An Intelligent System Integrated with Fuzzy Ontology for Product Recommendation and Retrieval

Abstract: – Systems that are integrated with recommendation features to support decision processes can provide a more personalized and proactive retrieval experience for users to choose among retrieval alternatives and subsequent refinement of choices. Many real-world systems call for autonomous intelligent agents acting in the face of uncertain knowledge and limited computational resources. This paper explores the development of an intelligence multi-agent based system for book recommendation and retrieval. It enlists the features of our iJADE (intelligent Java Agent-based Development Environment platform) to provide convenient Java tools and libraries for the agent components. The system possesses fuzzy reasoning features suitable for comparing, evaluating and classifying the webagents behavior. The fuzzy decision-making is based on user preferences, product feature selection and system recommendation. We can make use of the platform to build a web system integrated with the mobile multi-agent system and fuzzy ontology. The use of mobile agent technology provides a specific solution for overcoming the problem of being overloaded with too much information. User can simply enter preference information through the web browser and the backend iJADE server gets the request and sends back the relevant recommendations to the client. The simulation testing shows promising results. Key-words:- Information retrieval, intelligent agents, fuzzy ontology, recommender system, eCommerce 1