Dynamic Evaluation of Consumer Satisfaction Based on the Grey Prediction Model

With the rapid development of e-commerce, a large number of product reviews have arisen on the Internet. Sentimental orientation analysis and mining of product reviews become important for the study of the product reputation. Sentiment analysis takes the form of analyzing and classifying the reviews contents about a product, event, and place etc into positive, negative or neutral opinion. In this paper, we establish a dynamic evaluation system of consumers satisfaction. And a grey prediction model is applied to discover the dynamic evolution of consumers' satisfaction. Finally, the sentiment orientation analysis of the product reviews is finished. The validity of the proposed system is proven through the experimental results on three JD product reviews datasets, which could effectively analyze the developing trends and the sentiment orientation of the product over a period of time, which provides valuable information to enterprise and consumer's decision making.