Improving Personalized Search through the use of User Profiles Represented by Wikipedia-based Concepts

Abstract:

The internet is a source of an increasingly enormous amount of information, and search engines assist in sifting through it to fi nd the desired information. Personalized search engines have been developed which re-order results based on the user’s interests; however, current implementations appear to base their personalization algorithms on the URL of previously visited web pages, which do not fully capture the user’s interests. Previous research has successfully used hierarchical structures of concepts to better learn the user’s interests, and this research attempts to improve on these methods by using a flat-structured knowledge base in the form of Wikipedia to provide better personalized search features. Initial results show that the method improves on standard web search, but has yet to match the performance off ered by previous methods. A number of problems and limitations have been outlined, correcting which may greatly improve the performance of this method.

Download.

Advertisements
Tagged , , , ,

Comment?

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: