SLFTD: A Subjective Logic Based Framework for Truth Discovery

Abstract

Finding truth from various conflicting candidate values provided by different data sources is called truth discovery, which is of vital importance in data integration. Several algorithms have been proposed in this area, which usually have similar procedure: iterativly inferring the truth and provider's reliability on providing truth until converge. Therefore, an accurate provider's reliability evaluation is essential. However, no work pays attention to "how reliable this provider continuously providing truth". Therefore, we introduce subjective logic, which can records both (1) the provider's reliability of generating truth, and (2) reliability of provider continuously doing so. Our proposed methods provides a better evaluation for data providers, and based which, truth are discovered more accurately. Our framework can handle both categorical and numerical data, and can identify truth in either a generative or discriminative way. Experiments on two popular real world datasets, Book and Population, validates that our proposed subjective logic based framework can discover truth much more accurately than state-of-art methods.

Fake News Detection Based on Subjective Opinions

“…Before this, it is widely used in many other areas, including trust network analysis [31], conditional inference [32], information provider reliability assessment [33], trust management in sensor networks [34]. In one of our earlier works [35], Subjective Opinions is introduced for Truth Discovery. With the data represented and recorded more comprehensively, our proposed Subjective Opinions based method is expected to have a better performance than the probability based method.…”

FaNDS: Fake News Detection System Using Energy Flow

“…For example, for a peer A, it has an initial energy, which propagates to the whole network, and the energy collected by another peer B is the trust that peer A has to peer B. Another flow model different from the above methods is SLFTD from [63], where a bigraph of data providers and entities is constructed, and trust on the data provider and the discrimination ability of each entity is iteratively updated as in HITS.…”