Authors:
Shanchan Wu
and
Wenyuan Wang
Affiliation:
Tsinghua University, China
Keyword(s):
data mining, web mining, probability model, prediction
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Biomedical Engineering
;
Business Analytics
;
Data Engineering
;
Data Mining
;
Databases and Information Systems Integration
;
Datamining
;
Enterprise Information Systems
;
Health Information Systems
;
Sensor Networks
;
Signal Processing
;
Soft Computing
Abstract:
As the world-wide-web grows rapidly and a user's browsing experiences are needed to be personalized, the problem of predicting a user's behavior on a web-site has become important. In this paper, we present a probability model to utilize path profiles of users from web logs to predict the user's future requests. Each of the user's next probable requests is given a conditional probability value, which is calculated according to the function presented by us. Our model can give several predictions ranked by the values of their probability instead of giving one, thus increasing recommending ability. Based on a compact tree structure, our algorithm is efficient. Our result can potentially be applied to a wide range of applications on the web, including pre-sending, pre-fetching, enhancement of recommendation systems as well as web caching policies. The experiments show that our model has a good performance.