, we explore Representative Skyline concepts, that is, the Skyline routes which best describe the trade-offs among different POI features. To evaluate the effectiveness and efficiency of the proposed algorithms, we have conducted extensive experiments on real location-based social network datasets, and the experiment results show that our methods do indeed demonstrate good performance compared to state-of-the-art works.

Sequential pattern mining is an important subfield in data mining. Recently, discovering patterns from interval events has attracted considerable efforts due to its widespread applications. However, due to the complex relation between two intervals, mining interval-based sequences efficiently is a challenging issue. In this paper, we develop a novel algorithm, P-TPMiner, to efficiently discover two types of interval-based sequential patterns. Some pruning techniques are proposed to further reduce the search space of the mining process. Experimental studies show that proposed algorithm is efficient and scalable. Furthermore, we apply proposed method to real datasets to demonstrate the practicability of discussed patterns.

With the popularity of location-based social networks (LBSNs), users would like to share their check-ins with their friends for more social interactions. These check-in records reflect not only when and where they are, but also what they are doing. If we can capture the relations of the location, time, and activity factors in LBSNs, the location-based social platforms can provide more personalized location-based services to users. In this paper, we aim to infer individual activity and mobility based on their check-in records in LBSNs. For these two inference problems, we analyze check-in records, and utilize Bayesian network to represent the relations among location, time, and activity factors of check-in records. Based on the proposed network model, the two inference problems can be simplified to two modules, the activity-time and the location-activity relation. For the activity-time relation, we propose Order-1 Activity Transition Model to capture the activity-time relations of check-in records. Moreover, for the location-activity relation, we exploit the Gaussian mixture model to capture individual mobility features in different activities. To evaluate the proposed network model for the two inference problems, we conduct extensive experiments on two real datasets, and the experimental results show that our proposed Bayesian-based approach has higher performance than the state-of-the-art approaches for activity and mobility inference in LBSNs.

The increasing popularity and growth of mobile devices and location-based services enable us to utilize large-scale geo-tagged data to support novel location-based applications. This paper introduces a novel problem called the best region search (BRS) problem and provides efficient solutions to it. Given a set O of spatial objects, a submodular monotone aggregate score function, and the size a x b of a query rectangle, the BRS problem aims to find a x b rectangular region such that the aggregate score of the spatial objects inside the region is maximized. This problem is fundamental to support several real-world applications such as most influential region search (eg. the best location for a signage to attract most audience) and most diversified region search (eg. region with most diverse facilities). We propose an efficient algorithm called SliceBRS to find the exact answer to the BRS problem. Furthermore, we propose an approximate solution called CoverBRS and prove that the answer found by it is bounded by a constant. Our experimental study with real-world datasets and applications demonstrates the effectiveness and superiority of our proposed algorithms.

With the explosion of smartphones and social network services, location-based social networks (LBSNs) are increasingly seen as tools for businesses (e.g., restaurants and hotels) to promote their products and services. In this article, we investigate the key techniques that can help businesses promote their locations by advertising wisely through the underlying LBSNs. In order to maximize the benefit of location promotion, we formalize it as an influence maximization problem in an LBSN, i.e., given a target location and an LBSN, a set of k users (called seeds) should be advertised initially such that they can successfully propagate and attract many other users to visit the target location. Existing studies have proposed different ways to calculate the information propagation probability, that is, how likely it is that a user may influence another, in the setting of a static social network. However, it is more challenging to derive the propagation probability in an LBSN since it is heavily affected by the target location and the user mobility, both of which are dynamic and query dependent. This article proposes two user mobility models, namely the Gaussian-based and distance-based mobility models, to capture the check-in behavior of individual LBSN users, based on which location-aware propagation probabilities can be derived. Extensive experiments based on two real LBSN datasets have demonstrated the superior effectiveness of our proposals compared with existing static models of propagation probabilities to truly reflect the information propagation in LBSNs.

General health examination is an integral part of healthcare in many countries. Identifying the participants at risk is important for early warning and preventive intervention. The fundamental challenge of learning a classification model for risk prediction lies in the unlabeled data that constitutes the majority of the collected dataset. Particularly, the unlabeled data describes the participants in health examinations whose health conditions can vary greatly from healthy to very-ill. There is no ground truth for differentiating their states of health. In this paper, we propose a graph-based, semi-supervised learning algorithm called SHG-Health (Semi-supervised Heterogeneous Graph on Health) for risk predictions to classify a progressively developing situation with the majority of the data unlabeled. An efficient iterative algorithm is designed and the proof of convergence is given. Extensive experiments based on both real health examination datasets and synthetic datasets are performed to show the effectiveness and efficiency of our method.

In the recent years, several research works have been conducted on collecting context data from various sensors for activity inference. We observe that users perform several actions in their mobile phones: taking photos, performing check-ins, and accessing Wi-Fi networks. These actions generate spatial-temporal data that could be utilized to capture user activities. Spatial-temporal data could indicate that a user stays in a certain location at a particular time for a certain activity. In addition, by referring to social media data, one could also infer user activities. Three types of features are extracted for activity inference: 1) geographical feature, indicating where a user performs activities; 2) temporal feature, indicating when a user performs activities; and 3) semantic feature, showing the semantic concept of a place from location-based social networks. Here, we propose Spatial-Temporal Activity Inference Model (STAIM) to infer user activities from data with those three features. In addition, to determine the weight for each feature, we further propose three methods based on frequency, entropy, and entropy-frequency. Experimental results show that STAIM is able to effectively infer user activities, achieving 75% accuracy on average. Moreover, STAIM could infer user activities even when there is no training data (with some performance loss). Moreover, sensitive analysis of parameters is also conducted to select the most optimal parameter.

With the popularity of social media (e.g., Facebook and Flicker), users could easily share their check-in records and photos during their trips. In view of the huge amount of check-in data and photos in social media, we intend to discover travel experiences to facilitate trip planning. Prior works have been elaborated on mining and ranking existing travel routes from check-in data. We observe that when planning a trip, users may have some keywords about preference on his/her trips. Moreover, a diverse set of travel routes is needed. To provide a diverse set of travel routes, we claim that more features of Places of Interests (POIs) should be extracted. Therefore, in this paper, we propose a Keyword-aware Skyline Travel Route (KSTR) framework that use knowledge extraction from historical mobility records and the user's social interactions. Explicitly, we model the "Where, When, Who" issues by featurizing the geographical mobility pattern, temporal influence and social influence. Then we propose a keyword extraction module to classify the POI-related tags automatically into different types, for effective matching with query keywords. We further design a route reconstruction algorithm to construct route candidates that fulfill the query inputs. To provide diverse query results, we explore Skyline concepts to rank routes. To evaluate the effectiveness and efficiency of the proposed algorithms, we have conducted extensive experiments on real location-based social network datasets, and the experimental results show that KSTR does indeed demonstrate good performance compared to state-of-the-art works.

With the explosion of smartphones and social network services, location-based social networks (LBSNs) are increasingly seen as tools for businesses (e.g., restaurants, hotels) to promote their products and services. In this paper, we investigate the key techniques that can help businesses promote their locations by advertising wisely through the underlying LBSNs. In order to maximize the benefit of location promotion, we formalize it as an influence maximization problem in an LBSN, i.e., given a target location and an LBSN, which a set of k users (called seeds) should be advertised initially such that they can successfully propagate and attract most other users to visit the target location. Existing studies have proposed different ways to calculate the information propagation probability, that is how likely a user may influence another, in the settings of static social network. However, it is more challenging to derive the propagation probability in an LBSN since it is heavily affected by the target location and the user mobility, both of which are dynamic and query dependent. This paper proposes two user mobility models, namely Gaussian-based and distance-based mobility models, to capture the check-in behavior of individual LBSN user, based on which location-aware propagation probabilities can be derived respectively. Extensive experiments based on two real LBSN datasets have demonstrated the superior effectiveness of our proposals than existing static models of propagation probabilities to truly reflect the information propagation in LBSNs.

Owing to the great advent of sensor technology, the usage data of appliances in a house can be logged and collected easily today. However, it is a challenge for the residents to visualize how these appliances are used. Thus, mining algorithms are much needed to discover appliance usage patterns. Most previous studies on usage pattern discovery are mainly focused on analyzing the patterns of single appliance rather than mining the usage correlation among appliances. In this article, a novel algorithm, namely Correlation Pattern Miner (CoPMiner), is developed to capture the usage patterns and correlations among appliances probabilistically. CoPMiner also employs four pruning techniques and a statistical model to reduce the search space and filter out insignificant patterns, respectively. Furthermore, the proposed algorithm is applied on a real-world dataset to show the practicability of correlation pattern mining.

Sequential pattern mining is an important subfield in data mining. Recently, applications using time interval-based event data have attracted considerable efforts in discovering patterns from events that persist for some duration. Since the relationship between two intervals is intrinsically complex, how to effectively and efficiently mine interval-based sequences is a challenging issue. In this paper, two novel representations, endpoint representation and endtime representation, are proposed to simplify the processing of complex relationships among event intervals. Based on the proposed representations, three types of interval-based patterns: temporal pattern, occurrence-probabilistic temporal pattern, and duration-probabilistic temporal pattern, are defined. In addition, we develop two novel algorithms, Temporal Pattern Miner (TPMiner) and Probabilistic Temporal Pattern Miner (P-TPMiner), to discover three types of interval-based sequential patterns. We also propose three pruning techniques to further reduce the search space of the mining process. Experimental studies show that both algorithms are able to find three types of patterns efficiently. Furthermore, we apply proposed algorithms to real datasets to demonstrate the effectiveness and validate the practicability of proposed patterns.

In mobile social networks, users can communicate with each other over different telecom operators. Thus, for telecom operators, how to attract new customers is a significant issue. The work of churn prediction is to determine whether a customer would leave soon. Differing from churn prediction, our work is to find those users who are likely to join target services from the competitors in the near future, where these users are called potential users. To infer potential users, we propose a framework including feature extraction, feature selection, and classifier learning to solve the problem. First, we construct a heterogeneous information network from the call detail records of users. Then, we extract the explicit features from potential users' interaction behavior in the heterogeneous information network. Moreover, because users are influenced by their community, we extract community-based implicit features of potential users. After feature extraction, we explore the Information Gain to select the effective features. We use the effective explicit and implicit features to learn potential user classifiers, and use the classifiers to determine the potential users. Finally, we conduct experiments on real datasets. The results of our experiments show that the features extracted by our proposed method can improve the accuracy of inferring potential users.

Prior works have been elaborated on activity inferences from context information sensed on smart phones. Most of sensing computations are performed on CPU of smart phones. Thus, Sensor Hub is designed to avoid CPU involvement. However, Sensor Hub has several limitations, such as limited memory space and computation power. Since activity inference is a classification problem, prior works have already proposed some classifiers on smart phones. However, if one would like to build a classifier model in Sensor Hub, one challenging issue is to reduce the model size. One approach to reduce the model size of classifiers is feature selection. Feature selection reduces the model size by removing features in the feature set. Nonetheless, because sensor space is limited, removing features could reduce accuracy of classifier algorithm significantly. Therefore we explore feature-value selection concept, which considers the value rather than the feature to reduce the model size while preserving accuracy of classifiers. In this paper, we propose three feature-value selection methods, which consider confusion and redundancy among the feature-value. Due to the nature of feature-values, discretization of sensor data is important. We design a discretization method, LGD (Length Gini Discretization) and compare it with another method, MDLP (Minimum Description Length Discretization), to discretize sensor data using confusion metric to choose the cut point. Extensive experiments are conducted to evaluate our proposed feature-value selection methods. Feature-value removal allows us to reduce up to 80% of the model size and maintain average accuracy performance to 86%.

In this paper, we target the location promotion problem in location-based social networks (LBSNs). The location promotion problem is given a location, we select a set of users as seeds to influence as many users as possible who are likely to visit a selected location. Specifically, we model the location promotion problem as an influence maximization problem on a graph and explore the independent cascading diffusion model on the graph. To determine the propagation probability of the edges of our proposed graph, the relation between users and the selected location should be detected. A property of LBSN is that the major reason of users visiting a location is based on their mobility. Therefore, we propose a mobility model DMM (Distance-based Mobility Model) to model each user's mobility. DMM exploits random walk with restart and the power law property of users' movements. Based on DMM and the selected location, the propagation probability of edges can be derived. In the evaluation, we show the performance of our proposed algorithms on two real datasets.

Recently, with the advent of location-based social networking services (LBSNs), travel planning and location-aware information recommendation based on LBSNs have attracted much research attention. In this paper, we study the impact of social relations hidden in LBSNs, i.e., The social influence of friends. We propose a new social influence-based user recommender framework (SIR) to discover the potential value from reliable users (i.e., Close friends and travel experts). Explicitly, our SIR framework is able to infer influential users from an LBSN. We claim to capture the interactions among virtual communities, physical mobility activities and time effects to infer the social influence between user pairs. Furthermore, we intend to model the propagation of influence using diffusion-based mechanism. Moreover, we have designed a dynamic fusion framework to integrate the features mined into a united follow probability score. Finally, our SIR framework provides personalized top-k user recommendations for individuals. To evaluate the recommendation results, we have conducted extensive experiments on real datasets (i.e., The Go Walla dataset). The experimental results show that the performance of our SIR framework is better than the state-of the-art user recommendation mechanisms in terms of accuracy and reliability.

The advances in location-acquisition and smart phone technologies have led to a myriad of location-based social media. Therefore, analyzing the increasing amount of spatio-temporal data emerges as an important topic. Most studies on geographic data mining focus on exploring static mobility patterns. As the amount of incoming data streams increases, revealing the temporal aspect of user mobility patterns is worth investigating. This paper targets on mining user mobility patterns over time (referred to as mobility evolution) from streams of check-in records. Intuitively, at each time slot, a mobility pattern indicates spatial regions where users stay. Therefore, given a set of time slots, mobility evolution refers a sequence of spatial regions at each time slot. Note that nearby time slots may have similar spatial region distribution. Thus, given check-in datasets, we use the idea of data compression to obtain a sequence of representative segments, where each representative segment captures spatial region distribution at the corresponding time interval. To measure the quality of a segmentation result, we propose a representation cost function based on the Minimum Description Length (MDL) principle. In addition, because deriving the sequence of segments incurs expensive computational cost, we propose a family of greedy algorithms for segmentation to serve diverse requirements: efficient compression, informative compression, and cost-effective compression. Besides, to handle the massive amount of incoming check-in data, we also propose an incremental compression approach to incrementally update the mobility evolution. We conduct experiments on Foursquare datasets to demonstrate both the effectiveness and efficiency of our proposed algorithms.

In this paper, given a spatial range Q and a set of query points specified by users, the goal of this paper is to return the travel routes that fulfill two requirements: 1.) travel routes should contain all those query points specified, and 2.) travel routes should be within the spatial range Q. Furthermore, we claim that each query point may have its proper visiting time. As such, the travel routes should go through these query points at their corresponding proper visiting time. To avoid some redundant information in the travel routes, we utilize the skyline concept to retrieve travel routes with more diversity. Specifically, in our paper, we consider some factors, such as the visiting time information of POIs and the set of query points, in retrieving travel routes. These factors could be mapped into dimensional spaces. Then, each travel route is viewed as a data point in the dimensional space. Thus, skyline data points (referred to as skyline travel routes) are returned as the query result. Skyline travel routes could provide more diversity in the query result of trip route recommendations. To evaluate our proposed methods, we conducted extensive experiments on real datasets. The experimental results show that skyline travel routes indeed provide more diversity in the query result. In addition, we evaluate the efficiency of retrieving skyline travel routes.

In this paper, we propose a framework to infer different people's activity from the view of both the geographical habit and temporal habit of user. Such a personal activity inference framework is a crucial prerequisite for intelligent user experience, and power management of smart phones. By analyzing the real activity log data, we extract 3 kinds of features: 1) The geographical feature captures the user's activity preference of places, 2) The temporal feature records the routine habit of user's activity, 3) The semantic feature obtained from location-based social network can be used as an activity reference of public opinion for each location. Finally, we hybrid the features to build a Semantic-based Activity Inference Model (SAIM). To evaluate our proposed framework SAIM, we compared it with the state-of-art methods over a real dataset. The experimental results show that our framework could accurately inference user's activity and each feature of the three has different inferring ability for different user.

Mining trajectory data has been attracting significant interest in the last years. By analyzing trajectory data, we are able to discover the movement behavior and location-aware knowledge, and then develop many interesting applications such as movement behavior discovery, location prediction, traffic analysis, and so on. However, trajectory data mining is a challenge task because of the trajectory data is available with uncertainty. Furthermore, discovering the valuable knowledge from maritime trajectory is made even more difficult due to the maritime area is a free moving space. Unlike the vehicles' movements are constrained by road networks, there is no such a sea route for ships to follow in maritime area. A ship's movement may not exactly repeat the same trajectory even the ship has the similar movement behavior with others. In this work, Route Miner system provides a framework of ship route mining for maritime traffic analysis. Given a set of ship trajectories in a maritime area, Route Miner explore the movement behavior from those massive trajectories in a free moving space. Then, ship routes are detected based on those behavioral pattern. Finally, the system generates a set of ships routes to provide operators a better understanding from ship trajectory data. We conduct the experiments on real maritime trajectories to show the effectiveness of proposed Route Miner. In the future, Route Miner is going to serve as the photo type for exploring the solutions of the challenges those related to anomaly detection and traffic management in the maritime domain.

Given a social graph, the problem of influence maximization is to determine a set of nodes that maximizes the spread of influences. While some recent research has studied the problem of influence maximization, these works are generally too time consuming for practical use in a large-scale social network. In this article, we develop a new framework, community-based influence maximization (CIM), to tackle the influence maximization problem with an emphasis on the time efficiency issue. Our proposed framework, CIM, comprises three phases: (i) community detection, (ii) candidate generation, and (iii) seed selection. Specifically, phase (i) discovers the community structure of the network; phase (ii) uses the information of communities to narrow down the possible seed candidates; and phase (iii) finalizes the seed nodes from the candidate set. By exploiting the properties of the community structures, we are able to avoid overlapped information and thus efficiently select the number of seeds to maximize information spreads. The experimental results on both synthetic and real datasets show that the proposed CIM algorithm significantly outperforms the state-of-the-art algorithms in terms of efficiency and scalability, with almost no compromise of effectiveness.

In this paper, we tackle the problem of discovering movement-based communities of users, where users in the same community have similar movement behaviors. Note that the identification of movement-based communities is beneficial to location-based services and trajectory recommendation services. Specifically, we propose a framework to mine movement-based communities which consists of three phases: 1) constructing trajectory profiles of users, 2) deriving similarity between trajectory profiles, and 3) discovering movement-based communities. In the first phase, we design a data structure, called the Sequential Probability tree (SP-tree), as a user trajectory profile. SP-trees not only derive sequential patterns, but also indicate transition probabilities of movements. Moreover, we propose two algorithms: BF (standing for breadth-first) and DF (standing for depth-first) to construct SP-tree structures as user profiles. To measure the similarity values among users’ trajectory profiles, we further develop a similarity function that takes SP-tree information into account. In light of the similarity values derived, we formulate an objective function to evaluate the quality of communities. According to the objective function derived, we propose a greedy algorithm Geo-Cluster to effectively derive communities. To evaluate our proposed algorithms, we have conducted comprehensive experiments on two real data sets. The experimental results show that our proposed framework can effectively discover movement-based user communities.

Predicting Apps usage has become an important task due to the proliferation of Apps, and the complex of Apps. However, the previous research works utilized a considerable number of different sensors as training data to infer Apps usage. To save the energy consumption for the task of predicting Apps usages, only the temporal information is considered in this paper. We propose a Temporal-based Apps Predictor (abbreviated as TAP) to dynamically predict the Apps which are most likely to be used. First, we extract three Apps usage features, global usage feature, temporal usage feature, and periodical usage feature from the Apps usage trace. Then, based on those explored features, we dynamically derive an Apps usage probability model to estimate the current usage probability of each App in each feature. Finally, we investigate the usage probability in each feature and select k Apps with highest usage probability from the probability model. In this paper, we propose two selection algorithms, MaxProb and MinEntropy. To evaluate the performance of TAP, we use two real mobile Apps usage traces and assess the accuracy and efficiency. The experimental results show that the proposed TAP with the MinEntropy selection algorithm could have shorter response time of Apps prediction. Moreover, the accuracy reaches to 80% when k is 5, and when k is 7, the accuracy achieves almost 100% in both of the two real datasets.

With the increasing number of mobile Apps developed, they are now closely integrated into daily life. In this paper, we develop a framework to predict mobile Apps that are most likely to be used regarding the current device status of a smartphone. Such an Apps usage prediction framework is a crucial prerequisite for fast App launching, intelligent user experience, and power management of smartphones. By analyzing real App usage log data, we discover two kinds of features: The Explicit Feature (EF) from sensing readings of built-in sensors, and the Implicit Feature (IF) from App usage relations. The IF feature is derived by constructing the proposed App Usage Graph (abbreviated as AUG) that models App usage transitions. In light of AUG, we are able to discover usage relations among Apps. Since users may have different usage behaviors on their smartphones, we further propose one personalized feature selection algorithm. We explore minimum description length (MDL) from the training data and select those features which need less length to describe the training data. The personalized feature selection can successfully reduce the log size and the prediction time. Finally, we adopt the kNN classification model to predict Apps usage. Note that through the features selected by the proposed personalized feature selection algorithm, we only need to keep these features, which in turn reduces the prediction time and avoids the curse of dimensionality when using the kNN classifier. The results based on a real dataset demonstrate the effectiveness of the proposed framework and show the predictive capability for App usage prediction.

With the growth of location-based services and social services, low- sampling-rate trajectories from check-in data or photos with geo- tag information becomes ubiquitous. In general, most detailed mov- ing information in low-sampling-rate trajectories are lost. Prior works have elaborated on distant-time location prediction in high- sampling-rate trajectories. However, existing prediction models are pattern-based and thus not applicable due to the sparsity of data points in low-sampling-rate trajectories. To address the sparsity in low-sampling-rate trajectories, we develop a Reachability-based prediction model on Time-constrained Mobility Graph (RTMG) to predict locations for distant-time queries. Specifically, we de- sign an adaptive temporal exploration approach to extract effective supporting trajectories that are temporally close to the query time. Based on the supporting trajectories, a Time-constrained mobility Graph (TG) is constructed to capture mobility information at the given query time. In light of TG, we further derive the reacha- bility probabilities among locations in TG. Thus, a location with maximum reachability from the current location among all possi- ble locations in supporting trajectories is considered as the predic- tion result. To efficiently process queries, we proposed the index structure Sorted Interval-Tree (SOIT) to organize location records. Extensive experiments with real data demonstrated the effective- ness and efficiency of RTMG. First, RTMG with adaptive tempo- ral exploration significantly outperforms the existing pattern-based prediction model HPM [2] over varying data sparsity in terms of higher accuracy and higher coverage. Also, the proposed index structure SOIT can efficiently speedup RTMG in large-scale trajec- tory dataset. In the future, we could extend RTMG by considering more factors (e.g., staying durations in locations, application us- ages in smart phones) to further improve the prediction accuracy.

With the growth of location-based services and social services, low-sampling-rate trajectories from check-in data or photos with geo-tag information becomes ubiquitous. In general, most detailed moving information in low-sampling-rate trajectories are lost. Prior works have elaborated on distant-time location prediction in high-sampling-rate trajectories. However, existing prediction models are pattern-based and thus not applicable due to the sparsity of data points in low-sampling-rate trajectories. For example, it becomes difficult to derive trajectory patterns, let alone utilizing trajectory patterns for distant-time location prediction. In this paper, given a query time, the current location and time, we aim to predict the location of an object at the query time. To address the sparsity in low-sampling-rate trajectories, we develop a Reachability-based prediction model on Time-constrained Mobility Graph (abbreviated as RTMG) to predict locations for distant-time queries. Specifically, we design an adaptive temporal exploration approach to extract effective supporting trajectories that are temporally close to the query time. These data points are then represented as a Time-constrained user mobility Graph (refers to as TG). In light of TG, we further derive the reachability probabilities among locations in TG. Thus, a location with maximum reachability from the current location among all possible locations in supporting trajectories is considered as the prediction result. To efficiently process queries, we proposed an index structure SOIT to organize location records for on-line query processing. We conduct extensive experiments on real low-sampling-rate datasets and demonstrate the effectiveness and efficiency of RTMG.

This paper presents a novel framework for dynamic circle recommendation for a query user at a given time point from historical communication logs. We identify the fundamental factors that govern interactions and aim to automatically form friend circles for scenarios, such as, who should I share the photo with in the early morning? Whose post should be listed on top of my Facebook Wall feed at night? We develop a temporal probabilistic model that not only captures temporal tendencies between the query user and each friend candidate but also blends frequency and recency into circle formation. Experimental results on Enron dataset and Call Detail Records prove the effectiveness of dynamic circle formation with proposed temporal probabilistic model.

The advance of mobile technologies leads to huge volumes of spatio-temporal data collected in the form of trajectory data streams. In this study, we investigate the problem of discovering object groups that travel together (i.e., traveling companions ) from trajectory data streams. Such technique has broad applications in the areas of scientific study, transportation management, and military surveillance. To discover traveling companions, the monitoring system should cluster the objects of each snapshot and intersect the clustering results to retrieve moving-together objects. Since both clustering and intersection steps involve high computational overhead, the key issue of companion discovery is to improve the efficiency of algorithms. We propose the models of closed companion candidates and smart intersection to accelerate data processing. A data structure termed traveling buddy is designed to facilitate scalable and flexible companion discovery from trajectory streams. The traveling buddies are microgroups of objects that are tightly bound together. By only storing the object relationships rather than their spatial coordinates, the buddies can be dynamically maintained along the trajectory stream with low cost. Based on traveling buddies, the system can discover companions without accessing the object details. In addition, we extend the proposed framework to discover companions on more complicated scenarios with spatial and temporal constraints, such as on the road network and battlefield. The proposed methods are evaluated with extensive experiments on both real and synthetic datasets. Experimental results show that our proposed buddy-based approach is an order of magnitude faster than the baselines and achieves higher accuracy in companion discovery.

With the popularity of positioning devices, Web 2.0 technology, and trip sharing services, many users are willing to log and share their trips on the Web. Thus, trip planning Web sites are able to provide some new services by inferring Regions-Of-Interest (ROIs) and recommending popular travel routes from trip trajectories. We argue that simply providing some travel routes consisting of popular ROIs to users is not sufficient. To tour around a wide geographical area, for example, a city, some users may prefer a trip to visit as many ROIs as possible, while others may like to stop by only a few ROIs for an in-depth visit. We refer to a trip fitting the former user group as an in-breadth trip and a trip suitable for the latter user group as an in-depth trip . Prior studies on trip planning have focused on mining ROIs and travel routes without considering these different preferences. In this article, given a spatial range and a user preference of depth/breadth specified by a user, we develop a Pattern-Aware Trajectory Search (PATS) framework to retrieve the top K trajectories passing through popular ROIs. PATS is novel because the returned travel trajectories, discovered from travel patterns hidden in trip trajectories, may represent the most valuable travel experiences of other travelers fitting the user's trip preference in terms of depth or breadth. The PATS framework comprises two components: travel behavior exploration and trajectory search . The travel behavior exploration component determines a set of ROIs along with their attractive scores by considering not only the popularity of the ROIs but also the travel sequential relationships among the ROIs. To capture the travel sequential relationships among ROIs and to derive their attractive scores, a user movement graph is constructed. For the trajectory search component of PATS, we formulate two trajectory score functions, the depth-trip score function and the breadth-trip score function, by taking into account the number of ROIs in a trajectory and their attractive scores. Accordingly, we propose an algorithm, namely, Bounded Trajectory Search (BTS), to efficiently retrieve the top K trajectories based on the two trajectory scores. The PATS framework is evaluated by experiments and user studies using a real dataset. The experimental results demonstrate the effectiveness and the efficiency of the proposed PATS framework.

A Cyber-Physical System (CPS) integrates physical devices (e.g., sensors, cameras) with cyber (or informational) components to form a situation-integrated analytical system that may respond intelligently to dynamic changes of the real-world situations. CPS claims many promising applications, such as traffic observation, battlefield surveillance and sensor-network based monitoring. One important research topic in CPS is about the atypical event analysis, i.e., retrieving the events from large amount of data and analyzing them with spatial, temporal and other multi-dimensional information. Many traditional approaches are not feasible for such analysis since they use numeric measures and cannot describe the complex atypical events. In this study, we propose a new model of atypical cluster to effectively represent those events and efficiently retrieve them from massive data. The micro-cluster is designed to summarize individual events, and the macro-cluster is used to integrate the information from multiple event. To facilitate scalable, flexible and online analysis, the concept of significant cluster is defined and a guided clustering algorithm is proposed to retrieve significant clusters in an efficient manner. We conduct experiments on real datasets with the size of more than 50 GB, the results show that the proposed method can provide more accurate information with only 15% to 20% time cost of the baselines.

The advance of object tracking technologies leads to huge volumes of spatio-temporal data collected in the form of trajectory data stream. In this study, we investigate the problem of discovering object groups that travel together (i.e., traveling companions) from trajectory stream. Such technique has broad applications in the areas of scientific study, transportation management and military surveillance. To discover traveling companions, the monitoring system should cluster the objects of each snapshot and intersect the clustering results to retrieve moving-together objects. Since both clustering and intersection steps involve high computational overhead, the key issue of companion discovery is to improve the algorithm's efficiency. We propose the models of closed companion candidates and smart intersection to accelerate data processing. A new data structure termed traveling buddy is designed to facilitate scalable and flexible companion discovery on trajectory stream. The traveling buddies are micro-groups of objects that are tightly bound together. By only storing the object relationships rather than their spatial coordinates, the buddies can be dynamically maintained along trajectory stream with low cost. Based on traveling buddies, the system can discover companions without accessing the object details. The proposed methods are evaluated with extensive experiments on both real and synthetic datasets. The buddy-based method is an order of magnitude faster than existing methods. It also outperforms other competitors with higher precision and recall in companion discovery.

The advances in location-acquisition technologies have led to a myriad of spatial trajectories. These trajectories are usually generated at a low or an irregular frequency due to applications' characteristics or energy saving, leaving the routes between two consecutive points of a single trajectory uncertain (called an uncertain trajectory). In this paper, we present a Route Inference framework based on Collective Knowledge (abbreviated as RICK) to construct the popular routes from uncertain trajectories. Explicitly, given a location sequence and a time span, the RICK is able to construct the top-k routes which sequentially pass through the locations within the specified time span, by aggregating such uncertain trajectories in a mutual reinforcement way (i.e., uncertain + uncertain → certain). Our work can benefit trip planning, traffic management, and animal movement studies. The RICK comprises two components: routable graph construction and route inference. First, we explore the spatial and temporal characteristics of uncertain trajectories and construct a routable graph by collaborative learning among the uncertain trajectories. Second, in light of the routable graph, we propose a routing algorithm to construct the top-k routes according to a user-specified query. We have conducted extensive experiments on two real datasets, consisting of Foursquare check-in datasets and taxi trajectories. The results show that RICK is both effective and efficient.

Due to the flexibility and scalability in cloud computing, cloud computing nowadays plays an important role to handle a large-scale data analysis. For data processing operations, several cloud data managements (CDMs), such as HBase and Cassandra, are developed. Such CDMs usually provide key-value storages, where each key is used to access its corresponding value. Both HBase and Cassandra provide some basic operations (e.g., Get, Scan) to retrieve the values via keys specified by users. The exiting CDMs fully inherit the characteristics of cloud computing (i.e., high scalability and availability). With the aforementioned characteristics of cloud computing, CDMs are widely employed for Web data, especially for search engines. However, with the proliferation of smart phones and location-based services, data with spatial information, referring as spatial data, are dramatically increasing. Consequently, how to formulate keys for spatial data in the existing CDMs is a challenge issue. In this paper, we develop several key formulation schemes. In particular, we propose a novel Key formulation scheme based on R+-tree (abbreviated as KR+-index). With our design for keys of spatial data, the existing CDMs are able to efficiently retrieve spatial data. In light of KR+-tree, two spatial queries, k-NN query and range query, are designed. Moreover, we implement the proposed key formulation schemes on HBase and Cassandra, and import real spatial data for spatial queries. The experimental results demonstrate that KR+-tree outperforms other existing key formulations and MD-HBase.

To conserve energy, sensor nodes with similar readings can be grouped such that readings from only the representative nodes within the groups need to be reported. However, efficiently identifying sensor groups and their representative nodes is a very challenging task. In this paper, we propose a centralized algorithm to determine a set of representative nodes with high energy levels and wide data coverage ranges. Here, the data coverage range of a sensor node is considered to be the set of sensor nodes that have reading behaviors very close to the particular sensor node. To further reduce the extra cost incurred in messages for selection of representative nodes, a distributed algorithm is developed. Furthermore, maintenance mechanisms are proposed to dynamically select alternative representative nodes when the original representative nodes run low on energy, or cannot capture spatial correlation within their respective data coverage ranges. Using experimental studies on both synthesis and real data sets, our proposed algorithms are shown to effectively and efficiently provide approximate data collection while prolonging the network lifetime.

In this paper, we first introduce the traffic status estimation service, a location-based continuous range query application provided by the CarWeb system. The service provides users nearby realtime traffic estimation according to their locations and moving behaviors. This paper focuses on how to provide users with the most relevant information efficiently and accurately. We propose the Continuous Query with Dynamic Fan-Shape framework (CQ-DFS) which dynamically changes parameters of fan shape range query to efficiently acquire more useful query results without compromising the performance of the system. We conduct extensive experiments to demonstrate the effectiveness of CQ-DFS framework by using real world vehicle trajectories.

The mining of closed sequential patterns has attracted researchers for its capability of using compact results to preserve the same expressive power as conventional mining. However, existing studies only focus on time point-based data. Few research efforts have elaborated on discovering closed sequential patterns from time interval-based data, where each data persists for a period of time. Mining closed time interval-based patterns, also called closed temporal patterns, is an arduous problem since the pair wise relationships between two interval-based events are intrinsically complex. In this paper, an efficient algorithm, CEMiner is developed to discover closed temporal patterns from interval-based data. Algorithm CEMiner employs some optimization techniques to effectively reduce the search space. The experimental results on both synthetic and real datasets indicate that CEMiner not only significantly outperforms the prior interval-based mining algorithms in terms of execution time but also possesses graceful scalability. The experiment conducted on real dataset shows the practicability of time interval-based closed pattern mining.

Location prediction has attracted a significant amount of research effort. Given an object's recent movements and a future time, the goal of location prediction is to predict the location of this object at the future time specified. Prior works have elaborated on mining association relationships among regions, in which objects frequently appear, to predict locations. Association relationships among regions are represented as association rules. By exploring association relationships among regions, prior works are able to have a good accuracy for location prediction. However, with a large amount of trajectory data produced, a huge amount of association rules is expected. Furthermore, trajectory data has both the spatial and temporal information. To further enhance the accuracy of location prediction, one could utilize not only spatial information but also temporal information to estimate locations of objects. In this paper, we propose a spatial-temporal trajectory model (abbreviated as STT) to capture movement behaviors of objects. STT is represented as a probabilistic suffix tree with both spatial and temporal information of movements. Note that STT is able to discover sequential traversal relationships among regions and, for each region, STT derives the corresponding probabilities about the time when objects appear. With the nature of probabilistic suffix tree, we could use a compact structure to capture movement behavior of objects compared to association rules proposed. In light of STT, we further propose an algorithm to traverse STT for location prediction. By exploring both the spatial and temporal information of STT, the accuracy of location prediction is improved. To evaluate our proposed STT and prediction algorithm, we conduct experiments on the synthetic dataset generated from real datasets. Experimental results shows that our proposed STT is able to capture both spatial and temporal patterns of movement behaviors and, by exploring STT, our proposed prediction algorithm outperforms existing works.

Most studies on sequential pattern mining are mainly focused on time point-based event data. Few research efforts have elaborated on mining patterns from time interval-based event data. However, in many real applications, event usually persists for an interval of time. Since the relationships among event time intervals are intrinsically complex, mining time interval-based patterns in large database is really a challenging problem. In this paper, a novel approach, named as incision strategy and a new representation, called coincidence representation are proposed to simplify the processing of complex relations among event intervals. Then, an efficient algorithm, CTMiner (Coincidence Temporal Miner) is developed to discover frequent time-interval based patterns. The algorithm also employs two pruning techniques to reduce the search space effectively. Furthermore, experimental results show that CTMiner is not only efficient and scalable but also outperforms state-of-the-art algorithms.

A Cyber-Physical System (CPS) integrates physical devices (e.g., sensors, cameras) with cyber (or informational)components to form a situation-integrated analytical system that responds intelligently to dynamic changes of the real-world scenarios. One key issue in CPS research is trustworthiness analysis of the observed data: Due to technology limitations and environmental influences, the CPS data are inherently noisy that may trigger many false alarms. It is highly desirable to sift meaningful information from a large volume of noisy data. In this paper, we propose a method called Tru-Alarm which finds out trustworthy alarms and increases the feasibility of CPS. Tru-Alarm estimates the locations of objects causing alarms, constructs an object-alarm graph and carries out trustworthiness inferences based on linked information in the graph. Extensive experiments show that Tru-Alarm filters out noises and false information efficiently and guarantees not missing any meaningful alarms.

In this paper, we develop a framework of trajectory search called pattern-aware trajectory search (abbreviated as PATS). Given a set of trajectories, potential regions are extracted first and potential regions are viewed as popular regions interested by users. Furthermore, potential regions are organized as a region transition graph where each vertex is a potential region and edges capture sequential travels of potential regions from a set of trajectories given. By exploring the concept of random walk, the attractiveness of a potential region is derived. In light of attractiveness of potential regions the attractiveness of a trajectory is formulated and PATS will return top-K trajectories according to their attractiveness. We evaluated our framework by a real GPS dataset. Experimental results show that PATS is able to retrieve trajectories interested by users.

Wireless sensor networks have been proposed for facilitating various monitoring applications (e.g., environmental monitoring and military surveillance) over a wide geographical region. In these applications, spatial queries that collect data from wireless sensor networks play an important role. One such query is the K-Nearest Neighbor (KNN) query that facilitates collection of sensor data samples based on a given query location and the number of samples specified (i.e., K). Recently, itinerary-based KNN query processing techniques, which propagate queries and collect data along a predetermined itinerary, have been developed. Prior studies demonstrate that itinerary-based KNN query processing algorithms are able to achieve better energy efficiency than other existing algorithms developed upon tree-based network infrastructures. However, how to derive itineraries for KNN query based on different performance requirements remains a challenging problem. In this paper, we propose a Parallel Concentric-circle Itinerary-based KNN (PCIKNN) query processing technique that derives different itineraries by optimizing either query latency or energy consumption. The performance of PCIKNN is analyzed mathematically and evaluated through extensive experiments. Experimental results show that PCIKNN outperforms the state-of-the-art techniques.

Prior works have shown that probabilistic suffix trees (PST) could predict accurately the moving behaviors of objects for prediction-based object tracking sensor networks. However, maintaining PSTs for objects incurs a considerable amount of storage spaces for resource-constrained sensor nodes. In this paper, we derive a distance function between two PSTs and propose an algorithm to determine the similarity between them. By the distance between PSTs, we propose a clustering algorithm to partition objects with similar moving behaviors into groups. Furthermore, for each group, one PST is selected to predict movements of objects within one group. Experimental results show that our proposed approaches not only effectively reduce the storage cost but also provide good prediction accuracy.

Traffic status plays an important role in navigation systems. To estimate traffic status, expensive sensors are deployed, which is not cost efficient. In view of the growth of GPS navigation services, in this paper, we propose two algorithms to estimate traffic status of road segments in our CarWeb platform. To evaluate our proposed algorithms, we implement the proposed algorithms in our CarWeb platform that is used to collect GPS data points from cars. Extensive experiments are conducted on real datasets and experimental results indicate that our algorithms can provide desirable predictions of traffic status.

In this paper, we exploit the concept of sharing GPS data for estimate traffic information. Explicitly, we implemented CarWeb, a platform to collect real-time GPS data from cars. Once receiving a sufficient amount of GPS data from cars, we are able to estimate traffic information (i.e., speed information of roads). As such, users are able to obtain the update-to-date traffic information in CarWeb platform. In addition, we propose two algorithms to estimate traffic information with GPS data given. A prototype of CarWeb is implemented and functionalities of each component in CarWeb are demonstrated in this paper.

Recently, the growth of social bookmark sites (e.g., del.icio.us) brings a new way to organize and share Web pages. Specially, the social bookmarking sites contain many bookmarks of users, and users, who bookmark Web pages, would frequently browse these pages in the future. Therefore, we argue that social bookmarking sites provide the readers' perspective and are able to take the perspective into consideration in ranking Web pages. In this paper, we propose two ranking algorithms, ExpertVoteRank and RecommendationPageRank, to reveal the diverse information of Web pages in the social bookmarking sites. The concept of both algorithms are based on the views of readers: ExpertVoteRank takes advantage of experts of readers, while RecommendationPageRank applies recommendations from crowds to Web pages. Note that we collected about 90 millions data. Experiments show that both algorithms have effectiveness to rank Web pages according to the viewpoint of users.

Spatial queries for extracting data from wireless sensor networks are important for many applications, such as environmental monitoring and military surveillance. One such query is K Nearest Neighbor (KNN) query that facilitates sampling of monitored sensor data in correspondence with a given query location. Recently, itinerary-based KNN query processing techniques, that propagate queries and collect data along a pre-determined itinerary, have been developed concurrently [12] [14]. These research works demonstrate that itinerary-based KNN query processing algorithms are able to achieve better energy efficiency than other existing algorithms. However, how to derive itineraries based on different performance requirements remains a challenging problem. In this paper, we propose a new itinerary-based KNN query processing technique, called PCIKNN, that derives different itineraries aiming at optimizing two performance criteria, response latency and energy consumption. The performance of PCIKNN is analyzed mathematically and evaluated through extensive experiments. Experimental results show that PCIKNN has better performance and scalability than the state-of-the-art.

Dummy-based anonymization techniques for protecting location privacy of mobile users have been proposed in the literature. By generating dummies that move in humanlike trajectories, shows that location privacy of mobile users can be preserved. However, by monitoring long-term movement patterns of users, the trajectories of mobile users can still be exposed. We argue that, once the trajectory of a user is identified, locations of the user is exposed. Thus, it's critical to protect the moving trajectories of mobile users in order to preserve user location privacy. We propose two schemes that generate consistent movement patterns in a long run. Guided by three parameters in user specified privacy profile, namely, short- term disclosure, long-term disclosure and distance deviation, the proposed schemes derive movement trajectories for dummies. A preliminary performance study shows that our approach is more effective than existing work in protecting moving trajectories of mobile users and their location privacy.

In this paper, we propose a heterogeneous tracking model, referred to as HTM, to efficiently mine object moving patterns and track objects. Specifically, we use a variable memory Markov model to exploit the dependencies among object movements. Furthermore, due to the hierarchical nature of HTM, multi-resolution object moving patterns are provided. The proposed HTM is able to accurately predict the movements of objects and thus reduces the energy consumption for object tracking. Simulation results show that HTM not only is able to effectively mine object moving patterns but also save energy in tracking objects.

We propose in this paper an energy-conserved on-demand data broadcasting system by employing the data indexing technique. We also propose algorithm AIDOA to adjust the degree of buckets according to system workload. Ex-perimental results show that algorithm AIDOA is able to greatly reduce power consumption at the cost of slight in-crement in average access time and adjust the index and data organization dynamically to adapt to change of system workload.

With the cutting edge technology advance in wireless and mobile computers, the query processing in a mobile environment involves join processing among different sites, which include static servers and mobile computers. Because of the need for energy saving and also the presence of asymmetric features in a mobile computing environment, the conventional query processing for a distributed database cannot be directly applied to a mobile computing system. In this paper, we first explore three asymmetric features of a mobile environment. Then, in light of these features, we devise query processing methods for both join and query processing. Intuitively, employing semijoin operations in a mobile computing environment is able to further reduce both the amount of data transmission and energy consumption. A semijoin which is initiated by a mobile computer (respectively, the server) and is beneficial to reduce the cost of a join operation is termed a mobile-initiated or MI (respectively, server-initiated or SI) profitable semijoin. According to those asymmetric features of a mobile computing system, we examine three different join methods and devise some specific criteria to identify MI/SI profitable semijoins. For query processing, which refers to the processing of multijoin queries, we develop three query processing schemes. In particular, we formulate the query processing in a mobile computing system as a two-phase query processing procedure that can determine a join sequence and interleave that join sequence with SI profitable semijoins to reduce both the amount of data transmission and energy consumption. Performance of these join and query methods is comparatively analyzed and sensitivity analysis on several parameters: is conducted. Furthermore, we develop a systematic procedure to derive the characteristic functions of MI and SI profitable semijoins. It is noted that, given some system parameters, those characteristic functions are very important in determining which join method is the most appropriate one to employ in that configuration. It is shown by our simulation results that, by exploiting the three asymmetric features, these characteristic functions are very powerful in reducing both the amounts of energy consumption and data transmission incurred and can lead to the design of an efficient and effective query processing procedure for a mobile computing environment.

The growth in wireless communication technologies attracts a considerable amount of attention in mobile ad-hoc networks. Since mobile hosts in an ad-hoc network usually move freely, the topology of the network changes dynamically and disconnection occurs frequently. These characteristics make a mobile ad-hoc network be likely to be separated into several disconnected partitions, and the data accessibility is hence reduced. Several schemes are proposed to alleviate the reduction of data accessibility by replicating data items. However, little research effort was elaborated upon exploiting the group mobility where the group mobility refers to the phenomenon that several mobile nodes tend to move together. In this paper, we address the problem of replica allocation in a mobile ad-hoc network by exploring group mobility. We first analyze the group mobility model and derive several theoretical results. In light of these results, we propose a replica allocation scheme to improve the data accessibility. Several experiments are conducted to evaluate the performance of the proposed scheme. The experimental results show that the proposed scheme is able to not only obtain higher data accessibility but also produce lower network traffic than prior schemes.

In several mobile applications, the data items broadcast are dependent upon one another. However, most prior studies on broadcasting dependent data mainly consider single broadcast channel environments. In view of this, we explore the problem of broadcasting dependent data in multiple broadcast channels. By analyzing the model of dependent data broadcasting, we derive several theoretical properties for the average access time in a multiple channel environment. In light of the theoretical results, we develop a genetic algorithm to generate broadcast programs.

In this paper, we present a new data mining algorithm which involves incremental mining for user moving patterns in a mobile computing environment and exploit the mining results to develop data allocation schemes so as to improve the overall performance of a mobile system. First, we propose an algorithm to capture the frequent user moving patterns from a set of log data in a mobile environment. The algorithm proposed is enhanced with the incremental mining capability and is able to discover new moving patterns efficiently without compromising the quality of results obtained. Then, in light of mining results of user moving patterns and the properties of data objects, we develop data allocation schemes that can utilize the knowledge of user moving patterns for proper allocation of both personal and shared data. By employing the data allocation schemes, the occurrences of costly remote accesses can be minimized and the performance of a mobile computing system is thus improved. For personal data allocation, two schemes are devised: one utilizes the set level of moving patterns and the other utilizes their path level. Schemes for shared data are also developed. Performance of these schemes is comparatively analyzed.

In this paper we devise data allocation algorithms that can utilize the knowledge of user moving patterns for proper allocation of shared data in a mobile computing system. By employing the data allocation algorithms devised, the occurrences of costly remote accesses can be minimized and the performance of a mobile computing system is thus improved The data allocation algorithms for shared data, which are able to achieve local optimization and global optimization, are developed. Local optimization refers to the optimization that the likelihood of local data access by an individual mobile user is maximized whereas global optimization refers to the optimization that the likelihood of local data access by all mobile users is maximized By exploring the corresponding features, we devise algorithm SD-local and algorithm SD-global to achieve local optimization and global optimization, respectively. The simulation results show that the knowledge obtained from the user moving patterns is very important in devising effective data allocation algorithms which can lead to prominent performance improvement in a mobile computing system.

We explore in this paper the problem of dynamic data and channel allocations with the number of communication channels and the number of data items given. It is noted that the combined use of broadcast and on-demand channels can utilize the bandwidth effectively for data dissemination in a mobile computing environment. We first derive the an-alytical models of the expected delays when the data are requested through the broadcast and on-demand channels. Then, we transform this problem into to a guided search problem. In light of the theoretical properties derived, we devise an algorithm based on binary interpolation search, referred to as algorithm BIS, to obtain solutions of high quality efficiently. In essence, algorithm BIS is guided to explore the solution space with higher likelihood to be the optimal first, thereby leading to an efficient and effective search. It is shown by our simulation results that the solution obtained by algorithm BIS is of very high quality and is in fact very close to the optimal one. Sensitivity analysis on several parameters, including the number of data items and the number of communication channels, is conducted.

We explore in this paper the problem of generating hierarchical broadcast programs with the data access frequencies and the number of broadcast disks in a broadcast disk array given.Specifically, we first transform the problem of generating hierarchical broadcast programs into the one of constructing a channel allocation tree with variant-fanout.By exploiting the feature of tree generation with variantfanout, we develop a heuristic algorithm VF K to minimize the expected delay of data items in the broadcast program.Performance of these algorithms is analyzed.It is shown by our simulation results that by exploiting the feature of variant-fanout in constructing the channel allocation tree, the solution obtained by algorithm VF K is of very high quality and is in fact very close to the optimal one.