Verifying the claimed sale-ranking trustworthy: A maximum marginal relevance-based ranking method

摘要

Various online contents on Internet platforms or search engines are related to the corporatereputation. Facing the huge amount of online contents, we need a mining method that canautomatically extract and analyze a large number of network-related information and obtainthe real reliability of aspect for the content claimed by companies. In this paper, we proposeto generate a ranking model to verify whether the sales-rankings claimed by companies aretrustworthy. The key idea is that the company that has higher confidence score should besupported by the online media. We use a unique data set of public opinion data related witha specific company, which we supplement with data from various online news platform andretrieval webpages using a distributed and generic Web crawler. Meanwhile, basic informationand open financial data of companies are also collected for auxiliary analysis. We present aMaximal Marginal Relevance-based ranking model to compute the confidence score of eachcompany, taking into consideration the two technologies of word embedding and KL-Divergenceto filter the irrelevant documents. Extensive experiments show that the proposed methodoutperforms the state-of-the-art MMR-based method, and we showcase three representativecases about the corporate reputation built by us that gives positive, neutral, and negativesupport respectively to the sales-ranking claim of companies.