Users' Web Browsing Behaviour Pattern Mining And Its Application In The Online Teaching System

The online studying web server records the access-log of the users.Using the method of Data mining,we can find the users' studying behaviour patterns that hiden in the log file.The studying behaviour patterns will help us to reconstruct the web server or to personalize the web server.Reconstructing will make the studying server more human and more intelligent.Users will find that it is easyer to get the studying resource.When the users click one page,the other related pages' link(according to the users' behaviour pattern) will appear automatically.These tow applications apparently increase the users' experience in the online studying system.The preprocessing of the log-file is one of the most important and complited steps, and the following steps are besed on the result of it.During the preprocessing,Users identifying,Users' Session identifying and Users' episode constructing are the core-procedures.In the paper,we give a new method that base on the average reading speed of the users and the accessing time thredhold to split Users' Session sequence.Compared with the fixed max accessing time thredhold method,our method is more accurate.Our method can build long User Session.After the preprocessing,we try to find users' accessing behaviour pattern in the users' session sequence with WEKA,which is open-source.Based on these tow step,we give an online teaching system model with educational effect self-feedback founction.