Institutional Repository of Vilnius University
Classifying tax evaders by means of social network analytics /
Abstract
dc:descriptionThis thesis uses Logistic Regression, Decision Tree and Random Forest methods to classify companies into tax evading firms and tax compliant firms. Based on Lithuanian VAT law two relationships between persons are important – ownership and kinship. Based on these relationships’ networks are then constructed for each year in a dataset. From each of these networks their network features are then extracted. These include - network size, community structure and others. Regarding the model accuracy of all the models, it is clear that Random Forest model produced best results in all examined models. If these results would persist with different dataset, Random Forest model with AUC score around 0.7 could be useful when selecting potential auditing targets for Lithuanian State Tax Inspectorate.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2022
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Daraganas, Lukas,
Subjects
dc:subject × 1Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- eng
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD146235320&prefLang=en_US