Web search with personalization and knowledge

Abstract

Although many search engines provide relevantly good search results to the user, they do not consider personal, domain-specific preferences in their searching or ranking algorithms. In an intranet environment we could collect background information about users such as their expertise. If we can accumulate, categorize and personalize Web usage information, it can be used to help the user search Web pages efficiently and effectively. Data analysis and mining can further facilitate Web searching in an intelligent way. This paper describes Internet Search Advisor (ISA), a personalized, knowledge-driven search system that helps the user find informative Web sites. The ISA supports multi-dimensional data analysis and data mining based on association rules and sequential patterns.