Communication Dans Un Congrès Année : 2024

A new incremental pipeline for concept formation driven by prior knowledge: Application on the AI Act domain

Résumé

In the Ontology Learning research domain, despite recent advancements, the performance of current non or semi-supervised approaches for concept formation remains sub-optimal, particularly from a single, small-sized corpus for a specialized domain. In order to answer the performance drawback, this paper introduces a novel pipeline, called CO-ISSC (Core Ontology-based Incremental Semi-Supervised Clustering), for concept formation towards ontology learning. This pipeline uses a PLM (Pre-trained Language Model) and combines in an incremental manner a semi-supervised dimension reduction technique and a clustering technique, guided by core concepts as prior knowledge to align results with the ontology domain. Its incremental nature enhances prior knowledge and boosts its performance. The CO-ISSC pipeline’s performance is evaluated on the recent and significant AI Act text established by the European Union, which aims to ensure the safety, transparency, and non-discrimination of AI systems. To this end, we manually built a benchmark terminology for the AI Act domain given that no reference model exists yet. The results demonstrate promising performance of the CO-ISSC pipeline, outperforming baseline non-supervised or semi-supervised approaches such as DBSCAN, similarity measure based approaches, SVM and ANN.

Fichier principal
Vignette du fichier
1-s2.0-S1877050924026711-main.pdf (588.89 Ko) Télécharger le fichier
Origine Publication financée par une institution
Licence

Dates et versions

hal-04772437 , version 1 (08-09-2025)

Licence

Identifiants

Citer

Hongtao Ling, Mounira Harzallah, Margo Bernelin, Claudia Marinica, Patricia Serrano-Alvarado. A new incremental pipeline for concept formation driven by prior knowledge: Application on the AI Act domain. The 28th International Conference on Knowledge-Based and Intelligent Information & Engineering Systems (KES2024), KES international, Sep 2024, Seville, Spain. pp.2148-2157, ⟨10.1016/j.procs.2024.09.618⟩. ⟨hal-04772437⟩
540 Consultations
110 Téléchargements

Altmetric

Partager

  • More