Designing and evaluating patterns for relation acquisition from texts with Caméléon

Nathalie Aussenac-Gilles

Marie-Paule Jacques

Pattern-based approaches for knowledge identification in texts assume that linguistic regularities always characterise the same kind of knowledge, such as semantic relations. In this paper, we report the experimental evaluation of a large set of patterns using an ontology enrichment tool: Caméléon. Results emphasize the strong influence of the corpus on pattern efficiency and on their meaning. This influence confirms two of the hypotheses that motivated to define Caméléon as a support used in a human-driven process: (1) patterns and relations must be adapted to each project; (2) human interpretation is required to decide how to report the pieces of knowledge identified with patterns in the ontology.

Keywords: knowledge patterns for French, ontology engineering from text, knowledge engineering, conceptual relation, semi-automatic information extraction, terminology structuring, tool evaluation, dependence on textual genre and domain

Published online: 04 August 2008
https://doi.org/10.1075/term.14.1.04aus

Cited by

Cited by 4 other publications

Aussenac-Gilles, Nathalie, Jean Charlet & Chantal Reynaud

2020. In A Guided Tour of Artificial Intelligence Research, ► pp. 733 ff.

Mela, Augusta, Mathieu Roche & Mohamed El Amine Bekhtaoui

2012. In Natural Language Processing and Information Systems [Lecture Notes in Computer Science, 7337], ► pp. 366 ff.

Quesada-Martínez, Manuel, Jesualdo Tomás Fernández-Breis, Robert Stevens & Nathalie Aussenac-Gilles

2015. In Knowledge Engineering and Knowledge Management [Lecture Notes in Computer Science, 8982], ► pp. 172 ff.

Sambre, Paul & Cornelia Wermuth

2015. In Handbook of Terminology [Handbook of Terminology, 1], ► pp. 101 ff.

This list is based on CrossRef data as of 07 february 2022. Please note that it may not be complete. Sources presented here have been supplied by the respective publishers. Any errors therein should be reported to them.