BTD: An effective business-related hot topic detection scheme in professional social networks

Abstract

Professional social networks (PSNs) usually involve a large amount of valuable information for the business world. A heterogeneous network is constructed based on the structural characteristics of several communities from a PSN. Then, an effective business-related hot topic detection (BTD) scheme in PSNs is proposed, and this BTD scheme extracts business-related topics from posts found on the PSN. Furthermore, a business-related hot topic detection algorithm is proposed by extending the PageRank algorithm based on the heterogeneous network. The performance of the proposed method is evaluated by using real data from a PSN for B2B e-commerce. The experimental results show that the proposed method is able to detect business-related hot topics in heterogeneous networks from three aspects: affiliation relationships between posts and topics, users’ contributions to posts, and following relationships among users. The coverage rate is higher and the degree of distinction is greater than those of existing typical methods.