With the widespread adoption of smartphones and other mobile intelligent devices, Mobile Crowd Sensing (MCS) is widely used. Typically, the real locations of the workers and tasks must be submitted to the service platform to complete the task allocation. Therefore, the protection of location information has become a key factor in influencing user participation. To address the issue of location information leakage, we propose a location information protection method based on local differential privacy, which can protect the location privacy of workers and tasks while generating approximately accurate task allocation results. Firstly, we divide the region into$k$*$k$grids and merge girds with a similar dispersion to form clusters. Then, this paper utilizes the inverse sampling of the cumulative distribution function (CDF) of the flipped Huber distribution to generate a personalized noise location set for each cluster. Furthermore, the exponential mechanism is used to select the obfuscated location for each user. Finally, the platform selects workers based on the perturbed location to complete the task allocation. Theoretical analysis shows that our mechanism satisfies differential privacy and achieves an approximately accurate task allocation. Experimental results demonstrate that, compared to existing methods, this method exhibits superior performance across different datasets and effectively balances the utility of data and the protection of location privacy.

Auctions have been widely applied in resource allocation due to their fairness and efficiency. For instance, platforms receive requests from service requesters and utilize auction theory to select suitable service providers. Existing studies typically assume that winners are determined based on bidders’ true valuations by allowing arbitrary transactions between requesters and providers, which can lead to serious valuation privacy leakage issues and limitations in application scenarios. Although some research has addressed these concerns using differential privacy techniques, they mostly rely on a trusted platform, and the introduction of noise results in utility loss, making them unsuitable for restricted auction contexts. To overcome these limitations, we propose a restricted double auction mechanism based on local differential privacy. Specifically, we extract the characteristics of the valuation data and constrain the noise addition probability density function based on the data features. Then we design a novel exponential selection mechanism that ensures that the relative positions of the obfuscated bids remain unchanged compared to the original valuations, while satisfying ε-local differential privacy. Furthermore, we develop an auction matching mechanism that maintains properties such as truthfulness under restricted allocation. The simulation results demonstrate that the proposed bid obfuscation mechanism ensures that the relative positions of the interfered bids remain unchanged while incurring low time overhead. Compared to existing mechanisms, our restrictive auction mechanism can generate greater social welfare while reducing the risk of valuation privacy leakage.

The proliferation of homogeneous web services has necessitated the task of predicting vacant Quality of Service (QoS) for service-oriented downstream tasks. Existing approaches primarily focus on user-service invocations without considering temporal factors, limiting their applicability in QoS fluctuations over time. Moreover, some investigations are conducted to predict temporally missing QoS, which still suffers from two limitations. First, time-aware collaborative filtering (CF) approaches fail to well capture continuous temporal changes, which lowers the performance of time-aware QoS prediction. Second, they have paid less attention to the high sparsity of user-service QoS invocations across sequentially multiple time slices, which affects the calculation of temporal average QoS, thereby further reducing the accuracy of time-aware QoS prediction. To effectively mine the continuous temporal variations and solve the high sparsity of user-service QoS invocations, we propose a novel time-aware QoS prediction approach named Temporal Reinforced Collaborative Filtering (TRCF). We design temporal reinforced RBS and PCC to improve similarity evaluation that leads to better calculation of temporal average QoS and deviation migration for predicting time-aware QoS. We evaluate TRCF on a large-scale real-world temporal dataset WS-DREAM across 64 time slices and the results demonstrate its superior performance in time-aware QoS prediction, both under relatively dense and extremely sparse QoS situations.

Over the past years, Human Activity Recognition (HAR) has shown its great value and has been further developed with the help of deep learning. However, existing HAR systems that use deep learning methods to achieve the ideal accuracy of recognition heavily rely on massive amounts of labeled training samples. Unfortunately, it requires considerable human effort and is unrealistic for real-life applications. In this paper, we propose a novel system, which combines active learning with WiFi-based HAR. The system is capable of building a good activities recognizer in HAR with a limited amount of labeled training samples. We thus call the system ALSensing. To the best of our knowledge, ALSensing is the first system to apply active learning to WiFi-based HAR. We implement ALSensing using commercial WiFi devices and evaluated it with realistic data in several different environments. Our experimental results show that ALSensing achieves 52.83% recognition accuracy using 3.7% training samples, 58.97% recognition accuracy using 15% training samples and the baseline predicted with the existing method achieves 62.19% recognition accuracy using 100% training samples. When the performance of ALSensing is similar to that of the baseline, the required labeled samples are much less than that of the baseline.

Events spreading on social media platforms reflect current public concerns and emotions among public opinions. Heterogeneous elements of social networks and the sparse context of social messages bring significant challenges to the fine-grained social event detection task. Few existing methods can learn the inherent structure and rich semantics among social messages, nor can they effectively update the detection model in a dynamic scenario for continuously coming messages. In this paper, we design a novel Multi-Semantics Heterogeneous Graph Neural Network (MSGNN) to learn social events in a continuous detection framework. We apply the heterogeneous information network (HIN) to modeling social events, considering the heterogeneous elements and meta-paths in the social event data stream. We propose a dual-level messages aggregation mechanism to aggregate semantics between heterogeneous elements, which aggregates the local features of adjacent neighboring messages from the node level and the global semantics from the meta-path level to the current message. A semantic weight is designed for messages to filter out noise under social message streams. We conduct extensive experiments on two real-world social event datasets, and the experimental results demonstrate that our proposed model outperforms state-of-the-art models.

Non-IID data present a tough challenge for federated learning. In this paper, we explore a novel idea of facilitating pairwise collaborations between clients with similar data. We propose FedAMP, a new method employing federated attentive message passing to facilitate similar clients to collaborate more. We establish the convergence of FedAMP for both convex and non-convex models, and propose a heuristic method to further improve the performance of FedAMP when clients adopt deep neural networks as personalized models. Our extensive experiments on benchmark data sets demonstrate the superior performance of the proposed methods.

Deep learning has been applied in many recent advanced applications in the field of transportation, finance and medicine. These applications require significant computation resources and large-scale training samples. Cloud becomes a natural choice for conducting these learning tasks due to its abundant resources. However, deeper penetration of deep learning techniques in mission critical applications, like driverless car, calls for stricter time requirement to guarantee its interaction and larger amount of dataset for training to guarantee its accuracy, which cannot be easily satisfied by the cloud and makes the network transmission become the bottleneck. Edge learning emerges to be a promising direction to reduce data transmission time by processing and compressing the raw data at the edge of the network, while brings the concern of accuracy reduction at the meantime. To balance this tradeoff under cloud-edge architecture, we study a task scheduling problem for reducing weighted transmission time which takes learning accuracy into consideration. We also propose efficient scheduling algorithms which are able to achieve up to 50% reduction in makespan with extensive trace-driven simulations.

The state-of-the-art cloud computing platforms are facing challenges, such as the high volume of crowdsourced data traffic and highly computational demands, involved in typical deep learning applications. More recently, Edge Computing has been recently proposed as an effective way to reduce the resource consumption. In this paper, we propose an edge learning framework by introducing the concept of edge computing and demonstrate the superiority of our framework on reducing the network traffic and running time.