Research On The Web-based Data Mining And Personalized Information Service

With the rapid development of the Internet in the information flow, size, services and complexity, WWW has become an enormous scale, distribution and a wide range of global information service center. However, WWW is bringing a wealth of information and great convenience while also resulting in a number of pressing problems, personalized information services is one of them. To solve this problem, personalized information services have become a hot topic of current research, it is through a variety of methods, site content, structure, user behavior, such as data mining, the establishment of user interest model, providing better service to site users and raising competitiveness, and improving user loyalty to the site.In this paper, Web-based data mining in the realization of personalized information service carried out a number of key technologies study methods commonly used in data mining, Web data mining architecture, personalized access to information of the status quo and so on, studies of existing personality Information Service System Information filtering clustering algorithm to study the two types of PageGather improved algorithm based on an algorithm - a progressive PageGather algorithm, and distributed PageGather algorithm to study the effects of two kinds of application of the algorithms that these improvements to reduce the input data volume and complexity of the algorithm have a certain value.The value of this article also lies in developing web-based personalized recommendation system of the necessity, feasibility, software development and application of the status quo and system development and application of the practical difficulties encountered in the process have been studied, And put forward a program combining the benefits of users, personalized information systems developers, and Web site, as well as web-based data mining of personalized information systems prototype.