Ontology based video information mining tools

Abstract

Finding information on World Wide Web or web is the most popular and interactive searching medium to retrieve information these days. Web is easy to access and there are lots ofinformation available on the web. With growth of amount ofthe content or information, web has become increasingly difficult to find the data on the web quickly. The easiest way of searching on the web is search engines. Search engines returns thousands of results for small search keyword. It is difficult to filter out the required information from these search results. Web content mining is one of the most popular and effective way of filtering data. Users will be able to retrieve their required information through this kind of filtering mechanism. Web content mining is specific to particular area. The filtered data can be saved locally in order to user later. This information can be stored in a database system to use as required. Integrating relational database with ontology is an active research area these days. Ontology will be a core part of the next generation web. Current web (web 2.0) mostly relies on keyword based search. The performance of the system depends on the matching the keyword with available data. This study has focused on web content mining technologies and integrating ontology with relational database system to manipulate filtered data. The domain ofthe study is movies and related information. Movie related information is filtered using a web crawler and this information is saved in a database system after further cleaning. Web crawler is configured to get the movie related information from the web. Ontology is designed based on the movie related information. After designing the database and the ontology, these two layers should be mapped. This part is done using a data wrapper. This wrapper links ontology instances with database system. After mapping is done there should be a proper way to retrieve the data from ontology. SPARQL is used to query the ontology and SPARQL will get the data from the database through ontology mapping. Finally user interface is designed to interact with the user. User can search through two interfaces called general search advanced search. Using general search user will be able to search a keyword and through advanced search user can search using multiple parameters. Users will be able to search based on their actual searching criteria’s rather that searching a keyword.