Abstract: Due to the rapid growth of Internet, consumers obtain information about products and experiences from online rating platforms, which further influences their purchase decisions. The popularity, individuation, diversification, and heterogeneity of consumers contribute to the diversity of online review. This study examines the evolution process of opinions on the online review platform. Specifically, the network structure and confidence thresholds of traditional models are optimized based on the ground truth of Dianping via the combination of simulation and actual data. The proposed method can deepen the understanding for the evolution process of online opinion.