C4.5: programs for machine learning
C4.5: programs for machine learning
Machine Learning
The Random Subspace Method for Constructing Decision Forests
IEEE Transactions on Pattern Analysis and Machine Intelligence
Neural Networks: A Comprehensive Foundation
Neural Networks: A Comprehensive Foundation
On Bias, Variance, 0/1—Loss, and the Curse-of-Dimensionality
Data Mining and Knowledge Discovery
A Tutorial on Support Vector Machines for Pattern Recognition
Data Mining and Knowledge Discovery
ICPR '98 Proceedings of the 14th International Conference on Pattern Recognition-Volume 1 - Volume 1
ICDAR '95 Proceedings of the Third International Conference on Document Analysis and Recognition (Volume 1) - Volume 1
A brief introduction to boosting
IJCAI'99 Proceedings of the 16th international joint conference on Artificial intelligence - Volume 2
SVMs modeling for highly imbalanced classification
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics - Special issue on human computing
Learn++.MF: A random subspace approach for the missing feature problem
Pattern Recognition
Feature generation using genetic programming with application to fault classification
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
| Hi-index | 12.05 |
The Chlor-Alkali production is one of the largest industrial scale electro-synthesis in the world. Plants with more than 1000 individual reactors are common, where chlorine and hydrogen are only separated by 0.2mm thin membranes. Wrong operating conditions can cause explosions and highly toxic gas releases, but also irreversible damages of very expensive cell components with dramatic maintenance costs and production loss. In this paper, a Multi-Expert System based on first-order logic rules and Decision Forests is proposed to detect any abnormal operating conditions of membrane cell electrolyzers and to advice the operator accordingly. Robustness to missing data - which represents an important issue in industrial applications in general - is achieved by means of a Dynamic Selection strategy. Experiments performed with real-world electrolyzer data indicate that the proposed system can significantly detect the different operating modes, even in the presence of high levels of missing data - or ''wrong'' data, as a consequence of maloperation -, which is essential for precise fault detection and advice generation.