Research on E-commerce Customer Evaluation System in the Context of Big Data: Taking Amazon as an Example

Abstract

Amazon provides the customers an opportunity to assess their purchases. Manufacturers use this data to acquire further insights into the marketing the timing of participation, and potentially successful product designs. The first step is to preprocess the raw data by detecting outliers and using cubic spline method to complete them and verify the correctness. Then, the data is analyzed in time order, followed by data quantification and visualization. The conclusion is that the comment does affect the star rating. Finally, as a consultant hired by Sunshine, put forward some suggestions to the company’s marketing director: elaborating the establishment of relevant models, explaining the problems solved by the corresponding models, and summarizing the team’s analysis and results.