Analysis Of User Needs Of Mobile Reading App Based On Online Reviews

With the continuous expansion of the scale of mobile device users,mobile Internet has rapidly penetrated into many traditional industries with mobile applications(APPs)as the media,and the number of apps on the shelves in the application market is increasing.Developers have many competitors in the same field.Only by constantly meeting the needs of users and bringing good online experience to users,can they retain customers for a long time and maintain the competitiveness of the industry.Whether they can have a large user group and activity is the key to the success of mobile app.Therefore,how to effectively tap user needs and promote the better development of mobile digital product industry has become an urgent problem to be solved.Based on the above situation,this paper takes 10 typical App users' comments data in App Stone as the research sample to mine the comments features and analyze the users' needs.In the aspect of comment feature mining,this thesis focuses on the steps of data collection and data cleaning,transforming the original text into the text for analysis,and analyzing by the tools of word cloud graph,semantic network analysis,LDA subject model,etc,identify user-focused discussion topics and links between feature words.Based on the results of the above-mentioned review feature mining,the requirement analysis is carried out.Firstly,the attribute words of APP users' needs are extracted,and the attribute-affective word pairs are set up by dependency syntax,and the emotion words matching the attribute words are extracted,and the emotion tendency of attribute words and user comments are determined by sentiment analysis.Secondly,through the emotional tendency of attribute words and attribute words,the attribute emotional value matrix is established.On this basis,each user comment is regarded as a perceived stimulus to the attributes by the user using the App and the user's demand utility function model is established,and the affective tendency of each attribute word is brought into the demand utility function model,through the joint analysis method to determine user needs preferences and weights,based on Kano model to classify user needs,and verify the effectiveness.According to the visualization result of the topic model,the keyword with high TF-IDF value in the topic model is selected as the user requirement attribute word.After screening,the keywords such as font,note,page,share,check-in,eye protection,privacy,bookmark,check-in,content and so on are selected as the requirements attributes for this study.These requirements are representative in user comments,the method can also be used to extract the corresponding requirement attributes of App in other domains.Privacy,bookmarks,and the quality of the content of the books weigh heavily,and developers should be the focus of optimization.In addition,the study found that features such as turning pages,bookmarking,and content richness are essential attributes of apps,and users experience strong negative emotions when these features are absent;typography,privacy,and eye care are desirable attributes,if you can extend a user's reading time by providing personalized attributes,the user's experience will improve;notes,checkins,and sharing are attractive attributes that can dramatically increase a user's reading satisfaction.