FAWEHINMI, OLUMIDE ABIMBOLA and Covenant University, Theses (2018) HYBRID CREDIT CARD FRAUD DETECTION USING ANOMALY DETECTION AND GENETIC ALGORITHM. Masters thesis, COVENANT UNIVERSITY.
![[img]](http://eprints.covenantuniversity.edu.ng/style/images/fileicons/other.png) |
PDF
Download (171Kb) |
Abstract
The introduction of electronic (e-payment) technologies via credit card has significantly developed banking and other financial sector. This development has brought about reduction in the size of long queues and also the time mostly taken by customers in making and payment transaction. In spite of the growing trend of e-payment, the financial transaction has been marred with fraud This dissertation presents a machine learning based Hybrid Credit Card Fraud (HCCFD) model which uses anomy detection technique by applying multivariate normal distribution and genetic algorithm to detect fraudulent transaction on a credit card. HCCFD was trained using an imbalance dataset of credit card transaction with thirty thousand (30,000) observations each having 19 features and a target variable which indicate if a transaction is fraudulent or not. The dataset were preprocessed by normalizing it and then was subdivided into three (3) in the orders of 70% for training set, 15% for validation set and 15% testing set. The training set was used to compute Mean Vector and covariate matrix of the dataset while validation set was used by genetic algorithm to set the threshold between fraudulent and non-fraudulent transactions. The test set was used to evaluate the performance of HCCFD and it was found that HCCFD significantly outperformed its counterparts ( Support Vector Machine (SVM), Decision Tree and Artificial Neural Network (ANN)) trained on the same training set in terms of prediction accuracy and prediction time.
| Item Type: | Thesis (Masters) |
|---|---|
| Subjects: | Q Science > Q Science (General) Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software |
| Divisions: | Faculty of Engineering, Science and Mathematics > School of Electronics and Computer Science |
| Depositing User: | Mrs Hannah Akinwumi |
| Date Deposited: | 12 Mar 2020 10:29 |
| Last Modified: | 12 Mar 2020 10:29 |
| URI: | http://eprints.covenantuniversity.edu.ng/id/eprint/13199 |
Actions (login required)
 |
View Item |
