Web Mining Based on User Profile and Preferences
Harry Tursulistyono Yani Achsan · Annals of DAAAM for ... & proceedings of the ... International DAAAM Symposium · 2011
Finding documents and information in the Internet becomes easier after the emergence of various search engines such as AltaVista, HotBot, Yahoo, and Google.But none of these search engines can give documents relevant to the user background knowledge and interests.Nevertheless, these search engines can be used in a meta search engine application that will predict user interests.This paper describes the idea of how to obtain user profiles or user interests implicitly from the query terms/keywords when searching in the Internet and also from documents/web pages opened by the user.Terms frequency and inverse documents frequency (tf-idf) are used to create the user profiles.Currently, this paper concentrates only on long term user profiles and neglects ad hoc interest.The software/application developed in this research aims to collect user profiles in two tables.First table records query terms and its frequencies, and second table records tf-idf extracted from documents opened by the user.The query output generated by this application is ranked based on three extended query.From the experiments, the application can provide information relevant to its user profile