Abstract: Search results of web search engines are displayed according to their ranking function, which is a function of the in-links and the out-links of the web page. Users are generally overwhelmed by the thousands of results retrieved by the search engine, few of which are useful. Most of the search engine users are interested in the latest information about the searched keywords. Such pages containing the latest information about an event (or incident) are not always ranked high by the search engines due to the lack of sufficient in-links pointing to these web pages. For example, if a search query “DaWaK” is given to Google, the users are generally interested in the latest information i.e., the home page of DaWaK 2004 conference, which is not ranked in the top 10 results returned by Google. In this paper we address the problem of ranking the web pages based on events (or incidents) related to the searched keywords. We provide a method for finding patterns that constitute events in the search results returned by a conventional search engine and then rank the web pages based on these event patterns. We also describe a mechanism whereby our technique can be used inside the ranking function of a conventional search engine such as Google. We provide experimental results that validate the efficiency and use of our technique in capturing the user intentions.
Publication Year: 2004
Publication Date: 2004-01-01
Language: en
Type: book-chapter
Indexed In: ['crossref']
Access and Citation
Cited By Count: 1
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot