A team of two Chinese scientists from the Taiyuan University of Technology (Shanxi) have, using two distinct software – a search agent and a meta search engine – built a new search engine that intelligently searches through other search engines to give a result most closely suited to the query.
A search agent is defined as an intelligent virtual robot that can very quickly scan through data and look for keywords to assess the context of their findings.
According to the duo, using the new system, a fruit farmer would get information relating only to apples rather than Apple computers, when looking up for information on new strains of the fruit.
"Traditional search engines cannot cope easily with this rapid expansion of information resources,” said Chen Junjie, an information scientist from the university.
Meta searches involve scanning information, not from a single source like Google or Yahoo index, but from all available sources such as Bytesearch, MetaCrawler and Ixquick, well-known meta-search tools.
The intelligent agent determines the context of the user's search terms and chooses appropriate search engines to scan. It then retrieves the most relevant results.
According to Junjie, this approach boosts the precision and recall rate of traditional search engines and fulfil users' query requests well.
“Suppose you are interested in a specific football team, our meta-search engine would find your user profile containing your interest in football, and when you enter a keyword such as "football", the information on your favourite team from the various search engines will be retrieved preferentially,” explained team member Liu Wei.
Their study is described in Inderscience's International Journal of Agent-Oriented Software Engineering.