Evaluation Framework for Automatic Ontology Extraction Tools: An Experiment
Ontologies have become increasingly important in many areas. Building ontology, however, is a time-consuming activity which requires many resources. Consequently, the need for the automatic ontology extraction tool has been increased for the last two deca
- PDF / 264,618 Bytes
- 11 Pages / 430 x 660 pts Page_size
- 24 Downloads / 256 Views
Abstract. Ontologies have become increasingly important in many areas. Building ontology, however, is a time-consuming activity which requires many resources. Consequently, the need for the automatic ontology extraction tool has been increased for the last two decades, and many tools have been developed for this purpose. Yet, there is no comprehensive framework for evaluating such tools. In this paper, we identified important tool evaluation metrics and developed a set of criteria that guide us to evaluate the quality of ontology extraction tools. We carried out experiments and assessed four popular extraction tools using our proposed evaluation framework. The proposed framework can be applied as a useful benchmark when developers want to build ontology extraction tools.
1 Introduction In recent years, the Semantic Web receives great attention as the next generation Web. Adding semantics to the Web contents, it makes the Web meaningful, understandable, and machine-processable [1]. Ontologies are considered as a key component of the Semantic Web since it is the backbone of knowledge representation for this Web. Ontologies have been studied by many researchers in various fields, and, in particular, there has been a widespread use and application of ontology for data annotation, data integration, and intelligent system development. Although there have been many efforts to develop ontologies, several issues still need to be resolved in order to construct ontology effectively and efficiently. First of all, building ontology is a time-consuming activity which requires many resources. This is an ontology bottleneck problem, which results from the lack of a faster and efficient way to build ontology. Recently, several studies attempted to resolve this problem, and, as a result, many tools were developed to reduce the ontology construction time. These tools can be broadly classified into ontology editing tools, ontology merging tools, and ontology extraction tools. Ontology editing tools (e.g., protégé) help the ontologist to acquire, organize, and visualize the domain knowledge before and during the building ontology. Ontology merging tools (e.g., PROMPT) can be used to create single coherent ontology by unifying two or more existing ontologies. Ontology extraction tools (e.g., Text-To-Onto) support automatic creation R. Meersman, Z. Tari, P. Herrero et al. (Eds.): OTM 2007 Ws, Part I, LNCS 4805, pp. 511–521, 2007. © Springer-Verlag Berlin Heidelberg 2007
512
J. Park, W. Cho, and S. Rho
of ontologies by applying some techniques, such as natural language processing or machine learning. The second issue is related with tool evaluation. Several evaluation frameworks have been proposed to measure the performance of ontology editing tools [14] and ontology merging tools [12]. Although the ontology extraction tools are considered as a fundamental and best solution to resolve the ontology bottleneck problem when building an ontology from scratch, there is no evaluation framework for ontology extraction tools. The last issue
Data Loading...
