A Web Search Engine-Based Approach to Measure Semantic Similarity between Words

Abstract: A web search engine is software code that is designed to search for information on the World Wide Web. The search results are generally presented in a line of results often referred to as search engine results pages (SERP's). The information may be a specialist in web pages, images, information and other types of files. Some search engines also mine data available in data basesor open directories. Semantic similarity or semantic relatedness is a concept whereby a set of documents or terms within term lists are assigned a metric based on the likeness of their meaning / semantic content. The existing technique implemented for the sematic similarity of the words is based on novel pattern extraction algorithm and a pattern clustering algorithm. Here we are implementing an efficient algorithm for the searching of the similarity of the words using Machine Learning based rule approach.

Keywords: Web, Search Engine, Measure, Semantic.