Classification of durian characteristics for semantic representation from web documents

The Web contains enormous size of information that is represented in various document structures. The information is scattered and redundant. Currently, search engine is the main medium for retrieving this information. Yet, the most popular search engine cannot satisfy user query. Alternatively, semantic technology can alleviate this problem. In this paper, only relevant web HTML documents on durian also known as king of fruits are chosen. The characteristics of durian will be extracted from those HTML documents. These characteristics are then employed in semantic representation and stored along with their Uniform Resource Identifier (URI) in Resource Description Framework (RDF). The RDF provides the ontology link to many other web documents on durian. Experiment on 40 HTML documents provides eleven new characteristics of durian that can be represent in RDF for semantic search engine.