Multi-agent Meta-search Engine Based on Domain Ontology

This article describes a new approach of HTML pages search via Internet, which is based on the semantic understanding of pages content by means of multi-agent technology. Multi-agent text understanding system, which is the basis of the approach, converts an input query and pages, received from conventional search engines, to formalized semantic descriptors, and evaluates similarity of these descriptors. What makes the approach better than the classical one and gives it an ability to cope with current search-via-Internet difficulties more effectively in the text understanding algorithm, which is capable of understanding words synonymy and ambiguity in a text, detecting and resolving sense conflicts between different parts of text, acquiring additional implicit information and working with inter-phrasal context. Semantic descriptors of the text are represented as semantic networks, which include instances of problem domain ontology concepts. Both text understanding and descriptor comparison algorithms use the knowledge about problem domain, and this problem domain is stored in the form of ontology. Ontology contains the information about the language and the problem domain of the texts used, and it can be easily updated by experts in the course of work. The approach described above was applied to the analysis of web-pages related to car industry. As a result a meta-search engine was developed, capable of analyzing pages, retrieved from traditional search engines and sorting pages by their semantic relevance to the user request. In this article one will find description of the system, testing results and system future perspectives.