Institutional Repository of Vilnius University
Draudimo žalų sukčiavimo atvejų nustatymas, naudojant mašininio mokymosi metodus /
Abstract
dc:descriptionhe main aim of this study is to identify the most commonly used machine learning methods for detecting insurance claims fraud and apply them to the given data. Main objectives: conduct a literature analysis, prepare a dataset for the application of machine learning methods, implement chosen techniques, evaluate the classifiers performance and visualize the results. In this study data about insurance claims, registered in 2023 for physical persons and related to motor accidents in Lithuania is used. The suggested approach includes data preparation, natural language preprocessing and application of machine learning methods: k-means clustering, Random forest and Extreme Gradient Boosting classification. To address unbalanced data issue the paper proposes use of data balancing techniques, such as SMOTE or RUS.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2024
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Lankelytė, Ineta,
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD210643494&prefLang=en_US