Task assignment is a key issue in mobile crowdsensing (MCS). Most existing work in this area has focused on the task assignment for the single platform scenario, which can cause considerable waste of limited human resources or reduced task completion rate due to potential spatial mismatching between distributions of users and tasks. In this article, we study multiplatform cooperative task assignment. The design goal is to maximize the social welfare while ensuring cooperative and individual rationality. We formulate this problem, transform it to a maximum value flow problem, and prove its NP-hardness. We first propose a greedy-maximum-flow-based task matching (GMTA) mechanism for interplatform task matching. In GMTA, there are two phases in each time slot: 1) in the former phase, earliest-deadline-first-based intraplatform optimal task assignment is carried out at each individual platform and 2) in the second phase, greedy-maximum-flow-based task matching is carried out to perform interplatform cooperative task assignment for those overloaded tasks in the first phase. We then enhance GMTA by designing an iterative-maximum-flow-based task matching (IMTA) mechanism, which is to achieve enhanced social welfare at the cost of increased computational overhead. We deduce time complexities of both mechanisms, and prove that they satisfy cooperative and individual rationality. Extensive simulations are conducted and the simulation results demonstrate the effectiveness of our proposed mechanisms.

To improve test data compression efficiency, the order of the scan unit need be adjusted, which indirectly changes the content of the test pattern. Scan chain partitioning is a common method that utilises this concept. However, current scan chain partitioning methods can still be optimised in terms of entropy and test data compression efficiency, and lack universality. To enhance the efficiency of code-based compression efficiency, we propose a scan slice reordering algorithm that minimizes entropy. This algorithm first calculates the rank of each scan slice or column vector of the test set, and then dynamically adjusts the order of the scan slices according to the descending order of these ranks in the pursuit of minimizing the entropy of the test set. By iterating through this process, the optimal position of each scan slice in the test set is ultimately determined. Compression experiments should be performed on all test patterns using different code-based schemes. This not only reduces the entropy of the test set but also significantly improves the efficiency of different codes. Compared to traditional scan chain partitioning method, FDR encoding achieved an average compression ratio increase of 6.16%, and RL-Huffman encoding achieved an average compression ratio increase of 4.96%. The experimental results demonstrate that our proposed algorithm is feasible, effective, and universal.

Mobile crowdsensing (MCS) has been a promising paradigm for gathering sensing data from surrounding environment by leveraging smart devices carried by mobile users and also their subjective initiatives. In this sensing paradigm, mobile users can make full use of such sensors-rich smart devices for task executions. Recently, social mobile crowdsensing (SMCS) has received a lot of attention and much work has been carried out. Many incentive mechanisms exploit the social relations among users/workers for improving the system performance. However, most existing work in this area focused on offline and socially-aware scenarios. In this paper, we study both online socially-aware and socially-unaware scenarios for maximizing the platform utility. We formulate the problem of worker selection for maximizing the platform utility and prove this problem is NP-hard. For the socially-aware scenario, we propose an incentive mechanism (called SA-WGRA), which adopts sociality and capability based clustering algorithm for Worker Group formation and uses Reverse Auction for worker selection. For the socially-unaware scenario, we propose an incentive mechanism (called SUA-CGRA), which adopts Coalitional Game combined with Reversed Auction for worker selection. We prove that both mechanisms achieve computational efficiency, individual rationality, and platform rationality. Moreover, for SUA-CGRA, we prove that its formed coalitions satisfy coalition rationality, and further each of its formed coalitions is convex and hence the Shapley value is in the core solutions for profit distribution in each formed coalition. Simulations results show that both SA-WGRA and SUA-CGRA can effectively improve the platform utility.

Multi-access edge computing (MEC) has been an important and promising paradigm for offering computing services to mobile users with computation-intensive and latency-critical tasks. In this paper, we study a D2D collaboration based MEC system, where the service platform purchases resources from resource-rich collaborative D2D devices when the task arrival rate exceeds the platform’s capability for providing satisfactory QoS. The design objective is to maximize the platform profit while maximally satisfying the delay requirements of tasks. We define delay based utility functions for different participants and accordingly formulate the platform profit maximization problem as a Mixed Integer Non-Linear Programming (MINLP) problem. For the online case where future task arrivals are unknown in advance, we propose a reverse auction based task assignment and urgency-value based transmission scheduling algorithm (RAGM). We present the detailed algorithm design and deduce its computation complexity. We prove that RAGM satisfies individual rationality of all participants. We conduct extensive simulations and the results show the high performance of RAGM as compared with benchmark algorithms.

Opportunistic Mobile Crowdsensing (MCS) is an attractive and cost-effective sensing paradigm because it does not affect workers' daily routines. However, its opportunistic nature in task executions can lead to low task completion rate for deadline-sensitive tasks as compared with participatory sensing. Besides, existing work in opportunistic M CS lacks of study on how to effectively coordinate among multiple service platforms for idle worker sharing so as to improve the sensing performance. In this paper, we design a multi-platform cooperation based incentive mechanism (MPCIM) for deadline-sensitive task assignment in the context of opportunistic mobile crowdsensing. The design objective is to maximize the system profit while improving the task completion rate. In MPCIM, each platform first decides how many idle workers it can provide and also how many tasks it needs assistance at different locations; Then, a cross-platform managing entity is responsible for performing maximal matching between the idle workers and excessive tasks among different platforms to improve the task completion rate while respecting individual rationality and cooperative rationality. Extensive simulation results show that MPCIM can effectively improve the system profit and also task completion rate.

Incentive mechanism has been an important research direction in mobile crowdsensing. An effective incentive mechanism is critical to ensure the adequate number of participants/workers by providing them proper rewards. However, existing incentive mechanisms lack consideration on potential contributions of individual workers when recruiting new workers and retaining existing workers in the system. In this article, we propose a reverse auction-based incentive mechanism (RAIN), which considers participants' potential contributions when recruiting new workers, performing reverse auctions, and retaining existing workers. The design objective is to optimize the worker composition in the system while reducing the system cost. In RAIN, the potential contribution of a user to the system is measured as the degree at which the user's joining or staying in the system can remedy the inadequacy of workers for task auction/execution at the frequently visited locations of the user. We present design details of RAIN which includes selective worker recruitment, reverse auction based on biased bids, and selective retaining of auction losers, all based on individual users' potential contributions to the system. Extensive simulation results show that RAIN can effectively optimize the worker composition in a system and also effectively reduce the system cost.

Incentive mechanism design is a critical issue in mobile crowdsensing and a lot of work has been carried out. However, existing mechanisms in this area generally lack of consideration of individual worker/candidate's (potential) contribution to the system when recruiting new workers or when detaining existing workers. In this paper, we design a reverse auction based incentive mechanism. The design objective is to maximally reduce the system maintenance cost (including auction cost and recruitment cost) by optimizing the composition of workers in the system. For this purpose, in the recruiting process, candidates are queried in the descending order of their potential contributions to the system, while in the detaining process, likelydropping-out workers are rewarded with inner lottery whose amount is adjusted based on their usefulness to the system. In the auction process, prices are calculated based on workers' bids and also their usefulness to the system. We present detailed mechanism design. Simulation results show that our mechanism outperforms existing work.

Distant supervision for relation extraction is an efficient method to scale relation extraction to very large corpora which contains thousands of relations. However, the existing approaches have flaws on selecting valid instances and lack of background knowledge about the entities. In this paper, we propose a sentence-level attention model to select the valid instances, which makes full use of the supervision information from knowledge bases. And we extract entity descriptions from Freebase and Wikipedia pages to supplement background knowledge for our task. The background knowledge not only provides more information for predicting relations, but also brings better entity representations for the attention module. We conduct three experiments on a widely used dataset and the experimental results show that our approach outperforms all the baseline systems significantly.

We model knowledge graphs for their completion by encoding each entity and relation into a numerical space. All previous work including Trans(E, H, R, and D) ignore the heterogeneity (some relations link many entity pairs and others do not) and the imbalance (the number of head entities and that of tail entities in a relation could be different) of knowledge graphs. In this paper, we propose a novel approach TranSparse to deal with the two issues. In TranSparse, transfer matrices are replaced by adaptive sparse matrices, whose sparse degrees are determined by the number of entities (or entity pairs) linked by relations. In experiments, we design structured and unstructured sparse patterns for transfer matrices and analyze their advantages and disadvantages. We evaluate our approach on triplet classification and link prediction tasks. Experimental results show that TranSparse outperforms Trans(E, H, R, and D) significantly, and achieves state-of-the-art performance.

Guoliang Ji, Shizhu He, Liheng Xu, Kang Liu, Jun Zhao. Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2015.

The representation of a knowledge graph (KG) in a latent space recently has attracted more and more attention. To this end, some proposed models (e.g., TransE) embed entities and relations of a KG into a "point" vector space by optimizing a global loss function which ensures the scores of positive triplets are higher than negative ones. We notice that these models always regard all entities and relations in a same manner and ignore their (un)certainties. In fact, different entities and relations may contain different certainties, which makes identical certainty insufficient for modeling. Therefore, this paper switches to density-based embedding and propose KG2E for explicitly modeling the certainty of entities and relations, which learn the representations of KGs in the space of multi-dimensional Gaussian distributions. Each entity/relation is represented by a Gaussian distribution, where the mean denotes its position and the covariance (currently with diagonal covariance) can properly represent its certainty. In addition, compared with the symmetric measures used in point-based methods, we employ the KL-divergence for scoring triplets, which is a natural asymmetry function for effectively modeling multiple types of relations. We have conducted extensive experiments on link prediction and triplet classification with multiple benchmark datasets (WordNet and Freebase). Our experimental results demonstrate that our method can effectively model the (un)certainties of entities and relations in a KG, and it significantly outperforms state-of-the-art methods (including TransH and TransR).