Computer-aided Acquisition of Semantic Knowledge Project (CASK)

The Project is aimed at describing selected semantic fields using KDD (Knowledge Discovery in Databases) methods. The project is interdisciplinary involving linguists, logicians and computer-scientists. Establishing hierarchical (meta-)ontological definitions of Feature Structures usable as semantic information systems will contribute to build a common basis for describing semantic fields of different European languages.

Two complementary mono- and multi-lingual approaches have been adopted in order to enhance the possibility of finding semantic features which would reveal general enough for building the future meta-ontology of European languages semantics. Both approaches may be adopted by the research teams of CASK bilateral projects. The French-Polish CASK project was one such event.

The main idea of such a CASK "projects bundle” is to build a bank of semantic feature structures which will be based on the ontological inquiry of the most salient linguistic semantic fields of European Languages. To achieve this, linguists collect a sufficiently representative number of samples of linguistic expressions so that the latter can be described using interactively developed  semantic structures. Due to the very high precision of description tools, semantic fields of  ‘grammatical categories’ (such as Aspect, Modality, Tense) have been chosen prior to lexical items. However, some basic research on the lexical domain of Motion, namely displacement verb semantics, has already been made, too.

The KDD methods chosen presently allow to make very precise analyses using highly advanced technologies (and their combinations) such as algorithms of Decision Logic, Rough Set Theory and Formal Concept Analysis for symbolic data processing on the one hand, and Cluster and Factor Analyses algorithms for statistical data processing on the other hand. The above mentioned algorithms together with database building tools were implemented as SEMANA platform.

Possible future utilization of the results is development of Multi-Lingual Electronic Dictionaries and Machine Translation Systems. In order to get the largest and the deepest semantic knowledge of each studied European language, results of the research on the Japanese Language as a kind of “contrastive pivot language” will be extensively used.

Semana Software:
Free Licence User's Registration Form

Japanese version of this description


Back to CELTA (Centre de Linguistique Théorique et Appliquée)

© Copyright Université Paris Sorbonne (Paris 4)