Abstract
A large number of scientific papers are retrieved using Search Engines from the electronic databases. Some of these Engines are limited and others have designed for a general purpose. A number of researchers wish to prepare a survey of a particular topic. They are facing a problem to find the most related topics to a particular research title. The other problem is rising as a result of a search in an electronic database, where some Search Engines displays Dozens of pages and hundreds of results, it needs also more effort to be scanned manually and decide which results are relevant and which should be excluded. During the search process and matching the contents, the Search Engine maybe ignore some important documents. Some of these documents are excluded although, it is relevant to the subject and some results are included but not important. This research concentrates on a development of a Multi Scanning Filter (MSF) algorithm, that works on research documents found in various scientific databases, such as ISI, SCOPUS or EBSCO, etc. The idea of this research depends on the Google Search Engine, where the proposed algorithm consists of three parts. It maximizes the search space and works as a filter to Google results based on the similarity measure. This algorithm reduces the final search result list, make it more accurate, eliminate the problem of results' dispersion in traditional Search Engines, and helps developers improve current Search Engines, such as Google, this in turn will assist researchers everywhere gather the most related topics to a particular research title in a short time.