A Responsive web search system using word co-occurrence density and ontology: Independent study project search

Abstract

This research aims to develop a responsive Web search system for the independent study project search of the department of computer and information technology, Nakhon Sawan Rajabhat University. There are four features of the proposed search system. First, the responsive Web design using bootstrap framework is used for designing the Web site. Thus, the Web sites can adapt to the user’s viewing screen size; second, utilizing ontology for storing independent study project database in order to establish the data relationship and support semantic search; third, word segmentation using THSplitLib library for Thai language. Last, the proposed responsive Web search system was implemented in PHP and SPARQL for ontology search via RAP API. In addition, the retrieval decision making is based on co-occurrence density analysis. The search term can be one or more the one keyword or a sentence. The search results also display the project advisor information that is concurrent with the search term. Hence, students have more information for effective decision making. The experimental results show that the Web site can adapt to the various screen sizes and has high efficiency. The average F-measure is 95.02%, the average search precision is 94.70%, and the average recall is 95.35%.