Tools
Tools
Bhattacharya, Jinia (2025) Dating Application Fraud Profile Detection and Analysis using Data Mining. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (1MB) | Preview |
Preview |
PDF (Configuration Manual)
Download (1MB) | Preview |
Abstract
With the popularity that online dating application has gained, safety of users has been compromised by the increase of fraudulent activities such as catfishing, identity theft, and financial scams. This work focuses on detecting and analysing fraudulent profiles using data mining and machine learning techniques. The K-Means clustering method of unsupervised learning techniques is used in this study to detect anomalies in user profiles. Text-mining approaches like sentiment analysis and topic modelling are employed to identify the deception patterns in the detected anomalies within the profile. In addition, Support Vector Machine (SVM) as classification models are used to forecast fraudulent profiles. Experimental results show that combining clustering, sentiment analysis, and classification is more accurate at detecting fraud resulting in higher precision in SVM. Future aspects will aim to improve the class-balance and integrate advanced NLP models including real-time datasets can make fraud detection in dating applications reliable.
Actions (login required)
 |
View Item |