Abstract—Graph Pattern Matching (GPM) plays a significant role in social network analysis, which has been widely used in, for example, experts finding, social community mining and social position detection. Given a pattern graph GQ and a data graph GD, a GPM algorithm finds those subgraphs, GM, that match GQ in GD. However, the existing GPM methods do not consider the multiple constraints on edges in GQ, which are commonly exist in various applications such as, crowdsourcing travel, social network based e-commerce and study group selection, etc. In this paper, we first conceptually extend Bounded Simulation to Multi-Constrained Simulation (MCS), and propose a novel NP-Complete Multi-Constrained Graph Pattern Matching (MC-GPM) problem. Then, to address the efficiency issue in large-scale MC-GPM, we propose a new concept called Strong Social Component (SSC), consisting of participants with strong social connections. We also propose an approach to identify SSCs, and propose a novel index method and a graph compression method for SSC. Moreover, we devise a heuristic algorithm to identify MC-GPM results effectively and efficiently without decompress-ing graphs. An extensive empirical study on five real-world large-scale social graphs has demonstrated the effectiveness, efficiency and scalability of our approach. I.

Abstract—Online Social Networks (OSNs) have become an integral part of daily life in recent years. OSNs contain impor-tant participants, the trust relations between participants, and the contexts in which participants interact with each other. All of these have a great influence on the prediction of the trust between a source participant and a target participant, which is important for a participant’s decision-making process in many applications, such as seeking service providers. However, predicting the trust from a source participant to a target one based on the whole social network is not really feasible. Thus, prior to trust prediction, the extraction of a small-scale sub-network containing most of the important nodes and contextual information with a high density rate could make trust predic-tion more efficient and effective. However, extracting such a sub-network has been proved to be an NP-Complete problem. To address this challenging problem, we propose BiNet: a social context-aware trust sub-network extraction model to search for near-optimal solutions effectively and efficiently. In this model, we first capture important factors that affect the trust between participants in OSNs. Next, we define a utility function to measure the trust factors of each node in a social network. At last, we design a novel binary ant colony algorithm with newly designed initialization and mutation processes for sub-network extraction incorporating the utility function. The experiments, conducted on two popular datasets of Epinion and Slashdot, demonstrate that our approach can extract sub-networks covering important participants and contextual information while keeping a high density rate. Our approach is superior to the state-of-the-art approaches in terms of the quality of extracted sub-networks within the same execution time. Keywords-Trust; Sub-network extraction; Trust prediction; I.

Abstract — On a crowdsourcing platform consisting of task publishers and workers, it is critical for a task publisher to select trustworthy workers to solve human intelligence tasks (HITs). Currently, the prevalent trust evaluation mechanism employs the overall approval rate of HITs, with which dishonest workers can easily succeed in pursuing the maximal profit by quickly giving plausible answers or counterfeiting HITs approval rates. In crowdsourcing environments, a worker’s trustworthiness varies in contexts, i.e. it varies in different types of tasks and different reward amounts of tasks. Thus, we propose two classifications based on task types and task reward amount re-spectively. On the basis of the classifications, we propose a trust evaluation model, which consists of two types of context-aware trust: task type based trust (TaTrust) and reward amount based trust (RaTrust). Then, we model trustworthy worker selection as a multi-objective combinatorial optimization problem, which is NP-hard. For solving this challenging problem, we propose an evolutionary algorithm MOWS GA based on NSGA-II. The results of experiments illustrate that our proposed trust evaluation model can effectively differentiate honest workers and dishonest workers when both of them have high overall HITs approval rates.

Abstract In recent years, Online Social Networks (OSNs) with numerous partici-pants have been used as the means for rich activities. For example, employers could use OSNs to investigate potential employees, and participants could use OSNs to look for movie recommendations. In these activities, trust is one of the most impor-tant indication of participants decision making, greatly demanding the evaluation of the trustworthiness of a service provider along certain social trust paths from a service consumer. In this chapter, we first analyze the characteristics of the current generation of functional websites and the current generation of online social net-works based on their functionality and sociality, and present the properties of the new generation of social network based web applications. Then we present a new selection model considering both adjacent and end-to-end constraints, based on a novel concept Quality of Trust and a novel complex social network structure. More-over, in order to select the optimal one from massive social trust paths yielding the most trustworthy trust evaluation result, this chapter presents an effective and ef-ficient heuristic algorithm for optimal social trust path selection with constraints, which is actually an NP-Complete problem. Experimental results illustrate that the proposed method outperforms existing models in both efficiency and the quality of delivered solutions. This work provides key techniques to potentially lots of service-oriented applications with social networks as the backbone.