Heterogeneous network-based trust analysis

Abstract

Different information sources publish information with different degrees of correctness and originality. False information can often result in considerable damage. Hence, trustworthinessof information is an important issue in this datadriven world economy. Reputation of different agents in a network has been studied earlier in a variety of domains like e-commerce, social sciences, sensor networks, and P2P networks. Recently there has been work in the data mining community on performing trust analysis based on the data provided by multiple information providers for different objects, and such agents and their provided information about data objects form a multi-typed heterogeneous network. The trust analysis under such a framework is considered as heterogeneous network-based trust analysis. This paper will survey heterogeneous network-based trust analysis models and their applications. We would conclude with a summary and some thoughts on future research in the area.