Optimasi Parameter DBSCAN menggunakan Metode Differential Evolution untuk Deteksi Anomali pada Data Transaksi Bank

Rifqi Karunia Ibadirachman; Yulison Herry Chrisnanto; Puspita Nurul Sabrina

doi:10.54914/jtt.v10i1.1189

Penulis

Rifqi Karunia Ibadirachman Universitas Jendral Achmad Yani
Yulison Herry Chrisnanto Universitas Jendral Achmad Yani
Puspita Nurul Sabrina Universitas Jendral Achmad Yani

DOI:

https://doi.org/10.54914/jtt.v10i1.1189

Kata Kunci:

Anomali, Data Transaksi Bank, DBSCAN, Differential Evolution, Optimasi

Abstrak

Anomali dalam data transaksi bank sering kali menandakan adanya aktivitas penipuan atau kesalahan. Penelitian ini bertujuan mendeteksi anomali pada data transaksi bank dengan optimasi parameter DBSCAN menggunakan metode Differential Evolution (DE) karena terdapat kekurangan yaitu sulitnya menentukan parameter yang tepat untuk membuat cluster yang tepat agar dapat mendeteksi anomali pada data transaksi bank dengan baik. Data yang digunakan adalah data transaksi dari Bank XYZ dengan lebih dari 1011 record data. Tahapan penelitian meliputi pengumpulan data, preprocessing data (pembersihan, normalisasi, dan transformasi data), perancangan sistem, implementasi algoritma, serta analisa dan pengujian menggunakan metode Silhouette score dan Z-score . Metode DE digunakan untuk menentukan parameter optimal MinPts dan Epsilon secara otomatis. Hasil penelitian menunjukkan bahwa penggunaan DE dapat menghasilkan parameter yang optimal, dengan peningkatan akurasi deteksi anomali menggunakan DBSCAN. Evaluasi dengan Silhouette score menunjukkan akurasi rata-rata 0.7916 dan menggunakan DBI mencapai 0.19791 paling rendah, sementara pengukuran Z-score dan MSE menunjukkan kepadatan cluster yang tinggi dengan akurasi deteksi anomali mencapai 98.41% dan 0.555537. Pendekatan DE untuk pemilihan parameter ini efektif dalam meningkatkan performa DBSCAN dalam mendeteksi anomali pada data transaksi bank. Saran untuk penelitian selanjutnya adalah meningkatkan jumlah record data dan lakukan percobaan pada variasi atribut data yang lebih beragam.

Unduhan

Data unduhan belum tersedia.

Referensi

P. Jain, M. S. Bajpai, and R. Pamula, “A Modified DBSCAN Algorithm for Anomaly Detection in Time-series Data with Seasonality,” International Arab Journal of Information Technology, vol. 19, no. 1, pp. 23–28, Jan. 2022, doi: 10.34028/iajit/19/1/3.

M. Tatusch, G. Klassen, M. Bravidor, and S. Conrad, “Predicting Erroneous Financial Statements Using a Density-Based Clustering Approach,” in ACM International Conference Proceeding Series, Association for Computing Machinery, Aug. 2020, pp. 89–94. doi: 10.1145/3418653.3418673.

I. Kamil and B. Pharmasetiawan, “Fingerprint Presence Fraud Detection Using Tight Clustering on Employee’s Presence and Activity Data,” 2019. doi: DOI: 10.1109/ICITISEE48480.2019.9003914.

Mohamad Zamini and Gholamali Montazer, Credit Card Fraud Detection using autoencoder based clustering. 2018 9th International Symposium on Telecommunications (IST’2018), 2018. doi: DOI: 10.1109/ISTEL.2018.8661129.

M. Yang and D. Ergu, “Anomaly Detection of Vehicle Data Based on LOF Algorithm,” Frontiers in Signal Processing, vol. 4, no. 1, Jan. 2020, doi: 10.22606/fsp.2020.41007.

D. Deng, “Research on Anomaly Detection Method Based on DBSCAN Clustering Algorithm,” in Proceedings - 2020 5th International Conference on Information Science, Computer Technology and Transportation, ISCTT 2020, Institute of Electrical and Electronics Engineers Inc., Nov. 2020, pp. 439–442. doi: 10.1109/ISCTT51595.2020.00083.

P. Jain, M. S. Bajpai, and R. Pamula, “A Modified DBSCAN Algorithm for Anomaly Detection in Time-series Data with Seasonality,” International Arab Journal of Information Technology, vol. 19, no. 1, pp. 23–28, Jan. 2022, doi: 10.34028/iajit/19/1/3.

Z. Zhang, L. Chen, Q. Liu, and P. Wang, “A Fraud Detection Method for Low-Frequency Transaction,” IEEE Access, vol. 8, pp. 25210–25220, 2020, doi: 10.1109/ACCESS.2020.2970614.

M. Z. Hossain, M. J. Islam, M. W. R. Miah, J. H. Rony, and M. Begum, “Develop a dynamic DBSCAN algorithm for solving initial parameter selection problem of the DBSCAN algorithm,” Indonesian Journal of Electrical Engineering and Computer Science, vol. 23, no. 3, pp. 1602–1610, Sep. 2021, doi: 10.11591/ijeecs.v23.i3.pp1602-1610.

K. I. Ghathwan and A. J. Mohammed, “Intelligent Bat Algorithm for Finding Eps Parameter of DbScan Clustering Algorithm,” Iraqi Journal of Science, vol. 63, no. 12, pp. 5572–5580, 2022, doi: 10.24996/ijs.2022.63.12.41.

G. Dominico and R. S. Parpinelli, “Multiple global optima location using differential evolution, clustering, and local search,” Appl Soft Comput, vol. 108, Sep. 2021, doi: 10.1016/j.asoc.2021.107448.

M. T. Guerreiro et al., “Anomaly detection in automotive industry using clustering methods—a case study,” Applied Sciences (Switzerland), vol. 11, no. 21, Nov. 2021, doi: 10.3390/app11219868.

Manoj Kumar Reddy Mallidi and Yeshwanth Zagabathuni, “Analysis of Credit Card Fraud detection using Machine Learning models on balanced and imbalanced datasets,” International Journal of Emerging Trends in Engineering Research, vol. 9, no. 7, pp. 846–852, Jul. 2021, doi: 10.30534/ijeter/2021/02972021.

H. Sayed Ramadan, H. Amin Maghawry, M. El-Eleamy, and K. El-Bahnasy, “A Heuristic Novel Approch For Determination Of Otimal Epsilon For DBSCAN Clustering Algorithm,” J Theor Appl Inf Technol, vol. 15, no. 7, 2022, [Online]. Available: www.jatit.org

N. Gholizadeh, H. Saadatfar, and N. Hanafi, “K-DBSCAN: An improved DBSCAN algorithm for big data,” Journal of Supercomputing, vol. 77, no. 6, pp. 6214–6235, Jun. 2021, doi: 10.1007/s11227-020-03524-3.

H. Liu, Y. Wang, and W. G. Chen, “Anomaly detection for condition monitoring data using auxiliary feature vector and density-based clustering,” IET Generation, Transmission and Distribution, vol. 14, no. 1, pp. 108–118, Jan. 2020, doi: 10.1049/iet-gtd.2019.0682.

X. Olive and L. Basora, “Identifying Anomalies in past en-route Trajectories with Clustering and Anomaly Detection Methods,” 2019. [Online]. Available: https://hal.science/hal-02345597