Application of Domain-Specific Search Method in Meta-Search Engine on Internet

With the Internet's prosperity, more surfers find it is difficult to retrieve information efficiently. Domain-specific search engine (DSSE) is one of solutions to alleviate such inefficient situation. This paper proposes a new method for building DSSE based on meta-search engine (MSE) on Internet. First, odds ratio method is used to refine the domain-specific keywords, which are weighted by means of TF-IDF method. Second, domain query expression is constructed with the keywords by Decision Tree method to reflect the features of domain documents. Combining user's plain query with the domain query expression, the MSE constructs a domain-specific query expression and submits it to general-purpose search engines (GSE). The Extended Boolean Model is employed to rank the search results back from the GSE. Finally, the ranked results are returned to the query user. The experiment results show that the proposed domain-specific search method is effective in searching domain documents. The designed DSSE is more simple and efficient than the dedicated domain-specific search engines.