Research on the Method of Identifying Opinion Leaders Based on Online Word-of-Mouth

Opinion leaders are attracting increasing attention on practitioners and academics. Opinion leaders’ online Word-of-Mouth (WOM) plays a guiding and decisive role in reducing risks and uncertainty faced by users in online shopping. It is of great significance of businesses and enterprises to effectively identify opinion leaders. This study proposes an integrated method by looking at not only essential indicators of reviewers but also the review characteristics. The RFM model is used to evaluate the activity of reviewers. Four variables L (text length), T (period time), P (with or without a picture) and S (sentiment intensity) are derived to measure review helpfulness from review text. And two effective networks are built using the Artificial Neural Network (ANN). This study utilizes a real-life data set from Dianping.com for analysis and designs three different experiments to verify the identification effect. The results show that this method can scientifically and effectively identify the opinion leaders and analyze the influence of opinion leaders.