Analysis of User Needs on Downloading Behavior of English Vocabulary APPs Based on Data Mining for Online Comments

With highly developed social media, English learning Applications have become a new type of mobile resources, and online comments posted by users after using them not only an important source intellectual competition for enterprises, but can also help understand customers’ requirements, thereby improving product functionalities service quality, solve the pain points iteration innovation. Based on this, this paper crawled user three typical APPs (BaiCiZhan, MoMoBeiDanCi BuBeiDanCi), through emotion analysis hotspot mining technology, to obtain requirements then K-means clustering method was used analyze requirements. Finally, quantile regression is find out which needs impact downloads vocabulary APPs. The results show that: (1) Positive more significant users’ behavior than negative comments. (2) with higher downloads, both 5-star ratings increase emotional requirement effect in APP while enterprise’s improvement has positive downloads. (3) Regarding average or high adaptability Appearance (4) functional improve products will