Integrating knowledge from DEX hierarchies into a logistic regression stacking model for predicting ski injuries. (15th May 2018)

• Record Type: Journal Article
• Authors: Delibašić, Boris; Radovanović, Sandro; Jovanović, Miloš; Bohanec, Marko; Suknović, Milija
• Is Part Of: Journal of decision systems. Volume 27(2018)Supplement 1
• Page Start: 201
• Page End: 208
• Publication Date: 2018-05-15
• Subjects: Domain knowledge -- machine learning -- multi-criteria DEX model -- logistic regression -- stacking -- ski injury prediction; Decision support systems -- Periodicals; Management information systems -- Periodicals; Information resources management -- Periodicals; Information storage and retrieval systems -- Periodicals; Management -- Communication systems -- Periodicals; Decision support systems; Information resources management; Information storage and retrieval systems; Management -- Communication systems; Management information systems; Periodicals; 658.40305
• DOI: 10.1080/12460125.2018.1460164
• Languages: English
• ISSNs: 1246-0125

Abstract

Abstract: Machine learning models are often unaware of the structure that exists between attributes. Expert models, on the other hand, provide structured knowledge that is readily available, yet not often used in machine learning. This paper proposes the integration of expert knowledge, represented in the form of multi-criteria DEX (Decision EXpert) hierarchies or attributes, in a logistic regression stacking framework. We show that integrating expert knowledge into a machine learning framework can improve the quality of models. We tested our hypothesis on the problem for predicting ski injury occurrence, an important decision-support task in ski-resort management. Our results suggest that using a DEX hierarchy of attributes and stacking improves the AUC (area under the curve) compared to logistic regression models unaware of the DEX hierarchy from 1 to 4%.