Method for Measuring the Semantic-similarity of Textual Document and Web-pages

Abstract

In this article, the process of semantic comparison of web pages in the global network is addressed. Comparing process is done based on set of words in text. Google Ajax Search and Yandex-API services are used to search and find web pages in global network. The local database is used for the determination of synonyms of the words. For the experiment, we took a document from Global network and its text was exchanged to synonyms and generated new document. The generated document compared with the given algorithm and it showed 90-95% similarity. The developed system with this algorithm used in Tashkent University of Information Technologies named after Muhammad Al-Khwarizmi for verifying students Graduation Qualification works for plagiarism.