Community comment emotion tendency analysis-based mobile phone game ranking list construction method

Abstract

The invention discloses a community comment emotion tendency analysis-based mobile phone game ranking list construction method. The method comprises the following steps of: 1) pre-processing an evaluation text input by a user; 2) carrying out word segmentation and part-of-speech tagging on pre-processed mobile phone game comments; 3) selecting high-frequency emotion words as emotion characteristics so as to generate all characteristic vectors required by machine learning; 4) inputting the characteristic vectors and evaluation polarity labels into an SVM classifier to carry out model training; (5) arbitrarily inputting a segment of evaluation text by the user, and after carrying out operations of word segmentation, part-of-speech tagging and characterization, automatically carrying out evaluation polarity estimation on the evaluation text by a training model of the SVM classifier; and (6) generating an evaluation situation corresponding to each mobile phone game, and constructing a corresponding mobile phone game positive comment ranking list according to the quantities of positive comments and negative comments. The method for researching the comment polarity through emotion tendency analysis can extend to various fields and has wide application prospects.

Description

technical field [0001] The present invention relates to the technical field of sentiment analysis, in particular to a mobile game leaderboard construction method based on community comment sentiment analysis. Background technique [0002] With the continuous popularization of smart phones and the continuous improvement of the mobile Internet, the development of mobile games is also thriving. Mobile games have become a major interest pillar of mobile phone users nowadays, bringing huge benefits and broad development space to major game developers and middlemen. In the face of the rapidly growing mobile game user base, how to recommend games that meet the players’ interests and preferences, attract as many users as possible, and develop them to retain them has become an important issue for research and discussion in current mobile game applications. [0003] It is obvious that each game has a different number of user reviews, whether in mobile game forum communities or major ...

AI Technical Summary

Problems solved by technology

Therefore, although these comments are categorized with emotional polarity, these labels cannot correctly reflect the real emotional orientation of the comment, thus misleading other players and causing difficulties to a certain extent in constructing a mobile game praise ranking

[0005] At present, most researches on the field of Chinese sentiment analysis focus on Weibo, product reviews, news reviews and public opinion monitoring, but there are few studies on the use of mobile game reviews for sentiment analysis and judgment.

Images