Evolutionary machine learning for Web mining

Abstract

The Internet is an extensive source of information and searching it exhaustively is inefficient in terms of time complexity. Web search is one of the most universal and influential applications on the Internet. A new search mechanism called SmartSeek is introduced. Inspired by machine learning concepts, this new technique employs a genetic algorithm (GA) for adapting to the user's interests. The system accepts user feedback for fitness evaluation. The mutation operation involved in the process ensures that the search is not confined to a limited domain.