An Intelligent System for Semantic Information Retrieval Information from Textual Web Documents

Abstract

Text data, which are represented as free text in World Wide Web (WWW), are inherently unstructured and hence it becomes difficult to directly process the text data by computer programs. There has been great interest in text mining techniques recently for helping users to quickly gain knowledge from the Web. Text mining technologies usually involve tasks such as text refining which transforms free text into an intermediate representation form which is machine-processable and knowledge distillation which deduces patterns or knowledge from the intermediate form. These text representation methodologies consider documents as bags of words and ignore the meanings and ideas their authors want to convey. As terms are treated as individual items in such simplistic representations, terms lose their semantic relations and texts lose their original meanings. In this paper, we propose a system that overcomes the limitations of the existing technologies to retrieve the information from the knowledge discovered through data mining based on the detailed meanings of the text. For this, we propose a Knowledge representation technique, which uses Resources Description Framework (RDF) metadata to represent the semantic relations, which are extracted from textual web document using natural language processing techniques. The main objective of the creation of RDF metadata in this system is to have flexibility for easy retrieval of the semantic information effectively. We also propose an effective SEMantic INformation RETrieval algorithm called SEMINRET algorithm. The experimental results obtained from this system show that the computations of Precision and Recall in RDF databases are highly accurate when compared to XML databases. Moreover, it is observed from our experiments that the document retrieval from the RDF database is more efficient than the document retrieval using XML databases.