User comment analysis method based on text and picture contents

Abstract

The invention discloses a user comment analysis method based on text and picture contents. The method comprises the following steps: 1, inputting weight data of current users participating in ranking; 2, establishing a user ranking method according to the weight data; 3, calculating the user ranking score; 4 storing the user data in a dictionary and storing the ranking score in a skip list, (5) calculating the user ranking according to the skip list and obtaining the corresponding user data, and (6) displaying the user ranking in a game. The ranking system can quickly rank mass users of a farm game. According to the system, the ranking of the users in the game takes the money input ratio and the consumption time input ratio of the consumer products of the users as the ranking basis, and meanwhile, the skip list and the dictionary are taken as the bottom layer data structure of the ranking system, so that the problems of too low fairness of user ranking under multiple weights and low ranking efficiency under mass data can be effectively solved.

Description

technical field [0001] The invention relates to the technical field related to e-commerce, in particular to a user comment analysis method based on text and picture content. Background technique [0002] In recent years, with the continuous development of the Internet, e-commerce platforms have been used in various aspects such as clothing, cosmetics, books, food and agricultural by-products. Among them, user reviews after consumption are the most authentic responses to merchants and products, which can provide more transparent information for other consumers and help consumers further understand the real situation of merchants and products. However, user reviews are often mixed. High-quality user reviews will correctly guide consumers to breed and shop in the game, while poor-quality reviews will reduce consumer confidence in consumption and favorability of e-commerce platforms. In addition, the analysis and screening of user comments by current e-commerce platforms are o...

AI Technical Summary

Problems solved by technology

This method is limited to extracting product nominal information, and has a single function, and cannot deeply mine effective information in user reviews.

[0004] CN104123358A and CN108776864A respectively extract and match keywords from the text content of user comments and use neural network training models to analyze and evaluate user comments, but the data source is single, and the importance of picture comments in user comments is not considered

Images