A Study On The Subject Discovery Of Motion APP User's Comment Demand

With the development of Internet technology and mobile technology,application(APP)has emerged.The emergence of APP has brought great changes to people's life and work style.Sports app can stimulate people's enthusiasm for sports,with scientific guidance methods,including various types of fitness projects,reasonable diet suggestions,and provide users with personalized recommendation and big data Professional personalized health guidance is highly praised by users.At the same time,app developers and operators are trying their best to constantly improve the functions of app,therefore boost their competitiveness and gain more advantages in similar apps.This paper focuses on the research of sports app: 1)according to the user characteristics of sports app,combined with the theory and method of user portrait,this paper discusses the reason and method of constructing sports app user demand portrait model,and expounds the processing and construction process of relevant data in the construction of portrait,so as to provide method reference for accurate recommendation of APP platform;2)in the case of big data In this context,it is of great significance to mine valuable information from a large number of user comments.Based on the sports app user comments,this paper puts forward the research idea of sports app user online comment topic discovery.Taking the sports app Gudong as the research object,this paper collects the user comment data of Gudong software,introduces the topic recognition model,and uses LDA topic model to mine useful information from the user comment data User comments 3)research on demand aggregation based on Sports app reviews,the purpose of which is to aggregate the core needs of users,so as to facilitate the sports app platform to efficiently obtain the core needs of users,improve the platform functions and products in order,and use k-means clustering technology to classify the user review data,In addition,according to the experimental results,this paper puts forward the strategy suggestions of APP platform,which provides reference and ideas for platform developers and operators to improve the service level.The experimental results show that the research ideas and methods proposed in this paper can effectively identify the topic of sports app user reviews,complete the purpose of sports app user demand aggregation research,and on this basis,put forward corresponding suggestions for the development of APP platform.Using LDA topic model,this paper studies the text topic discovery of Gudong app users' online review data from January 1,2020 to December 31,2020,summarizes the top five topics of the review data and the corresponding top six keywords with the highest frequency,and finds that users' experience in the process of using the app is different,and their views are divided.In topic 1,the words "easy to use","very good","good" and so on are used It shows that users have a good sense of experience,and the keywords such as "flash back","no","advertisement" in theme 3 reflect users' dissatisfaction with the platform.Therefore,the platform should optimize the app function on the existing basis.Research on user requirements aggregation.Demand aggregation can better integrate the uneven distribution of users' needs.On the one hand,it can quickly focus on users' needs.On the other hand,it can help the platform analyze the types of needs and formulate measures in an orderly manner.In this paper,we use k-means clustering technology to classify the user's comment data,and get the user's comment demand cluster.Secondly,we show the comment cluster word cloud,which is intuitive and concise,and can clearly and accurately analyze the user's demand.Based on the results,we put forward the corresponding suggestions for the app platform: 1.Platform developers should continue to optimize app functions,such as optimizing programs,to avoid the phenomenon of flashback Secondly,we should set reasonable charging items,crack down on arbitrary charging behavior,and create a good and healthy network environment for users.Good app functions can make users feel friendly and comfortable,make them feel happy in the process of using,and the platform can also get a high reputation.2.If conditions permit,the platform should actively meet the needs of users.In the data of this paper,besides paying attention to basic sports,users also have certain expectations for sports coaches.If more professional coach teaching functions can be developed on the platform,users will get good feedback.In addition,if the platform can continue to explore on this basis,such as the types of private education,the introduction of preferential charging policies and other functions,users' choice will be greatly increased The right to choose.If these needs are met,it will enhance the user's happiness,because these implied needs gradually turn into real needs.3.In the data of this paper,the platform does not need to pay attention to the difference of user's age and gender,because in the comments,there is no data about the influence of age and gender on their own needs.However,the platform should follow up the user's attitude towards irrelevant needs,so as to make timely and accurate adjustments when user needs change.The experimental results show that the research ideas and methods proposed in this paper can effectively identify the topic of sports app user reviews,complete the purpose of sports app user demand aggregation research,and on this basis,put forward corresponding suggestions for the development of APP platform.