Analysis Of Network User Behavior Based On The Big Data

Nowadays,due to the rapid development of the Internet and the popularization of the application,the data volume of web is increasing constantly,most users will search engine as the first choice for information retrieval.Search engines can be chaotic information integration,for the user to establish an orderly index documents,users can quickly and easily retrieve information.However,in the Internet and produced using the search engine search logs more users so the Internet companies or search companies hope that through the search log analysis and mining of accurate and effective to launch the user's behavior,so as to enhance the customer satisfaction of search engine.However,the analysis of the user's behavior is faced with two major problems.First of all,how to solve the storage andpreprocessingof a large number of search logs;secondly,how to design the user behavior analysis model is used to obtain the user behavior characteristics,in the query log data of user access,how to choose the appropriate research platform to study user behavior.According to the above analysis and reference to a large number of literature,this paper uses the user behavior analysis system based on Hadoop platform.Hadoop is mainly used as the analysis platform,the use of HBase to store massive logs and the use of MapReduce computing model for user behavior analysis.Sogou user's search log as the analysis of the data source.Combined with the data mining algorithm,the characteristics of user retrieval behavior are analyzed.A massive user behavior analysis platform based on query log is designed,which includes three modules,namely,data processing module,data storage module,data analysis module,data analysis module is the key of the whole system,mainly from the keyword ranking,URL ranking,time andother aspects of the analysis of statistics the query log,and Web text mining process for ideas for the user query log analysis.Finally,this paper introduces the development and deployment of Hadoop cluster development environment.Through the detailed analysis of the user's search log data,the analysis results can be obtained.