INTELLIGENT AGENT FOR INFORMATION EXTRACTION BASED ON PATTERN DISCOVERY AND ONTOLOGY
Yousef S. Abuzir
Pages: 723-731 Published: 2 Jun 2014
Views: 2,130 Downloads: 780
Abstract: In recent years, several approaches have been proposed to extract information from web pages on the internet. In this research, a key technique focused on crawling and ontology used to discover knowledge from web. In this paper, we present intelligent crawling system that uses pattern and ontology to extract particular information from WEB sites. The system developed as an efficient tool to construct researcher’s profile automatically from web pages. Moreover, some searching and indexing methods, text mining and computational linguistics for underlying this problem are exploited. We evaluated the performance of our system on an information extraction task from different real academic web sites. Experimental results show that with the extraction rules based on pattern discovery and ontology, our system achieves 84.90 % average of overall precision.
Keywords: information extraction, knowledge discovery, web mining, ontology, agent, crawl