The Problem of Learning Non-taxonomic Relationships of Ontologies from Text

Abstract

Manual construction of ontologies by domain experts and knowledge engineers is a costly task. Thus, automatic and/or semi-automatic approaches to their development are needed. Ontology Learning aims at identifying its constituent elements, such as non-taxonomic relationships, from textual information sources. This article presents a discussion of the problem of Learning Non-Taxonomic Relationships of Ontologies and defines its generic process. Four techniques representing the state of the art of Learning Non-Taxonomic Relationships of Ontologies are described and the solutions they provide are discussed along with their advantages and limitations.

PARNT: A Statistic based Approach to Extract Non-Taxonomic Relationships of Ontologies from Text

“…The solutions adopted by PARNT for each phase of the generic process of LNTRO [9] illustrated in Fig. 1 are summarized in Table 1 and described in sections A to E. A. Corpus construction PARNT does not define a specific solution to be adopted in this phase and the specialist is the one responsible for choosing the one that best suits the needs for that situation.…”

“…1 [9]. The corpus construction task consists in selecting documents of the domain from which relationships can be extracted.…”

Learning Non-Taxonomic Relations on Demand for Ontology Extension

“…\Category: Songs by artist"-\Category: The Beatles songs", with \artist" as property and \The Beatles" as value). Recent work related to non-taxonomic relations learning includes Mohamed et al [23] and Serra et al [24]. Mohamed et al [23] propose an approach for automatically discovering relevant relations, given a large text corpus plus an initial ontology de¯ning hundreds of noun categories (e.g.…”

“…Athlete, Musician and Instrument). Serra et al [24] describe three representative techniques for non-taxonomic relations learning and discuss their advantages and limitations.…”

Reviewing the Problem of Learning Non-Taxonomic Relationships of Ontologies From Text

“…Shared, reflects the notion that an ontology captures consensual knowledge, that is, it's not private to some individual but accepted by a group. Currently, ontologies are applied in areas such as communication of software agents [11], integration of information [2], composition of Web Services [23], knowledge management [17], description of contents to support information retrieval from textual sources [14] [16], in Semantic Web applications [3] and knowledge-based systems [6].…”