ICA k-means based time series clustering analysis of online word-of-mouth

Online word-of-mouth activity is a very typical index of the lifecycle evolution model of a product, and understanding product lifecycle can help corresponding decision makers with their formulation of marketing strategies. In this paper, the data sets for the online comments on various types of products are studied; based on management theory and economics theory, and by applying such methods as independent component analysis (ICA) and clustering, the online word-of-mouth activities for different products or similar products are analyzed through clustering; meanwhile, the lifecycle curves for some representative products are extracted, and typical lifecycles are profoundly analyzed. This paper aims to effectively improve the effect of online word-of-mouth information on e-commerce marketing management and decision making, and to obtain internationally advanced achievement in corresponding field.