Narrow search
Showing 1 - 10 of 1,411
Cover Image
Learning cost sensitive binary classification rules accounting for uncertain and unequal misclassification costs
Rybizki, Lydia - 2014
supervised learning. In addition, we investigate the effects of cost sensitive adaptations in CART and boosting and conclude that …
Persistent link: https://www.econbiz.de/10010334218
Cover Image
Learning cost sensitive binary classification rules accounting for uncertain and unequal misclassification costs
Rybizki, Lydia - 2014
supervised learning. In addition, we investigate the effects of cost sensitive adaptations in CART and boosting and conclude that …
Persistent link: https://www.econbiz.de/10010250581
Cover Image
Performance of credit risk prediction models via proper loss functions
Figiani, Silvia Figini; Maggi, Mario - Dipartimento di Scienze Economiche e Aziendali, … - 2014
recently been proposed that can be seen as refinements of discrimination measures, including variants of the AUC (Area Under … the ROC curve), such as the H index. It is widely recognized that AUC suffers from lack of coherency especially when ROC …
Persistent link: https://www.econbiz.de/10010842822
Cover Image
A boosting method for maximization of the area under the ROC curve
Komori, Osamu - In: Annals of the Institute of Statistical Mathematics 63 (2011) 5, pp. 961-979
Persistent link: https://www.econbiz.de/10009325767
Cover Image
Examining the information content of residuals from hedonic and spatial models using trees and forests
Pace, R. Kelley; Hayunga, Darren - In: The journal of real estate finance and economics 60 (2020) 1/2, pp. 170-180
Persistent link: https://www.econbiz.de/10012226654
Cover Image
Data analytics for non-life insurance pricing
Wüthrich, Mario V.; Buser, Christoph - 2017 - Draft lecture notes, version January 27, 2017
regression trees, bagging, random forest, boosting machines and neural networks. Finally, we provide methodologies for analysing …
Persistent link: https://www.econbiz.de/10011625588
Cover Image
Improved customer choice predictions using ensemble methods
van Wezel, van Wezel, M.C.; Potharst, Potharst, R. - Faculteit der Economische Wetenschappen, Erasmus … - 2005
In this paper various ensemble learning methods from machine learning and statistics are considered and applied to the customer choice modeling problem. The application of ensemble learning usually improves the prediction quality of flexible models like decision trees and thus leads to improved...
Persistent link: https://www.econbiz.de/10010731879
Cover Image
Improved customer choice predictions using ensemble methods
Wezel, M.C. van; Potharst, R. - Erasmus University Rotterdam, Econometric Institute - 2005
In this paper various ensemble learning methods from machine learning and statistics are considered and applied to the customer choice modeling problem. The application of ensemble learning usually improves the prediction quality of flexible models like decision trees and thus leads to improved...
Persistent link: https://www.econbiz.de/10005450872
Cover Image
AUK: a simple alternative to the AUC
Kaymak, Uzay; Ben-David, Ben-David, A.; Potharst, … - Erasmus Research Institute of Management (ERIM), … - 2010
The area under Receiver Operating Characteristic (ROC) curve, also known as the AUC-index, is commonly used for ranking … the performance of data mining models. The AUC has many merits, such as objectivity and ease of interpretation. However … proposed AUK-index compensates for the above basic flaw of the AUC by being sensitive to the class distribution. Therefore it …
Persistent link: https://www.econbiz.de/10010731206
Cover Image
AUK: a simple alternative to the AUC
Kaymak, Uzay; Ben-David, A.; Potharst, R. - Erasmus Research Institute of Management (ERIM), ERIM … - 2010
The area under Receiver Operating Characteristic (ROC) curve, also known as the AUC-index, is commonly used for ranking … the performance of data mining models. The AUC has many merits, such as objectivity and ease of interpretation. However … proposed AUK-index compensates for the above basic flaw of the AUC by being sensitive to the class distribution. Therefore it …
Persistent link: https://www.econbiz.de/10008484125