Machine learning-powered liability determination in motor vehicle accidents
الملخص
The determination of liability in motor vehicle accidents is a complex problem influenced by various dynamic factors. Frequently an initial assessment of liability must be made early in the claims settlement process. However, at this stage, decision-makers regularly find themselves in a situation with incomplete information, which is why a decision must be made under uncertainty. Machine learning can facilitate and improve these decision situations by learning from historical data and thereby generating decision criteria. This master thesis explores how the decision of liability determination in motor vehicle accidents can be improved by machine learning. To achieve this goal, this thesis analyzes to what extent can a machine learning model predict the liability constellation in motor vehicle accidents. An adapted variant of the CRISP-DM methodology was applied to achieve the thesis goal. In the first step, the underlying business problem was examined, and the business goals of the desired solution were defined. In the second step, the data relevant to the solution was collected, processed, and analyzed. In the third step, concrete models were developed. Finally, the third step results were discussed and evaluated with the goals from the first step. The outcomes of this work are two artifacts. On the one hand, a model that is trained to predict the liability constellation. On the other hand, an analysis of how the problem can be solved using different machine learning techniques.
الكلمات المفتاحية
غير متوفر
التخصصات
رسائل وأطاريح مشابهة
- Application of reliability and resilience models to machine learning: a thesis in Computer Engineering Zakaria Faddi · 2023
- Image analysis and machine learning in patients with brain tumors Alice Neimantaite · Gothenburg University · 2026
- Hybrid Machine Learning and Time-Series Modelling to Forecast Product Demand: A CRISP-DM Approach Niklas Virtanen · University of Vaasa · 2026
- Predictive Scheduling of Incoming Goods Using Machine Learning and Deep Learning Methods Johannes Simon Michael Rechenauer · TU Wien · 2026
- Insurance Fraud Detection Using Supervised Machine Learning and Explainable Artificial Intelligence Paavo Laine · University of Turku · 2026
- Predicting refractive index structure parameter in the marine wave boundary layer using machine learning models: a thesis in Data Science Bhavana Chikkamuduvadi Renuka Gowda · 2026
