TY - GEN
T1 - Convolutional Neural Networks for Structured Industrial Data
AU - Moles, Luis
AU - Boto, Fernando
AU - Echegaray, Goretti
AU - Torre, Iván G.
N1 - Publisher Copyright:
© 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.
PY - 2023
Y1 - 2023
N2 - Regression methods aim to predict a numerical value of a target variable given some input variables by building a function f: Rn→ R. In Industry 4.0 regression tasks, tabular data-sets are especially frequent. Decision Trees, ensemble methods such as Gradient Boosting and Random Forest, or Support Vector Machines are widely used for regression tasks with tabular data. However, Deep Learning approaches are rarely used with this type of data, due to, among others, the lack of spatial correlation between features. Therefore, in this research, we propose two Deep Learning approaches for working with tabular data. Specifically, two Convolutional Neural Networks architectures are tested against different state of the art regression methods. We perform an hyper-parameter tuning of all the techniques and compare the model performance in different industrial tabular data-sets. Experimental results show that both Convolutional Neural Network approaches can outperform the commonly used methods for regression tasks.
AB - Regression methods aim to predict a numerical value of a target variable given some input variables by building a function f: Rn→ R. In Industry 4.0 regression tasks, tabular data-sets are especially frequent. Decision Trees, ensemble methods such as Gradient Boosting and Random Forest, or Support Vector Machines are widely used for regression tasks with tabular data. However, Deep Learning approaches are rarely used with this type of data, due to, among others, the lack of spatial correlation between features. Therefore, in this research, we propose two Deep Learning approaches for working with tabular data. Specifically, two Convolutional Neural Networks architectures are tested against different state of the art regression methods. We perform an hyper-parameter tuning of all the techniques and compare the model performance in different industrial tabular data-sets. Experimental results show that both Convolutional Neural Network approaches can outperform the commonly used methods for regression tasks.
UR - http://www.scopus.com/inward/record.url?scp=85141692719&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-18050-7_35
DO - 10.1007/978-3-031-18050-7_35
M3 - Conference contribution
AN - SCOPUS:85141692719
SN - 9783031180491
T3 - Lecture Notes in Networks and Systems
SP - 361
EP - 370
BT - 17th International Conference on Soft Computing Models in Industrial and Environmental Applications, SOCO 2022, Proceedings
A2 - García Bringas, Pablo
A2 - Pérez García, Hilde
A2 - Martinez-de-Pison, Francisco Javier
A2 - Villar Flecha, José Ramón
A2 - Troncoso Lora, Alicia
A2 - de la Cal, Enrique A.
A2 - Herrero, Alvaro
A2 - Martínez Álvarez, Francisco
A2 - Psaila, Giuseppe
A2 - Quintián, Héctor
A2 - Corchado Rodriguez, Emilio S.
PB - Springer Science and Business Media Deutschland GmbH
T2 - 17th International Conference on Soft Computing Models in Industrial and Environmental Applications, SOCO 2022
Y2 - 5 September 2022 through 7 September 2022
ER -
