Soft fusion of information accesses

In this paper the definition of a meta search engine model, named SOFIA (soft fusion of information accesses), is proposed that applies a flexible and soft fusion of the ranked lists of documents retrieved by distinct search engines available over the Internet. The decision function that specifies the fusion criterion is expressed by a linguistic quantifier modeled by an induced ordered average (IOWA) operator. Fuzzy quantifiers make soft fusion criteria possible. Further flexibility is obtained by computing the distinct fitness scores of the search engines based on a relevance feedback mechanism and using these scores to define the IOWA reorder vector. In this way the search engines with highest fitness influence more heavily the ranking of the documents in the fused list