Rating and Evaluation Model Based on Grey Correlation Analysis and Time Series

Nowadays, large-scale online markets are rapidly emerging, such as Tmall, Amazon and other online platforms, whose transaction value is up to several billion dollars. One of the most important principles of such a successful online shopping platform is the principle of customer obsession. Many platforms provide users with product rating opportunities, and users can give 1-5 points for products to reflect their satisfaction with this transaction. The large amount of data of these ratings provides a lot of useful information to be mined for decision makers. In this paper, a Rating and Evaluation Model Based on Grey Correlation Analysis and Time Series is put forward, Based on the product transaction information obtained through data mining, it first constructs an evaluation model of commodities based on the gray correlation analysis, three main consideration indicators are selected, namely, Comment, Rating value and Full-mark quantity, the corresponding weights are obtained in combination with the method of analytic hierarchy process, the gray weighted correlation degree of each commodity in a period of time is then calculated, which represents the degree of popularity of the commodity. Subsequently, the ARIMA model was also established by using time series tools to predict the grey weighted correlation of different commodities in the future, and the trend of the popularity of commodities in the future is also obtained. The overall trend of shopping and the popularity of commodities on Tmall shopping platform before and after the epidemic time are compared and analyzed, which is conducive to the further development of China’s online shopping platform from a statistical perspective.