Back to search

Institutional Repository of Vilnius University

Identification of fraudulent messages using artificial intelligence methods /

Abstract

dc:description

The wide spread use and high engagement rate of SMS-based (Short Message Service) communication has made it a prime target for fraudulent activities, such as smishing (SMS phishing), spam, and social engineering attacks with increasing sophistication in methods used. This aim of this research is to develop a machine-learning based fraud detection system, "Cybrr". The study systematically evaluates five distinct machine learning approaches - Random Forest, Support Vector Machines, Naive Bayes, K-Nearest Neighbors, and AdaBoost - through different relevant metrics. The models were trained on a newly created, balanced dataset consisting of 83,886 records, which was created by merging four existing datasets, applying data processing, tokenization and augmentation techniques to balance the dataset and improve the performance of the models. The results of the conducted experiments demonstrate that Support Vector Machine (SVM) achieves the best results in terms of various metrics, with an accuracy of 97% at a cost of 83 times longer training time compared to Random Forest. Random Forest achieves the best results in terms of balance between the evaluation metrics and the computational demands, with 96% accuracy and reasonable training time and resource requirements. The entire model training code responsible for model training and data processing is written in Python. The created system involves a modular architecture integrating FastAPI backend services with a React frontend, which enables users to select preferred detection algorithms and process both text and image inputs and receive real-time feedback on the classification results.

Degree

thesis:*
Grantor dc:publisher
Institutional Repository of Vilnius University
Year dc:date
2025

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Survila, Kernius,

Subjects

dc:subject × 1

Rights

dc:rights
Statement dc:rights
  • info:eu-repo/semantics/openAccess
Language dc:language
eng

Identifiers

dc:identifier.*

Chain of custody

source
Harvested from
Vilnius University
Base URL
epublications.vu.lt/oai
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Survila, Kernius,. Identification of fraudulent messages using artificial intelligence methods /. Institutional Repository of Vilnius University, 2025. https://repository.vu.lt/VU:ELABAETD229556706&prefLang=en_US