Skip to Content

Semantic Information Mining

The low user entry barrier of the Web has resulted in massive amounts of unstructured and weakly structured data referring to objects, concepts, user interests and communities. Providing automated or semi-automated means of bridging the ‘semantic gap’ - the gap between unstructured or semi-structured data and structured semantic formats – is a significant challenge. The goals of this research stream are to develop techniques for the automatic semantic and structural analysis of Web data, enabling the extraction, integration, indexing and reuse of knowledge, and to provide analysis of the dynamic structures and behaviours underlying knowledge production in shared information spaces.

Research on the Semantic Information Mining stream is carried out in the context of projects within the Unit for Natural Language Processing (UNLP) as well as the Unit for Information Mining and Retrieval (UIMR) as outlined in the picture below: