Abstract

Relation extraction and classification, collectively referred to as relation discovery, are fundamental tasks in the ontology construction process. In recent years, numerous studies have dedicated their efforts to devising hybrid approaches that merge traditional methods with Machine Learning (ML) or Deep Learning (DL) techniques to tackle relation discovery tasks. This research examines the applicability and efficacy of four distinct models— Support Vector Machine (SVM), Naive Bayes (NB), k-Nearest Neighbors (KNN), and Convolutional Neural Network (CNN)—in relation extraction and classification. Two different datasets, ACL RD-TEC 2.0 and ACLRelAcS, are utilized, alongside two distinct text representation methodologies, BOW and TF-IDF-BOW. An in-depth analysis of the results, including the advantages and limitations of each model, is presented in the results and discussion sections.