A propositionalization method of multi-relational data based on Grammar-Guided Genetic Programming

• Published in: Expert systems with applications Vol. 168; p. 114263
• Main Authors: Quintero-Domínguez, Luis A., Morell, Carlos, Ventura, Sebastián
• Format: Journal Article
• Language: English
• Published: New York Elsevier Ltd 15.04.2021; Elsevier BV
• Subjects: Bag-of-Words; Boolean; Boolean algebra; Classification; G3Prop; Genetic algorithms; Grammar-Guided Genetic Programming; Machine learning; Multi-Relational Data Mining; Propositionalization; Representations; G3Prop; Propositionalization; Bag-of-Words; Multi-Relational Data Mining; Grammar-Guided Genetic Programming
• ISSN: 0957-4174; 1873-6793
• DOI: 10.1016/j.eswa.2020.114263

The propositionalization process tries to find distinctive features of the examples in a database to transform such relational data into a simpler representation. More informative features have a positive impact on the classification capabilities of the learning algorithms. In this work, we propose a new propositionalization method, which generates complex Boolean attributes using Grammar-Guided Genetic Programming (G3P). The generated attributes are compound formulas that combine word items coming from a Bag-of-Words (BoW) representation using Boolean operators. The proposal was assessed against three state-of-the-art simple-instance and multiple-instance propositionalization methods. The experimental results show that the proposed method achieves an improvement in terms of classification accuracy and a considerable reduction in the dimensionality of the resulting datasets. •Grammar-Guided Genetic Programming is used to generate complex attributes.•Words coming from a Bag-of-Words representation are combined using Boolean operators.•Simple-instance and multiple-instance propositionalization were analyzed.•A considerable reduction in the dimensionality of the resulting datasets is achieved.