Click-through Data Analysis and Use in Web Search Engine

Computer Engineering

While using a search engine, different users might look for different search results when they submit it to a search engine .If the query is ambiguous or covering a broad topic, it is difficult to guess the users interest. So we are proposing an approach which will analyse the search log and provide us users intent . Generally the generic search engines fail to properly outline the user’s interest. So these search engines are not adaptable to individual user. But in our approach user interest is used in the clustering process to achieve personalization effect. The goal of personalized IR (information retrieval) is to return search results that better match the user intent. So to implement this we work in two phases. First, to make use of feedback session to discover users goal. Second, we find the similarities between the first clicked URL and the other URLs available about the query and display the most similar URLs. By doing this we can achieve specialized searching .

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.