Chung-ju Huang, Huiqiang Zhao, Yuanpeng He, Lijian Li, Wenpin Jiao, Zhi Jin, Peixuan Chen, Leye Wang. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Online 3D Bin Packing (3D-BP) with robotic arms is crucial for reducing transportation and labor costs in modern logistics. While Deep Reinforcement Learning (DRL) has shown strong performance, it often fails to adapt to real-world short-term distribution shifts, which arise as different batches of goods arrive sequentially, causing performance drops. We argue that the short-term lookahead information available in modern logistics systems is key to mitigating this issue, especially during distribution shifts. We formulate online 3D-BP with lookahead parcels as a Model Predictive Control (MPC) problem and adapt the Monte Carlo Tree Search (MCTS) framework to solve it. Our framework employs a dynamic exploration prior that automatically balances a learned RL policy and a robust random policy based on the lookahead characteristics. Additionally, we design an auxiliary reward to penalize long-term spatial waste from individual placements. Extensive experiments on real-world datasets show that our method consistently outperforms state-of-the-art baselines, achieving over 10% gains under distributional shifts, 4% average improvement in online deployment, and up to more than 8% in the best case--demonstrating the effectiveness of our framework.

Scientific research heavily depends on suitable datasets for method validation, but existing academic platforms with dataset management like PapersWithCode suffer from inefficiencies in their manual workflow.To overcome this bottleneck, we present a system, called ChatPD, that utilizes Large Language Models (LLMs) to automate dataset information extraction from academic papers and construct a structured paper-dataset network.Our system consists of three key modules: paper collection, dataset information extraction, and dataset entity resolution to construct paper-dataset networks.Specifically, we propose a Graph Completion and Inference strategy to map dataset descriptions to their corresponding entities.Through extensive experiments, we demonstrate that ChatPD not only outperforms the existing platform PapersWithCode in dataset usage extraction but also achieves about 90% precision and recall in entity resolution tasks.Moreover, we have deployed ChatPD to continuously extract which datasets are used in papers, and provide a dataset discovery service, such as task-specific dataset queries and similar dataset recommendations.We open source ChatPD and the current paper-dataset network on GitHub.

Ubiquitous sensors and mobile devices have spurred the growth of Web-of-Things (WoT) services in smart cities, making accurate spatio-temporal traffic predictions increasingly crucial. Leveraging advances in deep learning, recent Spatio-Temporal Graph Neural Networks (STGNNs) have achieved remarkable results. However, these methods address scenario-specific spatio-temporal heterogeneity by designing model architectures, often overlooking the importance of selecting optimal spatio-temporal knowledge (i.e., model inputs). In this paper, we propose an automated framework for spatio-temporal knowledge optimization to address this challenge. Our framework seamlessly integrates with downstream models, enhancing their performance across various prediction tasks. Specifically, we design a knowledge search space composed of parameters that represent scenario-specific spatio-temporal correlations within data. Additionally, we employ a bandit-based multi-fidelity algorithm for knowledge optimization to solve the constraint of limited resource. Furthermore, we adopt a meta-learner to extract transferable meta-knowledge about optimal knowledge, facilitating efficient exploration of the search space. Extensive experiments on five widely used real-world datasets demonstrate the effectiveness of our proposed framework. To the best of our knowledge, we are the first to automatically optimize spatio-temporal knowledge for spatio-temporal traffic prediction.

Sparse mobile crowdsensing is a cost-effective sensing paradigm that infers global data by sensing data from partial areas in a city. With the rapid development of diverse autonomous mobile agents such as unmanned aerial vehicles (UAVs) and ground vehicles, they have been widely applied in sparse mobile crowdsensing. However, existing works often predefine the role structures and behavioral preferences of these agents in tasks, which significantly limits their flexibility and adaptability, and making it difficult to fully exploit the collaborative potential of crowdsensing agents to efficiently achieve high-quality data sensing. In this paper, we propose an adaptive role learning framework for sparse mobile crowdsensing (ARL-SMCS), which focuses on role recognition for heterogeneous agents and role refinement among homogeneous agents. This framework, based on a multi-agent reinforcement learning model, introduces a variational autoencoder to learn the latent role representations of agents and uses maximum mean discrepancy to distinguish the functionalities of different types of agents. Additionally, ARL-SMCS incorporates an evolutionary algorithm to further refine task preferences among homogeneous agents. This framework overcomes the limitations of static role assignment in adapting to dynamic environments and task conflicts during task execution, significantly improving sensing quality and resource utilization efficiency. Extensive experiments on two real-world datasets demonstrate that ARL-SMCS consistently outperforms other baseline methods under various conditions, including different numbers, endurance, and decision interval lengths.

Publishing graph data is widely desired to enable a variety of structural analyses and downstream tasks. However, it also potentially poses severe privacy leakage, as attackers may leverage the released graph data to launch attacks and precisely infer private information such as the existence of hidden sensitive links in the graph. Prior studies on privacy-preserving graph data publishing relied on heuristic graph modification strategies and it is difficult to determine the graph with the optimal privacy–utility trade-off for publishing. In contrast, we propose the first privacy-preserving graph structure learning framework against sensitive link inference attacks, named PPGSL, which can automatically learn a graph with the optimal privacy–utility trade-off. The PPGSL operates by first simulating a powerful surrogate attacker conducting sensitive link attacks on a given graph. It then trains a parameterized graph to defend against the simulated adversarial attacks while maintaining the favorable utility of the original graph. To learn the parameters of both parts of the PPGSL, we introduce a secure iterative training protocol. It can enhance privacy preservation and ensure stable convergence during the training process, as supported by the theoretical proof. Additionally, we incorporate multiple acceleration techniques to improve the efficiency of the PPGSL in handling large-scale graphs. The experimental results confirm that the PPGSL achieves state-of-the-art privacy–utility trade-off performance and effectively thwarts various sensitive link inference attacks.

Contextual features are important data sources for building citywide crowd mobility prediction models. However, the difficulty of applying context lies in the unknown generalizability of contextual features (e.g., weather, holiday, and points of interests) and context modeling techniques across different scenarios. In this paper, we present a unified analytic framework and a large-scale benchmark for evaluating context generalizability. The benchmark includes crowd mobility data, contextual data, and advanced prediction models. We conduct comprehensive experiments in several crowd mobility prediction tasks such as bike flow, metro passenger flow, and electric vehicle charging demand. Our results reveal several important observations: (1) Using more contextual features may not always result in better prediction with existing context modeling techniques; in particular, the combination of holiday and temporal position can provide more generalizable beneficial information than other contextual feature combinations. (2) In context modeling techniques, using a gated unit to incorporate raw contextual features into the deep prediction model has good generalizability. Besides, we offer several suggestions about incorporating contextual factors for building crowd mobility prediction applications. From our findings, we call for future research efforts devoted to developing new context modeling solutions.

Cross-hospital collaboration has the potential to mitigate disparities in medical resources across different regions. However, strict privacy regulations prohibit the direct sharing of sensitive patient information between hospitals. Vertical Federated Learning (VFL) provides a novel privacy-preserving machine learning paradigm designed to maximizes data utility across multiple hospitals. Nevertheless, traditional VFL methods primarily benefit patients with overlapping data, leaving non-overlapping patients without guaranteed improvements in distributed healthcare prediction services. While some existing knowledge transfer techniques attempt to improve prediction performance for non-overlapping patients, they fail to adequately address scenarios where overlapping and non-overlapping patients originate from different domains, resulting in challenges such as feature and label heterogeneity. To address these issues, we propose UniTrans, a unified vertical federated knowledge transfer framework for edge healthcare collaboration. Our framework consists of three key steps. First, we extract the federated representation of overlapping patients by employing an effective vertical federated representation learning method to model multi-party joint features online. Next, each hospital learns a local knowledge transfer module offline, enabling the domain-adaptive transfer of knowledge from the federated representation of overlapping patients to the enriched representation of local non-overlapping patients. Finally, hospitals utilize these enriched local representations to enhance performance across various downstream medical prediction tasks. Extensive experiments on real-world medical datasets demonstrate the effectiveness and scalability of UniTrans in both intra-domain and cross-domain knowledge transfer. The code of UniTrans is available athttps://github.com/Chung-ju/Unitrans.

The widespread deployment of WiFi infrastructure has facilitated the development of Time-of-Flight (ToF) based sensing applications. ToF estimation, however, is a challenging task due to the complexity of multipath effect. In this paper, we propose a phase difference based method for ToF estimation and uncover the potential of signal cancellation to mitigate the impact of multipath and noise on phase differences among subcarriers. To separate the moving target path from the complex multipath for ToF estimation, we suggest employing specific elimination methods tailored to the characteristics of different signal components. For dynamic multipath, we observe that when a given subcarrier propagates along two paths to the receiver, with path lengths differing by half a wavelength, the phase difference introduced by these two paths cancels each other out. Therefore, we propose two metrics to identify signals that satisfy this condition, utilizing both frequency diversity and spatial diversity. Additionally, we propose leveraging time diversity to eliminate the static multipath component and reduce the impact of noise. We implemented the methods with off-the-shelf WiFi devices and achieved mean errors of 15.36 cm and 21.05 cm for distance estimation in outdoor and indoor scenarios, outperforming state-of-the-art ToF estimation method by 50% error reduction.

Fire events threaten the safety of residents and the health of ecosystems in affected areas, and post-disaster recovery efforts also require a large investment of resources and time. In recent years, the rising frequency of fire events has motivated local governments to strengthen their monitoring and emergency response efforts. However, current fire event identification methods can only identify the presence of a fire, without the ability to distinguish its specific category. In addition, when a fire occurs, the lack of information about the affected areas makes it challenging for emergency management authorities to take timely and effective rescue measures. To address these issues, we propose a two-stage framework for fire event identification and assessment. Specifically, in the first stage, based on multi-band fused remote sensing images and heterogeneous environmental images, the proposed framework not only identifies various fire events but also accurately identifies the boundaries of the fire events. In the second stage, integrating the results of fire event identification with social media data and domain knowledge, we present a real-time assessment agent for fire events based on the large language model. This agent enables timely and accurate analysis of the impact of fires on the affected areas. We evaluate our method on a real-world authority dataset, and results show that our framework identifies fire events with an F1-score of 61.0$\%$and a mAP of 57.7$\%$, which outperforms state-of-the-art baseline methods. In addition, the assessment results of fire events in real cases indicate that the proposed fire event assessment agent can assist emergency responders in obtaining timely and accurate information.

Spatiotemporal prediction (STP) service is one of the key infrastructure applications in smart cities. Currently, most of the existing STP services are constructed following the workflow of building deep learning (DL) applications while neglecting the importance of domain knowledge and region partition. However, the performance and interpretability of STP are highly related to them. As a result, there is an urgent requirement to develop a thorough and tailored workflow for STP services. To address this gap, we propose a novel workflow including two factors above as intermediate procedures. Based on the workflow, we design and implement an STP toolbox called UCTB (Urban Computing Tool Box) assisting practitioners in the rapid construction of STP services, which can manage multiple spatiotemporal do-main knowledge, support various region partition algorithms, and possess state-of-the-art models simultaneously. The relevant code and supporting documents have been open-sourced at https://github.com/uctb/UCIB.

Self-supervised learning shows promise in harnessing extensive unlabeled data, but it also confronts significant privacy concerns, especially in vision. In this paper, we aim to perform membership inference on visual self-supervised models in a more realistic setting: self-supervised training method and details are unknown for an adversary when attacking as he usually faces a black-box system in practice. In this setting, considering that self-supervised model could be trained by completely different self-supervised paradigms, e.g., masked image modeling and contrastive learning, with complex training details, we propose a unified membership inference method called PartCrop. It is motivated by the shared part-aware capability among models and stronger part response on the training data. Specifically, PartCrop crops parts of objects in an image to query responses with the image in representation space. We conduct extensive attacks on self-supervised models with different training protocols and structures using three widely used image datasets. The results verify the effectiveness and generalization of PartCrop. Moreover, to defend against PartCrop, we evaluate two common approaches, i.e., early stop and differential privacy, and propose a tailored method called shrinking crop scale range. The defense experiments indicate that all of them are effective. Our code is available at https://github.com/JiePKU/PartCrop.

Temporal (ST) prediction is crucial for making informed decisions in urban location-based applications like ride-sharing. However, existing ST models often require region partition as a prerequisite, resulting in two main pitfalls. Firstly, location-based services necessitate ad-hoc regions for various purposes, requiring multiple ST models with varying scales and zones, which can be costly to support. Secondly, different ST models may produce conflicting outputs, resulting in confusing predictions. In this paper, we propose One4All-ST, a framework that can conduct ST prediction for arbitrary modifiable areal units using only one model. To reduce the cost of getting multi-scale predictions, we design an ST network with hierarchical spatial modeling and scale normalization modules to efficiently and equally learn multi-scale representations. To address prediction inconsistencies across scales, we propose a dynamic programming scheme to solve the formulated optimal combination problem, minimizing predicted error through theoretical analysis. Besides, we suggest using an extended quad-tree to index the optimal combinations for quick response to arbitrary modifiable areal units in practical online scenarios. Extensive experiments on two real-world datasets verify the efficiency and effectiveness of One4All-ST in ST prediction for arbitrary modifiable areal units. The source codes and data of this work are available at https://github.com/uctb/One4All-ST.

Transferring knowledge across diverse data modalities is receiving increasing attention in machine learning. This paper tackles the task of leveraging expert-derived, yet expensive, tabular data to enhance image-based predictions when tabular data is unavailable during inference. The primary challenges stem from the inherent complexity of accurately mapping diverse tabular data to visual contexts, coupled with the necessity to devise distinct strategies for numerical and categorical tabular attributes. We propose CHannel tAbulaR alignment with optiMal tranSport (Charms), which establishes an alignment between image channels and tabular attributes, enabling selective knowledge transfer that is pertinent to visual features. Specifically, Charms measures similarity distributions across modalities to effectively differentiate and transfer relevant tabular features, with a focus on morphological characteristics, enhancing the capabilities of visual classifiers. By maximizing the mutual information between image channels and tabular features, knowledge from both numerical and categorical tabular attributes are extracted. Experimental results demonstrate that Charms not only enhances the performance of image classifiers but also improves their interpretability by effectively utilizing tabular knowledge.

Learning representations of user behavior sequences is crucial for various online services, such as online fraudulent transaction detection mechanisms. Graph Neural Networks (GNNs) have been extensively applied to model sequence relationships, and extract information from similar sequences. While user behavior sequence data volume is usually huge for online applications, directly applying GNN models may lead to substantial computational overhead during both the training and inference stages and make it challenging to meet real-time requirements for online services. In this paper, we leverage graph compression techniques to alleviate the efficiency issue. Specifically, we propose a novel unified framework called ECSeq, to introduce graph compression techniques into relation modeling for user sequence representation learning. The key module of ECSeq is sequence relation modeling, which explores relationships among sequences to enhance sequence representation learning, and employs graph compression algorithms to achieve high efficiency and scalability. ECSeq also exhibits plug-and-play characteristics, seamlessly augmenting pre-trained sequence representation models without modifications. Empirical experiments on both sequence classification and regression tasks demonstrate the effectiveness of ECSeq. Specifically, with an additional training time of tens of seconds in total on 100,000+ sequences and inference time preserved within 10−4seconds/sample, ECSeq improves the prediction [email protected] the widely used LSTM by ∼5%.

Spatio-temporal traffic prediction (STTP) plays a crucial role in the development of smart cities. Deep learning models have shown superior performance in traffic prediction, but their opacity and complexity of traffic data present challenges for researchers in tuning models. To tune models effectively, we propose a generalized error analysis pipeline and design a corresponding visualization system, STError-Copilot (Spatio-temporal Error Copilot). The pipeline analyzes multi-perspective spatio-temporal features to determine whether prediction errors originate from semantic or modeling levels, and subsequently tunes the model. STErrorCopilot provides a comprehensive data analysis solution, covering the entire workflow from data loading, processing and visualization to final tuning, delivering end-to-end services. We perform error analysis and tuning on two classic models using two real datasets, demonstrating that our method accurately identifies errors and provides appropriate tuning recommendations.

Text-to-image diffusion has attracted vast attention due to its impressive image-generation capabilities. However, when it comes to human-centric text-to-image generation, particularly in the context of faces and hands, the results often fall short of naturalness due to insufficient training priors. We alleviate the issue in this work from two perspectives. 1) From the data aspect, we carefully collect a human-centric dataset comprising over one million high-quality human-in-the-scene images and two specific sets of close-up images of faces and hands. These datasets collectively provide a rich prior knowledge base to enhance the human-centric image generation capabilities of the diffusion model. 2) On the methodological front, we propose a simple yet effective method called Mixture of Low-rank Experts (MoLE) by considering low-rank modules trained on close-up hand and face images respectively as experts. This concept draws inspiration from our observation of low-rank refinement, where a low-rank module trained by a customized close-up dataset has the potential to enhance the corresponding image part when applied at an appropriate scale. To validate the superiority of MoLE in the context of human-centric image generation compared to state-of-the-art, we construct two benchmarks and perform evaluations with diverse metrics and human studies. Datasets, model, and code are released at https://sites.google.com/view/mole4diffuser/.

Graph contrastive learning (GCL) has emerged as a state-of-the-art strategy for learning representations of diverse graphs including social and biomedical networks. GCL widely uses stochastic graph topology augmentation, such as uniform node dropping, to generate augmented graphs. However, such stochastic augmentations may severely damage the intrinsic properties of a graph and deteriorate the following representation learning process. We argue that incorporating an awareness of cohesive subgraphs during the graph augmentation and learning processes has the potential to enhance GCL performance. To this end, we propose a novel unified framework called CTAug, to seamlessly integrate cohesion awareness into various existing GCL mechanisms. In particular, CTAug comprises two specialized modules: topology augmentation enhancement and graph learning enhancement. The former module generates augmented graphs that carefully preserve cohesion properties, while the latter module bolsters the graph encoder's ability to discern subgraph patterns. Theoretical analysis shows that CTAug can strictly improve existing GCL mechanisms. Empirical experiments verify that CTAug can achieve state-of-the-art performance for graph representation learning, especially for graphs with high degrees. The code is available at https://doi.org/10.5281/zenodo.10594093, or https://github.com/wuyucheng2002/CTAug.

Recent advancements in Wi-Fi-based sensing technologies have enabled effective hand gesture recognition. However, most studies focus on single gesture recognition and fail to recognize naturally performed continuous gestures without pauses in transitions. The main challenges include diverse and uncertain transitions in continuous gesture recognition, making it difficult to segment and identify gestures from a stream of continuous hand movements. In this paper, we introduce a new method to recognize continuously performed gestures from a set of predefined gestures (e.g., digits) without requiring a pause in transitions. Instead of segmenting gestures at the gesture-transition level, we segment the stream into basic fractions that depict exclusive moving patterns of gestures. We propose a novel feature called meta motion, which geometrically characterizes different basic hand movements. Leveraging this feature, we use a back-tracking searching-based algorithm to identify gestures from the sequence of meta motions. Based on this approach, we develop a prototype system, WiCGesture, on commodity Wi-Fi devices. WiCGesture is the first system engaging in continuous gesture recognition using Wi-Fi signals. Evaluation results show that WiCGesture effectively recognizes continuous gestures from two gesture sets, significantly outperforming state-of-the-art methods.

Sparse mobile crowdsensing (MCS) is a cost-effective data collection paradigm that aims to recruit users to collect data from a part of sensing subareas and infer the rest. In a more realistic scenario, users participate in real-time and collect data along the way. For missing data inference, the significance of data collected from different subareas often varies over time. However, since users’ trajectories are uncertain, recruiting users who can cover important spatio-temporal subareas presents a challenge. Additionally, how to segment the budget wisely during recruitment is another challenge. To tackle these challenges, we propose a dual reinforcement learning (RL)-based online user recruitment strategy with adaptive budget segmentation, called DualRL-U, which consists of two alternating decision steps, i.e., the user recruitment decision and the budget retention decision. Specifically, for the user recruitment decision, we use RL to connect the user with data inference accuracy to estimate their contributions. For the budget retention decision, we use RL to connect the budget with the number of times the user can sense to evaluate the cost effectiveness. In this way, a dual RL model is constructed to achieve effective recruitment by alternately executing user recruitment decisions and budget retention decisions. Extensive experiments on real-world sensing data sets show the effectiveness of DualRL-U.

Network embedding represents network nodes by a low-dimensional informative vector. While it is generally effective for various downstream tasks, it may leak some private information of networks, such as hidden private links. In this work, we address a novel problem ofprivacy-preserving network embedding against private link inference attacks. Basically, we propose to perturb the original network by adding or removing links, and expect the embedding generated on the perturbed network can leak little information about private links but hold high utility for various downstream tasks. Towards this goal, we first propose general measurements to quantify privacy gain and utility loss incurred by candidate network perturbations; we then design aPrivacy-PreservingNetworkEmbedding (i.e., PPNE) framework to identify the optimal perturbation solution with the best privacy-utility trade-off in an iterative way. Furthermore, we propose many techniques to accelerate PPNE and ensure its scalability. For instance, as the skip-gram embedding methods including DeepWalk and LINE can be seen as matrix factorization with closed-form embedding results, we devise efficient privacy gain and utility loss approximation methods to avoid the repetitive time-consuming embedding training for every candidate network perturbation in each iteration. Experiments on real-life network datasets (with up to millions of nodes) verify that PPNE outperforms baselines by sacrificing less utility and obtaining higher privacy protection.

Despite achieving remarkable performance, Federated Learning (FL) encounters two important problems, i.e., low training efficiency and limited computational resources. In this article, we propose a new FL framework, i.e., FedDUMAP, with three original contributions, to leverage the shared insensitive data on the server in addition to the distributed data in edge devices so as to efficiently train a global model. First, we propose a simple dynamic server update algorithm, which takes advantage of the shared insensitive data on the server while dynamically adjusting the update steps on the server in order to speed up the convergence and improve the accuracy. Second, we propose an adaptive optimization method with the dynamic server update algorithm to exploit the global momentum on the server and each local device for superior accuracy. Third, we develop a layer-adaptive model pruning method to carry out specific pruning operations, which is adapted to the diverse features of each layer so as to attain an excellent tradeoff between effectiveness and efficiency. Our proposed FL model, FedDUMAP, combines the three original techniques and has a significantly better performance compared with baseline approaches in terms of efficiency (up to 16.9 times faster), accuracy (up to 20.4% higher), and computational cost (up to 62.6% smaller).

Urban events, such as hurricanes, floods, and epidemic outbreaks, usually have a significant impact on crowd behaviors that may dramatically change people’s mobility patterns, interaction manners, and etc. Accurately foreseeing the abnormal crowd behaviors triggered by urban events can help authorities to dynamically schedule urban resources and services, such as providing shuttle buses, temporal shelters, and psychological first aid. However, traditional crowd behavior prediction models tend to learn regular trends and patterns of crowd traffic, and usually fail to predict the abnormal traffic fluctuations triggered by urban events. In this work, we propose a context-aware framework to accurately predict event-triggered abnormal crowd traffic by explicitly modeling event contexts and their impact. We start by training a state-of-the-art spatiotemporal multi-graph model (MG) leveraging a multi-graph convolutional network (MGCN) to model city-wide venues and their connections, as well as gated recurrent unit (GRU) to capture the crowd traffic patterns within venues. In order to model the time-occasional impact of sudden influential event context (e.g., weather conditions), we propose a multi-task fusion diagram to co-train the MG model by learning a temporal impact embedding with an attention mechanism and recurrent neural network to obtain the MG-MT model. In order to model the hyper-space impact of irregularly distributed event context (e.g., epicenters and trajectories), we propose a multi-view fusion diagram to fine-tune the above-mentioned model by learning a spatial impact graph with a stimulus-response mechanism to obtain the spatiotemporal crowd traffic model (STORM). Experiments using real-world urban data collected from Xiamen, China show that our approach improves over the state-of-the-art baseline by more than 6% in predicting crowd traffic and by more than 40% in the abnormal crowd traffic part.

Evaluation is a systematic approach to assessing how well a system achieves its intended purpose. Federated learning (FL) is a novel paradigm for privacy-preserving machine learning that allows multiple parties to collaboratively train models without sharing sensitive data. However, evaluating FL is challenging due to its interdisciplinary nature and diverse goals, such as utility, efficiency, and security. In this survey, we first review the major evaluation goals adopted in the existing studies and then explore the evaluation metrics used for each goal. We also introduceFedEval, an open-source platform that provides a standardized and comprehensive evaluation framework for FL algorithms in terms of their utility, efficiency, and security. Finally, we discuss several challenges and future research directions for FL evaluation.

Sparse crowdsensing collects data from a subset of the sensing area and infers data for unsensed areas, reducing data collection costs. Previous works have primarily focused on independently collecting and inferring single types of data. However, real-world scenarios often involve multiple types of data that can complement each other by providing missing spatiotemporal distribution information. In this paper, we fully consider both intra-data correlations among data of the same type and inter-data correlations among data of different types, enabling collaborative execution of various tasks. In addition, we enhance the adaptability in practical application scenarios by utilizing real-time collected sparse data to guide task execution. For this purpose, we propose a multi-task adaptive budgeting framework for online sparse crowdsensing, called MTAB-SC. This framework consists of three parts: training data updating, data inference, and data collection. Firstly, we propose a multi-task data updating method to keep models up-to-date. Secondly, we design a data inference network for multi-task data joint inference. Finally, to allocate suitable budgets for each task and facilitate collaborative data collection across multiple tasks, we propose an Adaptive Budgeting for Collaborative Data Collection model (AB-CoDC). The effectiveness of our proposals is demonstrated through extensive experiments on two real-world datasets.

Contactless respiration monitoring using wireless signals has drawn much attention in recent years. Many approaches have been proposed, however, they may not work when there is a lack of signals directly reflected from target's chest, e.g., a target faces away from the transceiver or a target is blocked by furniture. In this paper, we design and implement a novel omnimonitoring system for human respiration,OmniRespMonitor, using a pair of speaker and microphone. Different from Radio Frequency (RF) signal, acoustic signals cannot penetrate through walls and furniture. The multipath reflection in an indoor environment will result in highly abundant acoustic signals. In this case, even though there are lack of acoustic signals directly reflected by a target's chest, indirectly-reflected acoustic signals can still be received by the microphone. We can therefore monitor the target's respiration by extracting this subtle variation of indirectly reflected signals. To achieve this, we model chest movement using truncated System Frequency Response (SFR). We then develop a global search method based on the autocorrelation function to extract minute chest movement from SFR sequences. Finally, we dynamically synthesize the chest movement information to recover the breathing wave in real time. We conduct extensive experiments with both humans and animals (goat), the results show thatOmniResMonitoris able to monitor single target's respiration within 5 meters in indoor environments in various challenging scenarios there are lack of directly-reflected acoustic signals.

The size of deep learning models in artificial intelligence (AI) software is increasing rapidly, which hinders the large-scale deployment on resource-restricted devices (e.g., smartphones). To mitigate this issue, AI software compression plays a crucial role, which aims to compress model size while keeping high performance. However, the intrinsic defects in the big model may be inherited by the compressed one. Such defects may be easily leveraged by attackers, since the compressed models are usually deployed in a large number of devices without adequate protection. In this paper, we try to address the safe model compression problem from a safety-performance co-optimization perspective. Specifically, inspired by the test-driven development (TDD) paradigm in software engineering, we propose a test-driven sparse training framework calledSafeCompress. By simulating the attack mechanism as the safety test, SafeCompress can automatically compress a big model to a small one following the dynamic sparse training paradigm. Then, considering two kinds of representative and heterogeneous attack mechanisms i.e., black-box membership inference attack and white-box membership inference attack, we develop two concrete instances called BMIA-SafeCompress and WMIA-SafeCompress. Further, we implement another instance called MMIA-SafeCompress by extending SafeCompress to defend against the occasion when attackers conduct black-box and white-box membership inference attacks simultaneously. Extensive experiments are conducted on five datasets for both computer vision and natural language processing tasks. The results verify the effectiveness and generalizability of our method. We also discuss how to adapt SafeCompress to other attacks besides MIA, demonstrating the flexibility of SafeCompress.

Most state-of-the-art deep domain adaptation techniques align source and target samples in a global fashion. That is, after alignment, each source sample is expected to become similar to any target sample. However, global alignment may not always be optimal or necessary in practice. For example, consider cross-domain fraud detection, where there are two types of transactions: credit and non-credit. Aligning credit and non-credit transactions separately may yield better performance than global alignment, as credit transactions are unlikely to exhibit patterns similar to non-credit transactions. To enable such fine-grained domain adaption, we propose a novel Knowledge-Inspired Subdomain Adaptation (KISA) framework. In particular, (1) We provide the theoretical insight that KISA minimizes the shared expected loss which is the premise for the success of domain adaptation methods. (2) We propose the knowledge-inspired subdomain division problem that plays a crucial role in fine-grained domain adaption. (3) We design a knowledge fusion network to exploit diverse domain knowledge. Extensive experiments demonstrate that KISA achieves remarkable results on fraud detection and traffic demand prediction tasks.

MAUP (modifiable areal unit problem) is a fundamental problem for spatial data management and analysis. As an instantiation of MAUP in online transportation platforms, region generation (i.e., specifying the areal unit for service operations) is the first and vital step for supporting spatiotemporal transportation services such as ride-sharing and freight transport. Most existing region generation methods are manually specified (e.g., fixed-size grids), suffering from poor spatial semantic meaning and inflexibility to meet service operation requirements. In this paper, we propose RegionGen, a data-driven region generation framework that can specify regions with key characteristics (e.g., good spatial semantic meaning and predictability) by modeling region generation as a multi-objective optimization problem. First, to obtain good spatial semantic meaning, RegionGen segments the whole city into atomic spatial elements based on road networks and obstacles (e.g., rivers). Then, it clusters the atomic spatial elements into regions by maximizing various operation characteristics, which is formulated as a multi-objective optimization problem. For this optimization problem, we propose a multi-objective co-optimization algorithm. Extensive experiments verify that RegionGen can generate more suitable regions than traditional methods for spatiotemporal service management.

There are significant regional inequities in health resources around the world. It has become one of the most focused topics to improve health services for data-scarce hospitals and promote health equity through knowledge sharing among medical institutions. Because electronic medical records (EMRs) contain sensitive personal information, privacy protection is unavoidable and essential for multi-hospital collaboration. In this paper, for a common disease in ICU patients, sepsis, we propose a novel cross-center collaborative learning framework guided by medical knowledge, SofaNet, to achieve early recognition of this disease. The Sepsis-3 guideline, published in 2016, defines that sepsis can be diagnosed by satisfying both suspicion of infection and Sequential Organ Failure Assessment (SOFA) greater than or equal to 2. Based on this knowledge, SofaNet adopts a multi-channel GRU structure to predict SOFA values of different systems, which can be seen as an auxiliary task to generate better health status representations for sepsis recognition. Moreover, we only achieve feature distribution alignment in the hidden space during cross-center collaborative learning, which ensures secure and compliant knowledge transfer without raw data exchange. Extensive experiments on two open clinical datasets, MIMIC-III and Challenge, demonstrate that SofaNet can benefit early sepsis recognition when hospitals only have limited EMRs.

Collaboration between healthcare institutions can significantly lessen the imbalance in medical resources across various geographic areas. However, directly sharing diagnostic information between institutions is typically not permitted due to the protection of patients’ highly sensitive privacy. As a novel privacy-preserving machine learning paradigm, federated learning (FL) makes it possible to maximize the data utility among multiple medical institutions. These feature-enrichment FL techniques are referred to as vertical FL (VFL). Traditional VFL can only benefit multi-parties’ shared samples, which strongly restricts its application scope. In order to improve the information-sharing capability and innovation of various healthcare-related institutions, and then to establish a next-generation open medical collaboration network, we propose a unified framework for vertical federated knowledge transfer mechanism (VFedTrans) based on a novel cross-hospital representation distillation component. Specifically, our framework includes three steps. First, shared samples’ federated representations are extracted by collaboratively modeling multi-parties’ joint features with current efficient vertical federated representation learning methods. Second, for each hospital, we learn a local-representation-distilled module, which can transfer the knowledge from shared samples’ federated representations to enrich local samples’ representations. Finally, each hospital can leverage local samples’ representations enriched by the distillation module to boost arbitrary downstream machine learning tasks. The experiments on real-life medical datasets verify the knowledge transfer effectiveness of our framework.

It is becoming more and more promising that marketers hire influencers to launch campaigns for spreading items (e.g., articles or videos about products) over social media platforms. Such social media influencer marketing may generate tremendous utility if the influencers persuade their followers to adopt the recommended items. This could further spur extensive spontaneous item propagation on social media. Although prior studies mainly focus on influencer-selection strategies by the influencers’ traits, marketers with a number of items are often requested to determine both influencers and marketing items. The appropriateness between influencers and items is critical, but rarely considered in prior influencer-identification methods. We thus formulate and solve a novel cost-effective social media influencer marketing problem to maximize marketers’ utility by selecting appropriate pairwise combinations of influencers and items (i.e., item-influencer pairs). In particular, we first model utility functions and propose a simulation-based method to estimate the appropriateness of arbitrarily given item-influencer pairs by their potential utility. With the estimated utility, we devise an algorithm to iteratively select appropriate item-influencer pairs under various realistic conditions, including marketers’ budget, influencers’ payments, item-user fitness, social propagation, and influencers’ marketing slots. We theoretically prove that the marketing utility achieved by our method is near-optimal. We also conduct extensive empirical experiments with three real-world data sets to verify the superiority of our method in terms of cost-effectiveness and computational efficiency. Lastly, we discuss insightful theoretical and practical implications. History: Accepted by Ram Ramesh, Area Editor for Data Science and Machine Learning. Funding: This study was partially funded by the National Natural Science Foundation of China [Grants 72071125, 72031001, and 61972008]. Supplemental Material: The online appendix is available at https://doi.org/10.1287/ijoc.2022.1246 .

Leveraging machine learning techniques for Sepsis early detection and diagnosis has attracted increasing interest in recent years. However, most existing methods require a large amount of labeled training data, which may not be available for a target hospital that deploys a new Sepsis detection system. More seriously, as treated patients are diversified between hospitals, directly applying a model trained on other hospitals may not achieve good performance for the target hospital. To address this issue, we propose a novel semi-supervised transfer learning framework based on optimal transport theory and self-paced ensemble for Sepsis early detection, called SPSSOT, which can efficiently transfer knowledge from the source hospital (with rich labeled data) to the target hospital (with scarce labeled data). Specifically, SPSSOT incorporates a new optimal transport-based semi-supervised domain adaptation component that can effectively exploit all the unlabeled data in the target hospital. Moreover, self-paced ensemble is adapted in SPSSOT to alleviate the class imbalance issue during transfer learning. In a nutshell, SPSSOT is an end-to-end transfer learning method that automatically selects suitable samples from two domains (hospitals) respectively and aligns their feature spaces. Extensive experiments on two open clinical datasets, MIMIC-III and Challenge, demonstrate that SPSSOT outperforms state-of-the-art transfer learning methods by improving 1-3% of AUC.

Knowing why people travel is meaningful for human mobility understanding and smart services development. Unfortunately, in real-world scenarios, trip purpose cannot be automatically collected on a large scale, thus calling for effective prediction models. Nevertheless, since passengers’ trip purposes in the city are diverse and complicated, the prediction is very difficult especially at a fine-grained level. Worse still, the informative data sources and real purpose-labels about trips are commonly limited for model learning. To resolve the dilemma, we propose a semi-supervised deep embedding framework for predicting fine-grained trip purposes on a large scale. Specifically, we first derive augmented trip contexts from the vehicle’s GPS trajectory and public POI check-in data, then convert POI contexts into the graph structure. We further establish aDual-AttentionGraphEmbedding Network withAutoencoder architecture (DAGE-A) to accomplish prediction and reconstruction simultaneously, in which category-aware graph attention networks are devised to model the POI semantics at trip’s origin/destination and extract complementary knowledge from unlabeled trips; and soft-attention is employed to aggregate different trip semantics appropriately for the final prediction. We conduct extensive experiments in Beijing and Shanghai, and results show our framework outperforms state-of-the-arts and could reduce labelling efforts by up to 20%. We also find that our model is generalized at different times and locations, and the performance varies for different trip purposes.

The Spatio-Temporal Traffic Prediction (STTP) problem is a classical problem with plenty of prior research efforts that benefit from traditional statistical learning and recent deep learning approaches. While STTP can refer to many real-world problems, most existing studies focus on quite specific applications, such as the prediction of taxi demand, ridesharing order, traffic speed, and so on. This hinders the STTP research as the approaches designed for different applications are hardly comparable, and thus how an application-driven approach can be generalized to other scenarios is unclear. To fill in this gap, this paper makes three efforts: (i) we propose an analytic framework, called STAnalytic, to qualitatively investigate STTP approaches regarding their design considerations on various spatial and temporal factors, aiming to make different application-driven approaches comparable; (ii) we design a spatio-temporal meta-model, called STMeta, which can flexibly integrate generalizable temporal and spatial knowledge identified by STAnalytic, (iii) we build an STTP benchmark platform including ten real-life datasets with five scenarios to quantitatively measure the generalizability of STTP approaches. In particular, we implement STMeta with different deep learning techniques, and STMeta demonstrates better generalizability than state-of-the-art approaches by achieving lower prediction error on average across all the datasets.

The size of deep learning models in artificial intelligence (AI) software is increasing rapidly, which hinders the large-scale deployment on resource-restricted devices (e.g., smartphones). To mitigate this issue, AI software compression plays a crucial role, which aims to compress model size while keeping high performance. However, the intrinsic defects in the big model may be inherited by the compressed one. Such defects may be easily leveraged by attackers, since the compressed models are usually deployed in a large number of devices without adequate protection. In this paper, we try to address the safe model compression problem from a safety-performance co-optimization perspective. Specifically, inspired by the test-driven development (TDD) paradigm in software engineering, we propose a test-driven sparse training framework called SafeCompress. By simulating the attack mechanism as the safety test, SafeCompress can automatically compress a big model to a small one following the dynamic sparse training paradigm. Further, considering a representative attack, i.e., membership inference attack (MIA), we develop a concrete safe model compression mechanism, called MIA-SafeCompress. Extensive experiments are conducted to evaluate MIA-SafeCompress on five datasets for both computer vision and natural language processing tasks. The results verify the effectiveness and generalization of our method. We also discuss how to adapt SafeCompress to other attacks besides MIA, demonstrating the flexibility of SafeCompress.

With the enactment of privacy-preserving regulations, e.g., GDPR, federated SVD is proposed to enable SVD-based applications over different data sources without revealing the original data. However, many SVD-based applications cannot be well supported by existing federated SVD solutions. The crux is that these solutions, adopting either differential privacy (DP) or homomorphic encryption (HE), suffer from accuracy loss caused by unremovable noise or degraded efficiency due to inflated data.

Recommender systems typically require the revelation of users’ ratings to the recommender server, which will subsequently use these ratings to provide personalized services. However, such revelations make users vulnerable to a broader set of inference attacks, allowing the recommender server to learn users’ private attributes, e.g., age and gender. Therefore, in this paper, we propose an efficient federated matrix factorization method that protects users against inference attacks. The key idea is that we obfuscate one user’s rating to another such that the private attribute leakage is minimized under the given distortion budget, which bounds the recommending loss and overhead of system efficiency. During the obfuscation, we apply differential privacy to control the information leakage between the users. We also adopt homomorphic encryption to protect the intermediate results during training. Our framework is implemented and tested on real-world datasets. The result shows that our method can reduce up to 16.7% of inference attack accuracy compared to using no privacy protections.

Sleep monitoring is essential to people's health and wellbeing, which can also assist in the diagnosis and treatment of sleep disorder. Compared with contact-based solutions, contactless sleep monitoring does not attach any device to the human body; hence, it has attracted increasing attention in recent years. Inspired by the recent advances in Wi-Fi-based sensing, this article proposes a low-cost and nonintrusive sleep monitoring system using commodity Wi-Fi devices, namely, WiFi-Sleep. We leverage the fine-grained channel state information from multiple antennas and propose advanced fusion and signal processing methods to extract accurate respiration and body movement information. We introduce a deep learning method combined with clinical sleep medicine prior knowledge to achieve four-stage sleep monitoring with limited data sources (i.e., only respiration and body movement information). We benchmark the performance of WiFi-Sleep with polysomnography, the gold reference standard. Results show that WiFi-Sleep achieves an accuracy of 81.8%, which is comparable to the state-of-the-art sleep stage monitoring using expensive radar devices.

In mobile crowdsourcing, organizers usually need participants' precise locations for optimal task allocation, e.g., minimizing selected workers' travel distance to task locations. However, the exposure of users' locations raises privacy concerns. In this paper, we propose a location privacy-preserving task allocation framework with geo-obfuscation to protect users' locations during task assignments. More specifically, we make participants obfuscate their reported locations under the guarantee of two rigorous privacy-preserving schemes, differential and distortion privacy, without the need to involve any third-party trusted entity. In order to achieve optimal task allocation with the differential-and-distortion geo-obfuscation, we formulate a mixed-integer non-linear programming problem to minimize the expected travel distance of the selected workers under the constraints of differential and distortion privacy. Moreover, a worker may be willing to accept multiple tasks, and a task organizer may be concerned with multiple utility objectives such as task acceptance ratio in addition to travel distance. Against this background, we also extend our solution to the multi-task allocation and multi-objective optimization cases. Evaluation results on both simulation and real-world user mobility traces verify the effectiveness of our framework. Particularly, our framework outperforms Laplace obfuscation, a state-of-the-art geo-obfuscation mechanism, by achieving up to 47 percent shorter average travel distance on real-world data under the same level of privacy protection.

Device-free hand gesture is one of the most natural ways to interact with everyday objects. However, existing WiFi-based gesture recognition solutions are typically restricted to indoor environments due to limited outdoor coverage. Furthermore, to achieve high sampling rates, they may interfere with normal data transmissions. In this paper, we aim to develop a robust dynamic gesture interaction system that can be ubiquitously deployed using Long-term Evolution (LTE) mobile terminals. Through both empirical studies and in-depth analysis using the Fresnel zone model, we reveal the key factors that contribute to the repeatability and discernibility of gestures. We show that the optimal location and orientation to perform gestures indeed exist and can be identified without prior knowledge of the position of LTE base stations (BSs) relative to a terminal. Guided by the design principles derived from Fresnel zone characteristics around a 4G terminal, we design highly repeatable and discernible gestures with salient received signal profiles. A gesture interaction system has been developed and implemented to achieve robust recognition with this careful design. Extensive experiments have been conducted in both indoor and outdoor environments, for different relative placements of mobile terminal and BS, and with different users. The proposed system can automatically identify the direction of BSs with a median error of less than 15 degrees and achieve gesture recognition accuracy as high as 98% in all scenarios without the need to acquire any training data.

Sparse Mobile Crowdsensing (MCS) has become a compelling approach to acquire and infer urban-scale sensing data. However, participants risk their location privacy when reporting data with their actual sensing positions. To address this issue, we propose a novel location obfuscation mechanism combining E-differential-privacy and δ-distortion-privacy in Sparse MCS. More specifically, differential privacy bounds adversaries' relative information gain regardless of their prior knowledge, while distortion privacy ensures that the expected inference error is larger than a threshold under an assumption of adversaries' prior knowledge. To reduce the data quality loss incurred by location obfuscation, we design a differential-and-distortion privacy-preserving framework with three components. First, we learn a data adjustment function to fit the original sensing data to the obfuscated location. Second, we apply a linear program to select an optimal location obfuscation function. The linear program aims to minimize the uncertainty in data adjustment under the constraints of E-differential-privacy, δ-distortion-privacy, and evenly-distributed obfuscation. We also design an approximated method to reduce the required computation resources. Third, we propose an uncertainty-aware inference algorithm to improve the inference accuracy for the obfuscated data. Evaluations with real environment and traffic datasets show that our optimal method reduces the data quality loss by up to 42% compared to the state-of-the-art methods with the same level of privacy protection; the approximated method incurs <; 3% additional quality loss than the optimal method, but only needs <; 1% of the computation time.

Task allocation is a major challenge in Mobile Crowd Sensing (MCS). While previous task allocation approaches follow either the opportunistic or participatory mode, this paper proposes to integrate these two complementary modes in a two-phased hybrid framework called HyTasker. In the offline phase, a group of workers (called opportunistic workers) are selected, and they complete MCS tasks during their daily routines (i.e., opportunistic mode). In the online phase, we assign another set of workers (called participatory workers) and require them to move specifically to perform tasks that are not completed by the opportunistic workers (i.e., participatory mode). Instead of considering these two phases separately, HyTasker jointly optimizes them with a total incentive budget constraint. In particular, when selecting opportunistic workers in the offline phase of HyTasker, we propose a novel algorithm that simultaneously considers the predicted task assignment for the participatory workers, in which the density and mobility of participatory workers are taken into account. Experiments on two real-world mobility datasets demonstrate that HyTasker outperforms other methods with more completed tasks under the same budget constraint.

Region-level demand forecasting is an essential task in ridehailing services. Accurate ride-hailing demand forecasting can guide vehicle dispatching, improve vehicle utilization, reduce the wait-time, and mitigate traffic congestion. This task is challenging due to the complicated spatiotemporal dependencies among regions. Existing approaches mainly focus on modeling the Euclidean correlations among spatially adjacent regions while we observe that non-Euclidean pair-wise correlations among possibly distant regions are also critical for accurate forecasting. In this paper, we propose the spatiotemporal multi-graph convolution network (ST-MGCN), a novel deep learning model for ride-hailing demand forecasting. We first encode the non-Euclidean pair-wise correlations among regions into multiple graphs and then explicitly model these correlations using multi-graph convolution. To utilize the global contextual information in modeling the temporal correlation, we further propose contextual gated recurrent neural network which augments recurrent neural network with a contextual-aware gating mechanism to re-weights different historical observations. We evaluate the proposed model on two real-world large scale ride-hailing demand datasets and observe consistent improvement of more than 10% over stateof-the-art baselines.

We consider the problem of mobilizing community effort to reposition indiscriminantly parked shared bikes in urban environments through crowdsourcing. We propose an ethically aligned incentive optimization approach WSLS which maximizes the rate of success for bike repositioning while minimizing cost and prioritizing users’ wellbeing. Realistic simulations based on a dataset from Singapore demonstrate that WSLS significantly outperforms existing approaches.

Drug-drug interactions (DDIs) are a major cause of preventable hospitalizations and deaths. Recently, researchers in the AI community try to improve DDI prediction in two directions, incorporating multiple drug features to better model the pharmacodynamics and adopting multi-task learning to exploit associations among DDI types. However, these two directions are challenging to reconcile due to the sparse nature of the DDI labels which inflates the risk of overfitting of multi-task learning models when incorporating multiple drug features. In this paper, we propose a multi-task semi-supervised learning framework MLRDA for DDI prediction. MLRDA effectively exploits information that is beneficial for DDI prediction in unlabeled drug data by leveraging a novel unsupervised disentangling loss CuXCov. The CuXCov loss cooperates with the classification loss to disentangle the DDI prediction relevant part from the irrelevant part in a representation learnt by an autoencoder, which helps to ease the difficulty in mining useful information for DDI prediction in both labeled and unlabeled drug data. Moreover, MLRDA adopts a multi-task learning framework to exploit associations among DDI types. Experimental results on real-world datasets demonstrate that MLRDA significantly outperforms state-of-the-art DDI prediction methods by up to 10.3% in AUPR.

Spatio-temporal prediction is a key type of tasks in urban computing, e.g., traffic flow and air quality. Adequate data is usually a prerequisite, especially when deep learning is adopted. However, the development levels of different cities are unbalanced, and still many cities suffer from data scarcity. To address the problem, we propose a novel cross-city transfer learning method for deep spatio-temporal prediction tasks, called RegionTrans. RegionTrans aims to effectively transfer knowledge from a data-rich source city to a data-scarce target city. More specifically, we first learn an inter-city region matching function to match each target city region to a similar source city region. A neural network is designed to effectively extract region-level representation for spatio-temporal prediction. Finally, an optimization algorithm is proposed to transfer learned features from the source city to the target city with the region matching function. Using citywide crowd flow prediction as a demonstration experiment, we verify the effectiveness of RegionTrans. Results show that RegionTrans can outperform the state-of-the-art fine-tuning deep spatio-temporal prediction models by reducing up to 10.7% prediction error.

Recent years have witnessed advances of Internet of Things technologies and their applications to enable contactless sensing and elderly care in smart homes. Continuous and real-time respiration monitoring is one of the important applications to promote assistive living for elders during sleep and attracted wide attention in both academia and industry. Most of the existing respiration monitoring systems require expensive and specialized devices to sense chest displacement. However, chest displacement is not a direct indicator of breathing and thus false detection may often occur. In this paper, we design and implement a real-time and contactless respiration monitoring system by directly sensing the exhaled airflow from breathing using ultrasound signals with off-the-shelf speaker and microphone. Exhaled airflow from breathing can be regarded as air turbulence, which scatters the sound wave and results in Doppler effect. Our system works as an acoustic radar which transmits sound wave and detects the Doppler effect caused by breathing airflow. We mathematically model the relationship between the Doppler frequency change and the direction of breathing airflow. Based on this model, we design a minimum description length-based algorithm to effectively capture the Doppler effect caused by exhaled airflow. We conduct extensive experiments with 25 participants (7 elders, 2 young kids, and 16 adults, including 11 females and 14 males) in four different rooms. The participants take four different sleep postures (lying on one's back, on right/left side, and on one's stomach) in different positions of the bed. Experiment results show that our system achieves a median error lower than 0.3 breaths/min (2%) for respiration monitoring and can accurately identify Apnea. The results also demonstrate that the system is robust to different respiration styles (shallow, normal, and deep), respiration rate variation, ambient noise, sensing distance variation (within 0.7 m), and transmitted signal frequency variation.

In online social networks, users always expect to share some information for benefits (e.g., personalized services) while hiding the others for privacy. Unfortunately, the hidden information is likely to be predicted by various powerful inference attacks with the rapid advances in machine learning. Then, what is the risk that a user's private information could be disclosed? What countermeasures can be taken to fight against the privacy violation for the user? To tackle these issues, this article proposes a general Framework for Private Attribute Disclosure estimation (F-PAD) including three steps: 1) private attribute prediction; 2) disclosure model training; 3) disclosure risk estimation. Not like most prior risk estimation studies focusing on one specific attack model and private attribute, F-PAD can estimate disclosure risk for individual users in terms of disclosure probability and risk level within a high confidence given a basket of potential inference attack models; furthermore, F-PAD can adapt to various attributes (e.g., gender, age) and offer countermeasures to help users lower the risk. Extensive experiment studies on two real social network datasets, Facebook and Book-Crossing, have verified the effectiveness of F-PAD in `current city', `gender' and `age' disclosure risk estimation.

This paper proposes a novel task allocation framework, PSTasker, for participatory sensing (PS), which aims to maximize the overall system utility on PS platform by coordinating the allocation of multiple tasks. While existing studies mainly optimize the task allocation from the perspective of the task organizer (e.g., maximizing coverage or minimizing incentive cost), PSTasker further considers diverse factors on the participants' side, including user work bandwidth, user availability, devices' sensor configuration, task completion likelihood, and mobility pattern. Furthermore, by considering the heterogeneity in three dimensions (i.e., task, time, and space), it adopts a novel model to measure task sensing quality and overall system utility. In PSTasker, it first calculates the utlity of a given task allocation plan by jointly fusing different participant-side factors into one unified estimation function, and then employs an iterative greedy process to optimize the task allocation. Extensive evaluations based on real-world mobility traces demonstrate that PSTasker outperforms the baseline methods under various settings.

Worker recruitment is a crucial research problem in Mobile Crowd Sensing (MCS). While previous studies rely on a specified platform with a pre-assumed large user pool, this paper leverages the influence propagation on the social network to assist the MCS worker recruitment. We first select a subset of users on the social network as initial seeds and push MCS tasks to them. Then, influenced users who accept tasks are recruited as workers, and the ultimate goal is to maximize the coverage. Specifically, to select a near-optimal set of seeds, we propose two algorithms, named Basic-Selector and Fast-Selector, respectively. Basic-Selector adopts an iterative greedy process based on the predicted mobility, which has good performance but suffers from inefficiency concerns. To accelerate the selection, Fast-Selector is proposed, which is based on the interdependency of geographical positions among friends. Empirical studies on two real-world datasets verify that Fast-Selector achieves higher coverage than baseline methods under various settings, meanwhile, it is much more efficient than Basic-Selector while only sacrificing a slight fraction of the coverage.

For real-world mobile applications such as location-based advertising and spatial crowdsourcing, a key to success is targeting mobile users that can maximally cover certain locations in a future period. To find an optimal group of users, existing methods often require information about users' mobility history, which may cause privacy breaches. In this paper, we propose a method to maximize mobile crowd's future location coverage under a guaranteed location privacy protection scheme. In our approach, users only need to upload one of their frequently visited locations, and more importantly, the uploaded location is obfuscated using a geographic differential privacy policy. We propose both analytic and practical solutions to this problem. Experiments on real user mobility datasets show that our method significantly outperforms the state-of-the-art geographic differential privacy methods by achieving a higher coverage under the same level of privacy protection.

One fundamental issue in managing bike sharing systems is bike flow prediction. Due to the hardness of predicting flow for a single station, recent research often predicts flow at cluster-level. However, they cannot directly guide fine-grained system management issues at station-level. In this paper, we revisit the problem of the station-level bike flow prediction, aiming to boost the prediction accuracy using the breakthroughs of deep learning techniques. We propose a multi-graph convolutional neural network model to predict flow at station-level, where the key novelty is viewing the bike sharing system from the graph perspective. More specifically, we construct multiple graphs for a bike sharing system to reflect heterogeneous inter-station relationships. Afterward, we fuse multiple graphs and apply the convolutional layers to predict station-level future bike flow. The results on realistic bike flow datasets verify that our multi-graph model can outperform state-of-the-art prediction models by reducing up to 25.1% prediction error.

Sparse Mobile CrowdSensing (MCS) is a novel MCS paradigm where data inference is incorporated into the MCS process for reducing sensing costs while its quality is guaranteed. Since the sensed data from different cells (sub-areas) of the target sensing area will probably lead to diverse levels of inference data quality, cell selection (i.e., choose which cells of the target area to collect sensed data from participants) is a critical issue that will impact the total amount of data that requires to be collected (i.e., data collection costs) for ensuring a certain level of quality. To address this issue, this paper proposes a Deep Reinforcement learning based Cell selection mechanism for Sparse MCS, called DR-Cell. We properly model the key concepts in reinforcement learning including state, action, and reward, and then propose to use a deep recurrent Q-network for learning the Q-function that can help decide which cell is a better choice under a certain state during cell selection. Experiments on various real-life sensing datasets verify the effectiveness of DR-Cell over the state-of-the-art cell selection mechanisms in Sparse MCS by reducing up to 15% of sensed cells with the same data inference quality guarantee.

Spatial crowdsourcing (SC) is an emerging paradigm of crowdsourcing, which commits workers to move to some particular locations to perform spatio-temporal-relevant tasks (e.g., sensing and activity organization). Task allocation or worker selection is a significant problem that may impact the quality of completion of SC tasks. Based on a conceptual model and generic framework of SC task allocation, this paper first gives a review of the current state of research in this field, including single task allocation, multiple task allocation, low-cost task allocation, and quality-enhanced task allocation. We further investigate the future trends and open issues of SC task allocation, including skill-based task allocation, group recommendation and collaboration, task composition and decomposition, and privacypreserving task allocation. Finally, we discuss the practical issues on real-world deployment as well as the challenges for large-scale user study in SC task allocation.

Mobile crowd sensing (MCS) is the special case of crowdsourcing, which leverages the smartphones with various embedded sensors and user's mobility to sense diverse phenomenon in a city. Task allocation is a fundamental research issue in MCS, which is crucial for the efficiency and effectiveness of MCS applications. In this paper, we specifically focus on the task allocation in MCS systems. We first present the unique features of MCS allocation compared to generic crowdsourcing, and then provide a comprehensive review for diversifying problem formulation and allocation algorithms together with future research opportunities.

Data quality and budget are two primary concerns in urban-scale mobile crowdsensing. Traditional research on mobile crowdsensing mainly takes sensing coverage ratio as the data quality metric rather than the overall sensed data error in the target-sensing area. In this article, we propose to leverage spatiotemporal correlations among the sensed data in the target-sensing area to significantly reduce the number of sensing task assignments. In particular, we exploit both intradata correlations within the same type of sensed data and interdata correlations among different types of sensed data in the sensing task. We propose a novel crowdsensing task allocation framework called SPACE-TA (SPArse Cost-Effective Task Allocation) , combining compressive sensing, statistical analysis, active learning, and transfer learning, to dynamically select a small set of subareas for sensing in each timeslot (cycle), while inferring the data of unsensed subareas under a probabilistic data quality guarantee. Evaluations on real-life temperature, humidity, air quality, and traffic monitoring datasets verify the effectiveness of SPACE-TA. In the temperature-monitoring task leveraging intradata correlations, SPACE-TA requires data from only 15.5% of the subareas while keeping the inference error below 0.25°C in 95% of the cycles, reducing the number of sensed subareas by 18.0% to 26.5% compared to baselines. When multiple tasks run simultaneously, for example, for temperature and humidity monitoring, SPACE-TA can further reduce ∼10% of the sensed subareas by exploiting interdata correlations.

In traditional mobile crowdsensing applications, organizers need participants' precise locations for optimal task allocation, e.g., minimizing selected workers' travel distance to task locations. However, the exposure of their locations raises privacy concerns. Especially for those who are not eventually selected for any task, their location privacy is sacrificed in vain. Hence, in this paper, we propose a location privacy-preserving task allocation framework with geo-obfuscation to protect users' locations during task assignments. Specifically, we make participants obfuscate their reported locations under the guarantee of differential privacy, which can provide privacy protection regardless of adversaries' prior knowledge and without the involvement of any third-part entity. In order to achieve optimal task allocation with such differential geo-obfuscation, we formulate a mixed-integer non-linear programming problem to minimize the expected travel distance of the selected workers under the constraint of differential privacy. Evaluation results on both simulation and real-world user mobility traces show the effectiveness of our proposed framework. Particularly, our framework outperforms Laplace obfuscation, a state-of-the-art differential geo-obfuscation mechanism, by achieving 45% less average travel distance on the real-world data.

Pedestrians distracted by smartphones are easy to meet with various dangers when crossing or walking on the street, such as the obstacles on the sidewalk (e.g., temporary parking and road repairing). Existing works about pedestrian safety are mostly based on the sensing capabilities from a single device. The surrounding information that can be learned, however, is quite limited or incomplete. Therefore, in many cases the dangers cannot be detected and the pedestrians cannot be alerted. In this paper, a novel system called CrowdWatch is proposed, which leverages mobile crowd sensing and crowd intelligence aggregation to detect temporary obstacles and make effective alerts for distracted walkers. To detect obstacles, we first study the regular rules of pedestrians' avoidance behaviors from the aspects of turn-making and visual contexts. The Dempster-Shafer evidence theory is then used to fuse the behavior and visual contexts, and further calculate the confidence of obstacle existence. Afterwards, we leverage the features of pedestrians' traces to characterize an appropriate dangerous area, which is used to alert distracted walkers. The conducted experiments with 36 participants and different obstacle settings indicate that the crowd-intelligencebased obstacle detection method is effective and the accuracy of reminding attains 83.3%.

Despite the great demand on and attempts at package express shipping services, online retailers have not yet had a practical solution to make such services profitable. In this paper, we propose an economical approach to express package delivery, i.e., exploiting relays of taxis with passengers to help transport package collectively, without degrading the quality of passenger services. Specifically, we propose a two-phase framework called crowddeliver for the package delivery path planning. In the first phase, we mine the historical taxi trajectory data offline to identify the shortest package delivery paths with estimated travel time given any Origin-Destination pairs. Using the paths and travel time as the reference, in the second phase we develop an online adaptive taxi scheduling algorithm to find the near-optimal delivery paths iteratively upon real-time requests and direct the package routing accordingly. Finally, we evaluate the two-phase framework using the real-world data sets, which consist of a point of interest, a road network, and the large-scale trajectory data, respectively, that are generated by 7614 taxis in a month in the city of Hangzhou, China. Results show that over 85% of packages can be delivered within 8 hours, with around 4.2 relays of taxis on average.

In mobile crowdsensing (MCS), one of the participants' main concerns is the cost for 3G data usage, which affects their willingness to participate in a crowdsensing task. In this paper, we present the design and implementation of an MCS data uploading mechanism-ecoSense-to help reduce additional 3G data cost incurred by the whole crowd of sensing participants. By considering the two most common real-life 3G price plans-unlimited data plan (UnDP) and pay as you go (PAYG), ecoSense partitions all the users into two groups corresponding to these two price plans at the beginning of each month, with the objective of minimizing the total refunding budget for all participants. The partitioning is based on predicting users' mobility patterns and sensed data size. The ecoSense mechanism is designed inspired by the observation that during the data uploading cycles, UnDP users could opportunistically relay PAYG users' data to the crowdsensing server without extra 3G cost, provided the two types of users are able to “meet” on a common local cost-free network (e.g., Bluetooth or WiFi direct). We conduct our experiments using both the Massachusetts Institute of Technology reality mining and the Small World In Motion (SWIM) simulation data sets. Evaluation results show that ecoSense could reduce total 3G data cost by up to ~50%, when compared to the direct-assignment method that assigns each participant to UnDP or PAYG directly according to the size of her sensed data.

Bike sharing is booming globally as a green transportation mode, but the occurrence of over-demand stations that have no bikes or docks available greatly affects user experiences. Directly predicting individual over-demand stations to carry out preventive measures is difficult, since the bike usage pattern of a station is highly dynamic and context dependent. In addition, the fact that bike usage pattern is affected not only by common contextual factors (e.g., time and weather) but also by opportunistic contextual factors (e.g., social and traffic events) poses a great challenge. To address these issues, we propose a dynamic cluster-based framework for over-demand prediction. Depending on the context, we construct a weighted correlation network to model the relationship among bike stations, and dynamically group neighboring stations with similar bike usage patterns into clusters. We then adopt Monte Carlo simulation to predict the over-demand probability of each cluster. Evaluation results using real-world data from New York City and Washington, D.C. show that our framework accurately predicts over-demand clusters and outperforms the baseline methods significantly.

Sparse Mobile Crowdsensing (MCS) has become a compelling approach to acquire and make inference on urban-scale sensing data. However, participants risk their location privacy when reporting data with their actual sensing positions. To address this issue, we adopt e-differential-privacy in Sparse MCS to provide a theoretical guarantee for participants' location privacy regardless of an adversary's prior knowledge. Furthermore, to reduce the data quality loss caused by differential location obfuscation, we propose a privacypreserving framework with three components. First, we learn a data adjustment function to fit the original sensing data to the obfuscated location. Second, we apply a linear program to select an optimal location obfuscation function, which aims to minimize the uncertainty in data adjustment. We also propose a fast approximated variant. Third, we propose an uncertaintyaware inference algorithm to improve the inference accuracy of obfuscated data. Evaluations with real environment and traffic datasets show that our optimal method reduces the data quality loss by up to 42% compared to existing differential privacy methods.

For participatory sensing, task allocation is a crucial research problem that embodies a tradeoff between sensing quality and cost. An organizer usually publishes and manages multiple tasks utilizing one shared budget. Allocating multiple tasks to participants, with the objective of maximizing the overall data quality under the shared budget constraint, is an emerging and important research problem. We propose a fine-grained multitask allocation framework (MTPS), which assigns a subset of tasks to each participant in each cycle. Specifically, considering the user burden of switching among varying sensing tasks, MTPS operates on an attention-compensated incentive model where, in addition to the incentive paid for each specific sensing task, an extra compensation is paid to each participant if s/he is assigned with more than one task type. Additionally, based on the prediction of the participants' mobility pattern, MTPS adopts an iterative greedy process to achieve a near-optimal allocation solution. Extensive evaluation based on real-world mobility data shows that our approach outperforms the baseline methods, and theoretical analysis proves that it has a good approximation bound.

Container ports are generally measured and compared using performance indicators such as container throughput and facility productivity. Being able to measure the performance of container ports quantitatively is of great importance for researchers to design models for port operation and container logistics. Instead of relying on the manually collected statistical information from different port authorities and shipping companies, we propose to leverage the pervasive ship GPS traces and maritime open data to derive port performance indicators, including ship traffic, container throughput, berth utilization, and terminal productivity. These performance indicators are found to be directly related to the number of container ships arriving at the terminals and the number of containers handled at each ship. Therefore, we propose a framework that takes the ships' container-handling events at terminals as the basis for port performance measurement. With the inferred port performance indicators, we further compare the strengths and weaknesses of different container ports at the terminal level, port level, and region level, which can potentially benefit terminal productivity improvement, liner schedule optimization, and regional economic development planning. In order to evaluate the proposed framework, we conduct extensive studies on large-scale real-world GPS traces of container ships collected from major container ports worldwide through the year, as well as various maritime open data sources concerning ships and ports. Evaluation results confirm that the proposed framework not only can accurately estimate various port performance indicators but also effectively produces port comparison results such as port performance ranking and port region comparison.

This paper first defines a novel spatial-temporal coverage metric, k-depth coverage, for mobile crowdsensing (MCS) problems. This metric considers both the fraction of subareas covered by sensor readings and the number of sensor readings collected in each covered subarea. Then iCrowd, a generic MCS task allocation framework operating with the energy-efficient Piggyback Crowdsensing task model, is proposed to optimize the MCS task allocation with different incentives and k-depth coverage objectives/ constraints. iCrowd first predicts the call and mobility of mobile users based on their historical records, then it selects a set of users in each sensing cycle for sensing task participation, so that the resulting solution achieves two dual optimal MCS data collection goals-i.e., Goal. 1 near-maximal k-depth coverage without exceeding a given incentive budget or Goal. 2 near-minimal incentive payment while meeting a predefined k-depth coverage goal. We evaluated iCrowd extensively using a large-scale real-world dataset for these two data collection goals. The results show that: for Goal.1, iCrowd significantly outperformed three baseline approaches by achieving 3-60 percent higher k-depth coverage; for Goal.2, iCrowd required 10.0-73.5 percent less incentives compared to three baselines under the same k-depth coverage constraint.

Data quality and budget are two primary concerns in urban-scale mobile crowdsensing applications. In this paper, we leverage the spatial and temporal correlation among the data sensed in different sub-areas to significantly reduce the required number of sensing tasks allocated (corresponding to budget), yet ensuring the data quality. Specifically, we propose a novel framework called CCS-TA, combining the state-of-the-art compressive sensing, Bayesian inference, and active learning techniques, to dynamically select a minimum number of sub-areas for sensing task allocation in each sensing cycle, while deducing the missing data of unallocated sub-areas under a probabilistic data accuracy guarantee. Evaluations on real-life temperature and air quality monitoring datasets show the effectiveness of CCS-TA. In the case of temperature monitoring, CCS-TA allocates 18.0-26.5% fewer tasks than baseline approaches, allocating tasks to only 15.5% of the sub-areas on average while keeping overall sensing error below 0.25°C in 95% of the cycles.

Link prediction for new users who have not created any link is a fundamental problem in Online Social Networks (OSNs). It can be used to recommend friends for new users to start building their social networks. The existing studies use cross-platform approaches to predict a new user's links on a certain OSN by porting his existing links from other OSNs. However, it cannot work when OSNs are not willing to share their data or users do not want to connect different OSN accounts. In this paper, we use a single-platform approach to carry out the link prediction. We explore the users' profile attributes (e.g., workplace, high school and hometown) which can be easily obtained during the new users' sign up procedure. Based on the limited available information from the new user, along with the attributes and links from existing users, we extract three types of social features: basic feature, derived feature and latent relation feature. We propose a link prediction model using these social features based on Support Vector Machines. Eventually, we rely on a large Facebook data set consisting of 479,000 users to evaluate our proposed model. The result reveals that our model outperforms the baselines by achieving the AUC value of 0.83; it also demonstrates that each of the proposed social features contribute significantly to the prediction model.

This paper proposes a novel task allocation framework, CrowdTasker, for mobile crowdsensing. CrowdTasker operates on top of energy-efficient Piggyback Crowdsensing (PCS) task model, and aims to maximize the coverage quality of the sensing task while satisfying the incentive budget constraint. In order to achieve this goal, CrowdTasker first predicts the call and mobility of mobile users based on their historical records. With a flexible incentive model and the prediction results, CrowdTasker then selects a set of users in each sensing cycle for PCS task participation, so that the resulting solution achieves near-maximal coverage quality without exceeding incentive budget. We evaluated CrowdTasker extensively using a large-scale real-world dataset and the results show that CrowdTasker significantly outperformed three baseline approaches by achieving 3%-60% higher coverage quality.

Mobile Crowdsensing (MCS) requires users to be motivated to participate. However, concerns regarding energy consumption and privacy—among other things—may compromise their willingness to join such a crowd. Our preliminary observations and analysis of common MCS applications have shown that the data transfer in MCS applications may incur significant energy consumption due to the 3G connection setup. However, if data are transferred in parallel with a traditional phone call, then such transfer can be done almost “for free”: with only an insignificant additional amount of energy required to piggy-back the data—usually incoming task assignments and outgoing sensor results—on top of the call. Here, we present an Energy-Efficient Mobile Crowdsensing (EEMC) framework where task assignments and sensing results are transferred in parallel with phone calls. The main objective, and the principal contribution of this article, is an MCS task assignment scheme that guarantees that a minimum number of anonymous participants return sensor results within a specified time frame, while also minimizing the waste of energy due to redundant task assignments and considering privacy concerns of participants. Evaluations with a large-scale real-world phone call dataset show that our proposed EEMC framework outperforms the baseline approaches, and it can reduce overall energy consumption in data transfer by 54--66% when compared to the 3G-based solution.

This paper proposes a novel mobile crowdsensing (MCS) framework called EMC3, which intends to reduce energy consumption of individual user as well as all participants in data transfer caused by task assignment and data collection of MCS tasks, considering the user privacy issue, minimal number of task assignment requirement and sensing area coverage constraint. Specifically, EMC3incorporates novel pace control and decision making mechanisms for task assignment, leveraging participants' current call, historical call records as well as predicted future calls and mobility, in order to ensure the expected number of participants to return sensed results and fully cover the target area, with the objective of assigning a minimal number of tasks. Extensive evaluation with a large-scale real-world dataset shows that EMC3assigns much less sensing tasks compared to baseline approaches, it can save 43%-68% energy in data transfer compared to the traditional 3G-based scheme.

Energy consumption and mobile data cost are two key factors affecting users’ willingness to participate in mobile crowd-sensing tasks. While data-plan (DP) users are mostly concerned with energy consumption, non-data-plan (NDP) users are more sensitive to data cost. Traditional ways of data uploading in mobile crowdsensing tasks often go to two extremes: either in real time or completely offline after the whole task is over. In this paper, we propose effSense—an energy-efficient and cost-effective data uploading framework, which utilizes adaptive uploading schemes within fixed data uploading cycles. In each cycle, effSense empowers the participants with a distributed decision making scheme to choose the appropriate timing and network to upload data. effSense reduces data cost for NDP users by maximally offloading data to Bluetooth/WiFi gateways or DP users encountered; it reduces energy consumption for DP users by piggybacking data on a call or using more energy-efficient networks rather than initiating new 3G connections. By leveraging the predictability of users’ calls and mobility, effSense selects proper uploading strategies for both user types. Our evaluation with the MIT reality mining and Nodobo datasets shows that effSense can reduce 55%–65% energy consumption for DP users, and 48%–52% data cost for NDP users, respectively, compared to traditional uploading schemes.

This paper presents a comprehensive empirical study on the correlations between users' interest similarity and various social features across three interest domains (i.e., movie, music and TV). This study relies on a large dataset, containing 479, 048 users and 5, 263, 351 user-generated interests, captured from Facebook. We identify the social features from three types of the users' information - demographic information (e.g., age, gender, location), social relations (i.e., friendship), and users' interests. The results reveal that the interest similarity follows the homophily principle. Particularly, the results show that two users are more likely to be alike in their interests 1) if they exhibit more similarity in their demographic characteristics (e.g., similar age, same gender, or close to each other geographically), or 2) if they are more intimate in their friendship, or 3) if they present a higher average interest individuality (i.e., a measurement for estimating the personalized characteristics of a user's interests). The empirical observations could be exploited to infer how two users are alike in their interests according to the social features, which could be further harnessed by various practical applications and services, such as recommendation system and advertisement service.

Traditionally, the port container throughput, a crucial measurement of regional economic development, was manually collected by port authorities. This requires a large amount of human effort and often delays publication of this important figure. In this paper, by leveraging ubiquitous positioning techniques and open data, we propose a two-phase approach to estimation of port container throughput in real-time. First, we obtain the number of container ships arriving at berth by analyzing the ships' GPS traces. Then we estimate the throughput of each ship, in terms of number of containers transshipped, by considering the ship's berthing time, capacity, length, breadth, and crane operation performance, as extracted from different data sources. Evaluation results using real-world datasets from Hong Kong and Singapore show that the proposed approach not only estimates the container throughput quite accurately, but also outperforms the baseline method significantly.

This paper proposes a novel participant selection framework, named CrowdRecruiter, for mobile crowdsensing. CrowdRecruiter operates on top of energy-efficient Piggyback Crowdsensing (PCS) task model and minimizes incentive payments by selecting a small number of participants while still satisfying probabilistic coverage constraint. In order to achieve the objective when piggybacking crowdsensing tasks with phone calls, CrowdRecruiter first predicts the call and coverage probability of each mobile user based on historical records. It then efficiently computes the joint coverage probability of multiple users as a combined set and selects the near-minimal set of participants, which meets coverage ratio requirement in each sensing cycle of the PCS task. We evaluated CrowdRecruiter extensively using a large-scale real-world dataset and the results show that the proposed solution significantly outperforms three baseline algorithms by selecting 10.0% -- 73.5% fewer participants on average under the same probabilistic coverage constraint.

Programmers often learn how to use an API by studying its usage examples. There are many usage examples scattered in web pages on the Internet. However, it often takes programmers much effort to find out the desired examples from a large number of web pages by web search. This paper proposes a tool named APIExample that can extract usage examples for java APIs from web pages on the Internet and recommend them to programmers. Given a java API, the tool collects its related web pages from the Internet, extracts java code snippets and their surrounding descriptive texts embedded in the pages, then assembles them into usage examples for programmers. Furthermore, in order to help programmers capture more kinds of usages of the target API by browsing fewer examples, our tool clusters and ranks the listed examples based on the target API's usage. Besides, as a practical tool, APIExample provides multiple aspects of frequently-used information about using the target API in a concise user interface with friendly user experience. Two kinds of user-interaction style, a web search portal and an Eclipse plug-in, are now both publicly available.