移动App用户的海量日志分析的优化策略与算法研究.doc

移动App用户的海量日志分析的优化策略与算法研究 摘　要 关键字： 视频推荐 关联规则 Apriori算法 MapReduce模型  OPTIMIZATION STRATEGY AND ALGORITHM RESEARCH ON MASS LOG ANALYSIS OF MOBILE APP USERS ABSTRACT With the rapid development of Internet ,especially mobile Internet, Internet of things. Currently we have been in the era of data，information overload. However ,The user is difficult to find what she is interested in,facing massive information . And with the Internet companies , operators gradually introduced many Internet applications , such as videos , games,community ,etc . How to analyze these massive log data generated by these products becomes very valuable. In the current research on data mining , the mining on association rules is a very important research area . The research involves many fields of information , health care ,transportation ,finance ,etc. In the association rules Algorithms ,the Apriori algorithm proposed by R.Agrawal and R.Srikant in 1994 is the most important and classic.The core idea is to generate a candidate set and plot closed down two-stage detection to mining frequent item sets. This paper focuses on the research on user massive data processing algorithm , which mainly include the following contents: The usage of Apriori algorithm in user video recommendation system . In this research , we acquisite and process a famous internet company’s massive log data , through mining the degree of association between different videos,and to recommend more relevant video to users . The research of improvement and application on Apriori algorithm , due to open database frequently and a large number of frequent itemsets generated leaving bad time performance.In this paper, we propose some optimization strategies of frequent item set encoding on the basis of a large number of log data analysis . The research of Apriori Based on MapReduce .Faced with TB, even PB-class massive log data ,stand-alone computing is difficult to meet the needs of large data mining . In this study ,we im