Hybrid profiling in information retrieval

One of the main challenges in search engine quality of service is how to satisfy the needs and the interests of individual users. This raises the fundamental issue of how to identify and select the information that is relevant to a specific user. This concern over generic provision and the lack of search precision have provided the impetus for the research into Web Search personalisation. In this paper a hybrid user profiling system is proposed - a combination of explicit and implicit user profiles for improving the web search effectiveness in terms of precision and recall. The proposed system is content-based and implements the Vector Space Model. Experimental results, supported by significance tests, indicate that the system offers better precision and recall in comparison to traditional search engines.