| The amount of data available on the Web has been growing explosively during the last years. The managing and analyzing of the web data becomes an increasingly important issue. To overcome the difficulties associated with semistructured nature of the web data, the ideas taken from the database techniques were proposed. WebOntEx is one of the projects ongoing in this direction, and it attempts to build a system for automatic extraction of ontology from the set of web documents.; According to the WebOntEx approach, web ontology can be extracted by analyzing both the structure of the layout of the web document and actual content. To uncover the hidden semantic information contained within a web document, WebOntEx relies on WordNet electronic library. Using a set of relations provided by WordNet, it is possible to get the meaning of the extracted words and relationships among them. In addition, the relations between concepts can be extracted by analyzing the layout of the document. The structure of the layout can be represented by HTML tag tree. This thesis focuses on different approaches to implement the tag tree. |
