Research On The Web Log Mining Of Teaching Resources Searching Platform

With the growing number of Web applications, Web database size is also expanding its data volume will gradually increase. Web log mining using data mining techniques to log mining web server log analysis, log in to explore the potential of rules and patterns that will eventually apply to aspects of site architecture design, personalized service. Web log mining process is usually divided into three phases:data preprocessing stage, pattern discovery phase and pattern analysis phase. Throughout the Web log mining process, the most important is the data preprocessing stage, it can directly affect the performance of the algorithm and the results back pattern recognition and pattern analysis. This is the main part of the session identification data preprocessing, but also the most basic, critical step.The main innovative research work includes:(1) Gives a Web session identification method based on dynamic time threshold. Several sessions of the currently used identification methods are described in detail, analyzing the advantages and disadvantages of each method, in reference to the basis of the time-based heuristic identification method is proposed on the site home page to begin a new session to dynamically session time threshold to decide improved identification session border, gives the algorithm flow chart and the specific implementation method. Experimental results show that the improved method can identify not only the session identification more real user session, but also effectively improve the accuracy of the whole session identification and recognition degrees.(2) Designed a teaching resources search platform based on Web log mining. The platform in GuangXi University of Chinese medicine school website IIS log as treatment object, select the log information one day in July 2013 as a systems analysis of the data. Designed the overall architecture of the system, the system function has carried on the detailed description of the main modules, the data table structure and flow charts of each link are given, the program can realized the prototype system.