The Research Of Network Resources Ranking Based On User Browsing Behavior

With the development of information society, Internet has become an important part of people's daily life and work, while the rapidly growing network information become a massive information resources, which should have provided more options for people, but due to the Internet resources has not a uniform distribution pattern, with too much useless information, led to it is difficult to find the need resources. Based on the existing condition, people are trying to find the way to solve the contradiction there are many difficulties when users search for resources they wanted while the resource is too rich. One way is to improve the page layout to enhance the usability by changing the order of network resources.The sort of network resources is an array which based on certain characteristics of resource user groups, so that user can easily find the necessary resources. This paper starts the analysis from the perspective of user community needs, and then mines the acts of users to find the importance of each resource and the relationship between them in a certain type resources. Take it as a basis for sorting of such resources in order to enable users to quickly find the resources. The paper will study from the following three aspects:(1) Analysis the connection between user browsing behavior and the importance of network resources, then as a basis, establish a scores model of network resources based on user browsing behavior, evaluate the importance of each resource, so that support the follow-up mine and final sort of resources.(2) As for the commonly association rules algorithm is based on the number of the project appears, while like the click event in the network is not valid to this situation, the paper make use of weighted association rules to mining the relationship between resources. The paper will take the network score as the weight, which calculated by the scores model of network resources.In order to improve the performance of weighted association rules, an improved weighted association rules is proposed.(3) In order to meet the users'requirement to the certain types of resource at the same time, make the maximum use of resource, this paper presents a sorting algorithm, which takes the page weight as main line, the association rules as basic unit.The proposed sorting model uses the resources importance based on user browsing behavior as the scoring mechanism, and improves the resource order by the resource relationship, which found by weighted association rules. The model provides a good reference value to improve the availability of the site.