Wirtschaft (MSB)
-
Wirtschaft
(2)
Refine
Year
Publication Type
- Lecture (212)
- Article (130)
- Part of a Book (103)
- Report (65)
- Book (64)
- Conference Proceeding (39)
- Contribution to a Periodical (5)
Language
- English (618) (remove)
Keywords
- European (3)
- business (3)
- university (3)
- university-business cooperation (3)
- Networking (2)
- Process-Driven Application (2)
- University-Business Cooperation (2)
- 3D printing (1)
- Academic Entrepreneurship (1)
- Activ Investor (1)
Faculty
A major requirement for credit scoring models is to provide a maximally accurate risk prediction. Additionally, regulators demand these models to be transparent and auditable. Thus, in credit scoring, very simple predictive models such as logistic regression or decision trees are still widely used and the superior predictive power of modern machine learning algorithms cannot be fully leveraged. Significant potential is therefore missed, leading to higher reserves or more credit defaults. This paper works out different dimensions that have to be considered for making credit scoring models understandable and presents a framework for making ``black box'' machine learning models transparent, auditable and explainable. Following this framework, we present an overview of techniques, demonstrate how they can be applied in credit scoring and how results compare to the interpretability of score cards. A real world case study shows that a comparable degree of interpretability can be achieved while machine learning techniques keep their ability to improve predictive power.
In the context of Continuous Software Engineering, it is acknowledged as best practice to develop new features on the mainline rather than on separate feature branches. Unfinished work is then usually prevented from going live by some kind of feature toggle. However, there is no concept of feature toggles for Process-Driven Applications (PDA) so far. PDAs are hybrid systems consisting not only of classical source code but also of a machine-interpretable business process model. This paper elaborates on a feature development approach that covers both the business process model and the accompanying source code artifacts of a PDA. The proposed solution, Toggles for Process-Driven Applications (T4PDA), equipped with an easy to use modeling tool extension, enables the developer to safely commit unfinished work on model and source code to the project’s mainline. It will be kept inactive during productive deployments unless the feature is finally released. During an AB/BA crossover design experiment, the T4PDA approach, including the provided tool support, showed higher software quality, a faster development process, and contented developers.