Low-Bias Extraction of Domain-Specific Concepts
Axel-Cyrille Ngonga Ngomo · 2010
The availability of domain-specific knowledge models in various forms has led to the development of several tools and applications specialized on complex domains such as bio-medecine, tourism and chemistry. Yet, most of the current approaches to the extraction of domain-specific knowledge from text are limited in their portability to other domains and languages. In this paper, we present and evaluate an approach to the low-bias extraction of domain-specific concepts. Our approach is based on graph clustering and makes no use of a-priori knowledge about the language or the domain to process. Therefore, it can be used on virtually any language. The evaluation is carried out on two data sets of different cleanness and size. Povzetek: Od jezika neodvisna metoda iz besedila izlušči termine in nato domensko odvisne koncepte. 1