Research On Statistical Laws Of User Network Behavior Diversity

With the rapid development of the information society,a large number of user behaviors have transferred from offline to online.This change has led to an explosive growth of users'online behavior data.Also,there are many human behavior rules behind such a huge amount of data.To discover these rules,research on users' behavior based on online data has attracted widespread attention from scholars.The analysis of users' behavior based on online data mining is useful since it could contribute to the user's network experience and the quality of people's lives.Therefore,this paper takes diet and knowledge transmission as examples,studies the diversity and evolution mechanism of users' online behaviors via data mining technology and multiple linear regression models.In the first part,this paper mainly studies the diversity of eating behaviors.Firstly,the principle of data collecting of Python is introduced.Then 1809 users' information on DaZhongDianPing is collected by Requests and BeautifulSoup package.Through these data,the correlation between the diversity of dietary behaviors and the characteristics of the users are analyzed systematically.Also,the relation between the diversity of dietary behaviors and consumption levels of users are investigated using multiple regression analysis.It has been found that there is a clear "inverted U-shaped" trend between the diversity of diet behaviors and the consumption level of the users,that is the diet diversity will increase with the rise of consumption level,however,after the consumption level exceed to a certain level,the relation between them will become negative.What's more,the gender,age,marriage and love status also influence the diversity of dietary behaviors.From the results of the first part,this paper finds that the online behavior of users is always inseparable from the spread of knowledge.Therefore,this paper further explores the relationship and evolution mechanism between users' online behavior and the dissemination of knowledge mechanism.Firstly,the method of data collection based on Fiddler is introduced,and the detailed steps are described by taking the application TikTok as an example.Then,this paper uses the Heuristic-Systematic model to analyze the relationship between users' online behavior and knowledge dissemination.Research proves that in the spreading of short knowledge videos,users will be more inclined to focus on the content of the video rather than the status of the author.So the social short video platform empowers individuals through the weakening of author information and motivates individuals to participate in the production and dissemination of knowledge.Based on Maslow's hierarchy of Needs and Heuristic-Systematic Mode of Information Processing,this paper combined with multiple linear regression models to provide new methods for the study of users' online behavior.It also concludes the common data collecting methods which will further contribute to the application of data mining and statistical analysis technology in users' behavior.