RT Conference Proceedings
T1 Machine learning and fuzzy measures: a real approach to individual classification
A1 Gutiérrez García-Pardo, Inmaculada
A1 Santos, Daniel
A1 Castro Cantalejo, Javier
A1 Hernández-Gonzalo, Julio Alberto
A1 Gómez González, Daniel
A1 Espínola Vílchez, María Rosario
AB In the field of machine learning, a crucial task is understanding the relative importance of the different input features in a predictive model. There is an approach in the literature whose aim is to analyze the predictive capacity of some features with respect to others. Can we explain a feature of the input space with others? Can we quantify this capacity? We propose a practical approach for analyzing the importance of features in a model and the explanatory capacity of some features over others. It is based on the adaptation of existing definitions from the literature that use the Shapley value and fuzzy measures. Our new approach aims to facilitate the understanding and application of these concepts by starting from a simple idea and considering well known methods. The main objective of this work is to provide a useful and practical approach for analyzing feature importance in real world cases.
SN 9783031399640
SN 0302-9743
YR 2023
FD 2023
LK https://hdl.handle.net/20.500.14352/107836
UL https://hdl.handle.net/20.500.14352/107836
LA eng
NO Gutiérrez, I. et al. (2023) «Machine Learning and Fuzzy Measures: A Real Approach to Individual Classification», en Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Springer Science and Business Media Deutschland GmbH, pp. 137-148. Disponible en: https://doi.org/10.1007/978-3-031-39965-7_12
NO Colección de libros: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (14069 LNCS)
NO Gobierno de España. Secretaría de Estado de Investigacion, Desarrollo e Innovacion
DS Docta Complutense
RD 30 sept 2024