Semoogle - An Ontology Based Search Engine

Abstract

In this thesis, we present a prototype for search engine to show how such a semantic search application based on ontology techniques contributes to save time for user, and improve the quality of relevant search results compared to a traditional search engine. This system is built as a query improvement module, which uses ontology and sorts the results search based on four predefined categories. The first and important part of the implementation of search engine prototype is to apply ontology to define the meaning and the relations between the queries in default domain of the study. Next, categorization of the results is carried out in order to improve the quality of result search presentation based on categorization-list. The ontology used in this search engine prototype includes sample of terms in safety and security domain, which is capable to be modified in this domain, or can be substituted by another ontology in the other fields of study. The process is continued by searching the enriched query through the Web using Google interface application search engine. The application uses ranking algorithms to categorize and organize the results of Google search in four categories, i.e. History, Mechanism, Prevention, and Case study. The predefined categories can be substituted to the other categories based on user preferences in other studies using different categorizes.