CNN-BiLSTM ve Momentum ile Optimize Edilmiş SGD Kullanılarak Gelişmiş Mobil Para Dolandırıcılığı Tespiti

Year 2025, Volume: 16 Issue: 3, 207 - 231, 31.08.2025

Niamatu Yussif , Kate Takyi , Rose-mary Owusuaa Mensah Gyening , Samuelson Israel Boadu-acheampong

https://doi.org/10.5824/ajite.2025.03.002.x

Abstract

Mobil para sistemlerinin hızla benimsenmesi, dolandırıcılık faaliyetlerinde önemli bir artışa yol açarak güvenliklerini ve itibarlarını tehlikeye atmıştır. Bu araştırma, Stokastik Gradyan İnişi (SGD) kullanarak momentumla bir CNN-BiLSTM modelini değiştirerek mobil para dolandırıcılığını tespit etmek için gelişmiş bir yöntem sunmaktadır. Önceden işlenmiş bir hibrit CNN-BiLSTM modeli kullanarak işlem verilerinden belirgin özellikleri hesapladık ve modeli coğrafi ve zamansal yönleri içeren verilerdeki eğilimleri belirlemek üzere eğittik. Model, endüstri standardı test yaklaşımlarını kullanarak dikkat çekici bir performans gösterdi: 0.9928'lik bir F1 puanı, 0.9927'luk bir hassasiyet, 0.9928'lık bir doğruluk ve 0.9929'lık bir geri çağırma. Önerilen model, sahtekârlığı tespit edebilir ve düşük bir yanlış pozitif oranına sahiptir. Çalışmaya göre, model özellik seçimini iyileştirir ve çeşitli optimizasyon tekniklerini birleştirerek onu daha esnek ve farklı mobil para sistemleri için uygun hale getirir.

Keywords

Sahtecilik Tespiti , Makine Öğrenmesi , Sinir Ağları , Stokastik Gradyan İnişi

Advanced Mobile Money Fraud Detection Using CNN-BiLSTM and Optimized SGD with Momentum

Abstract

The accelerated adoption of mobile money systems has significantly increased fraudulent activity, compromising their security and trustworthiness. This research presents an enhanced method for detecting mobile money fraud by modifying a CNN-BiLSTM model with momentum using Stochastic Gradient Descent (SGD). We computed salient features from transaction data using a pre-processed hybrid CNN-BiLSTM model and trained the model to identify trends in the data that included geographical and temporal aspects. The model performed remarkably using industry-standard testing approaches: an F1 score of0.9928, precision of 0.9927, accuracy of 0.9928, and recall of 0.9929. The proposed model can identify dishonesty and has a low false positive rate. According to the study, the model improves feature selection and incorporates various optimization techniques, making it more flexible and suitable for different mobile money systems.

Keywords

Fraud Detection , Machine Learning , Neural Networks , Stochastic Gradient Descent

References

• Agarwal, A., Iqbal M., Mitra, B., Kumar, V., & Lal, N.( 2021). “Hybrid CNN-BILSTM-attention based identification and prevention system for banking transactions,” … Essent. OILS J. …, vol. 8, no. 5, pp. 2552–2560, [Online]. Available: http://www.nveo.org/index.php/journal/article/view/809
• Alghofaili, Y., Albattah, A., & Rassam, M. A. (2020). “A Financial Fraud Detection Model Based on LSTM Deep Learning Technique,” J. Appl. Secur. Res., vol. 15, no. 4, pp. 498–516, https://doi.org/10.1080/19361610.2020.1815491.
• Almazroi, A. A., & Ayub, N. (2023) “Online Payment Fraud Detection Model Using Machine Learning Techniques,” IEEE Access, vol. 11, no. November, pp. 137188–137203, https://doi.org/10.1109/ACCESS.2023.3339226.
• Botchey, F. E., Qin, Z., Hughes-Lartey, K., & Ampomah, K. E.( 2021). “Predicting Fraud in Mobile Money Transactions using Machine Learning: The Effects of Sampling Techniques on the Imbalanced Dataset,” Inform., vol. 45, no. 7, pp. 45–56, https://doi.org/10.31449/inf.v45i7.3179.
• Chatterjee, M., & Namin, A. S. (2019). “Detecting Phishing Websites through Deep Reinforcement Learning.” Proceedings - International Computer Software and Applications Conference 2(1): 227–32.
• Chang, V., Doan, L. M. T., Di Stefano, A., Sun, Z., & Fortino, G. (2022). “Digital payment fraud detection methods in digital ages and Industry 4.0,” Comput. Electr. Eng., vol. 100, pp. 1–31, https://doi.org/10.1016/j.compeleceng.2022.107734.
• El Kafhali, S., Tayebi, M., & Sulimani, H.( 2024). “An Optimized Deep Learning Approach for Detecting Fraudulent Transactions,” Inf., vol. 15, no. 4, pp. 1–25, https://doi.org/10.3390/info15040227.
• Gibson, S., Issac, B., Zhang, L., & Jacob, S. M. (2020) “Detecting spam email with machine learning optimized with bio-inspired metaheuristic algorithms,” IEEE Access, vol. 8, pp. 187914–187932, https://doi.org/10.1109/ACCESS.2020.3030751.
• Gislain, Z. N. T., & Yurievich, K. E. (2025). Fraud detection using Kolmogorov-Arnold Network, pp. 1-14.
• Hambali Moshood, A.,  “Comparative Analysis of Decision Tree Algorithms for Predicting Undergraduate Students’ Performance in Computer Programming”.
• Igwesi, C. (2023). “Enhancing Authentication and Fraud Detection in Financial Technology,” no. December, 2023.
• Kaggle. (n.d.). Kaggle datasets: Paysim1. Retrieved [4th August, 2023], from https://www.kaggle.com/datasets/ealaxi/paysim1
• Lokanan, M. E. (2023).“Predicting mobile money transaction fraud using machine learning algorithms,” Appl. AI Lett., vol. 4, no. 2, pp. 1–24,  https://doi.org/10.1002/ail2.85.
• Mbunge, E., Makuyana, R., Chirara,  N., & Chingosho, A. (2015). “Fraud Detection in E-Transactions using Deep Neural Networks-A Case of Financial Institutions in Zimbabwe,” Int. J. Sci. Res., vol. 6, no. 9, pp. 2319–7064, https://doi.org/10.21275/ART20176804.
• Networks, C. (2023). “Retracted: Financial Fraud Detection in Healthcare Using Machine Learning and Deep Learning Techniques,” Secur. Commun. Networks, vol. 2023, pp. 1–1, https://doi.org/10.1155/2023/9758612.
• Sun, Q., Tang, T., Chai, H., Wu, J., & Chen, Y. (2021). “Boosting fraud detection in mobile payment with prior knowledge,” Appl. Sci., vol. 11, no. 10, pp. 1–17, https://doi.org/10.3390/app11104347.
• Yang, X., Zhang, C., Sun, Y., Pang, K., Jing, L., Wa, S., & Lv, C. (2023)  “FinChain-BERT: A High-Accuracy Automatic Fraud Detection Model Based on NLP Methods for Financial Scenarios,” Inf., vol. 14, no. 9, pp. 1–26, https://doi.org/10.3390/info14090499.
• Zhang, Z., Chen, L., Liu, Q., & Wang, P. (2020). “A Fraud Detection Method for Low-Frequency Transaction,” IEEE Access, vol. 8, pp. 25210–25220, , https://doi.org/10.1109/ACCESS.2020.2970614.
• Zhdanova, M., Repp, J., Rieke, R., Gaber, C., & Hemery, B. (2014).“No smurfs: Revealing fraud chains in mobile money transfers,” Proc. - 9th Int. Conf. Availability, Reliab. Secur. ARES 2014, pp. 11–20, https://doi.org/10.1109/ARES.2014.10.