Publication

Abstract : Text mining has wide range of applications from biotechnology to media and security. In the area of computer science, text mining can be used to improve the areas of search and to automate analysis processes. Topic modeling, which is a part of text mining, has been playing a vital part in web content and search. It is used to lookup information and analyzes user needs and services. In this paper, we are using topic modeling algorithm (LDA) to find user search preferences from web browsing histories. From the results obtained, we can find out the frequently browsed websites and hence can understand their search preferences. This method can also be used to develop better services for users and can be used to develop software that can even track terrorist activities. The preprocessing techniques, application of algorithm and analysis of the results are discussed in this paper.