Temporal Knowledge Bases Group
The Temporal Knowledge Bases Group at UJI
The ever-increasing amount of information generated by the Information Society is posing new challenges to the Database and Knowledge Engineering research areas. Especially, the irruption of the Web, with its billions of web pages, as well as the emerging Semantic Web, are demanding new methods along with automatic mechanisms for organising, structuring and giving semantics to huge collections of data and documents.
The TKBG group is mainly concerned with the investigation of novel methods for modelling, querying and managing very large collections of semi-structured documents. A semi-structured document combines data and textual contents, and it usually lacks an external schema describing its structure. Current Web languages such as HTML, XML and RDF fall into this category of documents.
Our approach for the effective exploitation of these large document collections combines techniques stemming from several research areas: Databases, Information Retrieval, Natural Language Processing and Pattern Recognition.
The main active research lines started by the group around this problem are the following ones:
- Knowledge(-based) extraction from large corpora.
- Mining very large knoweldge resources (KR).
- Multidimensional analysis of unconventional data.