DETEKSI POLA CANDLESTICK MENGGUNAKAN YOLOV8 UNTUK ANALISIS TEKNIKAL BERBASIS CITRA
Article Sidebar
Main Article Content
Abstract
Penelitian ini mengevaluasi kinerja model YOLOv8 dalam mendeteksi pola candlestick secara otomatis dari citra grafik keuangan. Pendekatan kuantitatif eksperimental digunakan dengan dataset 4.435 citra candlestick yang telah dianotasi dari Roboflow Universe. Model dilatih menggunakan konfigurasi standar YOLOv8 dengan learning rate 0.01, batch size 16, dan 100 epoch. Metrik evaluasi meliputi precision, recall, F1-score, dan mean Average Precision (mAP) pada rentang IoU [0.5:0.95]. Hasil menunjukkan YOLOv8 mencapai akurasi deteksi tinggi dengan precision 0.877, recall 0.898, dan mAP@[0.5:0.95] sebesar 0.903. Model menunjukkan kinerja kuat pada pola dengan ciri visual jelas seperti Bullish Engulfing dan Three Line Strike. Namun, tantangan sistematis muncul dalam membedakan pola dengan kemiripan visual tinggi, khususnya Morning Star–Morning Doji Star dan Evening Star–Evening Doji Star. Confusion matrix ternormalisasi mengungkapkan , model juga mampu membedakan pola yang mirip dengan hasil tertinggi 78% pada pola morning star dan nilai terendah 46% pada pola evening doji star, menunjukkan ambiguitas visual yang melekat pada representasi candlestick. Penelitian ini memberikan benchmark komprehensif untuk YOLOv8 dalam deteksi objek finansial dan menekankan perlunya peningkatan fitur berbasis konteks semantik untuk pasangan pola yang membingungkan.
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.
1. Open Access Policy
Articles published in the journal are open-access articles distributed under the terms and conditions of the Creative Commons license. This aligns with international standards for scholarly publishing, ensuring that all research content is freely available online to the global research community without subscription or paywall barriers immediately upon publication.
2. Creative Commons License
Unless otherwise specified, all accepted and published works are licensed under the Creative Commons Attribution-ShareAlike (CC BY-SA) 4.0 International License.
Under this license, users are free to:
Share: Copy and redistribute the material in any medium or format.
Adapt: Remix, transform, and build upon the material for any purpose, even commercially.
Subject to the following terms:
Attribution: You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
ShareAlike: If you remix, transform, or build upon the material, you must distribute your contributions under the same license as the original.
3. Author Rights & Archiving
Copyright Retention: Authors retain full ownership of the copyright for their published articles without restrictions, granting the journal a non-exclusive license to publish the work as the original publisher of record.
Self-Archiving: Authors are fully permitted to deposit all versions of their manuscript (preprint, accepted manuscript, and final published PDF) into institutional repositories, subject-specific repositories, or on their personal academic websites, provided that proper citation to the journal is maintained.
References
[1] S. Birogul, G. Temur, and U. Kose, “YOLO Object Recognition Algorithm and ‘Buy-Sell Decision’ Model over 2D Candlestick Charts,” IEEE Access, vol. 8, pp. 91894–91915, 2020, doi: 10.1109/ACCESS.2020.2994282.
[2] A. Bochkovskiy, C.-Y. Wang, and H.-Y. M. Liao, “YOLOv4: Optimal Speed and Accuracy of Object Detection,” Apr. 2020, [Online]. Available: http://arxiv.org/abs/2004.10934
[3] R. Padilla, W. L. Passos, T. L. B. Dias, S. L. Netto, and E. A. B. Da Silva, “A comparative analysis of object detection metrics with a companion open-source toolkit,” Electronics (Switzerland), vol. 10, no. 3, pp. 1–28, Feb. 2021, doi: 10.3390/electronics10030279.
[4] M. Liang, S. Wu, X. Wang, and Q. Chen, “A stock time series forecasting approach incorporating candlestick patterns and sequence similarity,” Expert Syst Appl, vol. 205, Nov. 2022, doi: 10.1016/j.eswa.2022.117595.
[5] “Huang (2022) Fast Candlestick Patterns Detection with Limited Training Samples”.
[6] S. Menggunakan, A. Yolo, and G. A. Prasetia, “Pendeteksi Pola Candlestick Chart Pada.”
[7] J.-H. Chen and Y.-C. Tsai, “Dynamic Deep Convolutional Candlestick Learner,” Jan. 2022, [Online]. Available: http://arxiv.org/abs/2201.08669
[8] C.-Y. Wang, A. Bochkovskiy, and H.-Y. M. Liao, “YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors,” Jul. 2022, [Online]. Available: http://arxiv.org/abs/2207.02696
[9] I. Uzun, M. Lobachev, V. Kharchenko, T. Schöler, and I. Lobachev, “Candlestick Pattern Recognition in Cryptocurrency Price Time-Series Data Using Rule-Based Data Analysis Methods,” Computation, vol. 12, no. 7, pp. 1–22, 2024, doi: 10.3390/computation12070132.
[10] P. Foret, A. Kleiner, H. Mobahi, and B. Neyshabur, “Sharpness-Aware Minimization for Efficiently Improving Generalization,” Apr. 2021, [Online]. Available: http://arxiv.org/abs/2010.01412
[11] E. Mienye, N. Jere, G. Obaido, I. D. Mienye, and K. Aruleba, “Deep Learning in Finance: A Survey of Applications and Techniques,” Dec. 01, 2024, Multidisciplinary Digital Publishing Institute (MDPI). doi: 10.3390/ai5040101.
[12] T. D. Nguyen, Q. B. Tran, D. A. Tran, T. D. Than, and Q. D. Tran, “Object Detection Approach for Stock Chart Patterns Recognition in Financial Markets,” in ACM International Conference Proceeding Series, Association for Computing Machinery, Feb. 2023, pp. 150–158. doi: 10.1145/3587828.3587851.
[13] X. Li, Q. Liu, Y. Hu, and H. Liu, “The Double-Layer Clustering Based on K-Line Pattern Recognition Based on Similarity Matching,” Information (Switzerland), vol. 15, no. 12, Dec. 2024, doi: 10.3390/info15120821.
[14] S. Vijayababu, S. Bennur, and S. Vijayababu, “Enhancing Financial Chart Analysis: Advanced Detection of Candlestick Patterns Using Deep Learning Models for Mastering Trend Recognition.” [Online]. Available: https://www.researchgate.net/publication/376685209
[15] S. Goutte, H.-V. Le, F. Liu, and H.-J. von Mettenheim, “Deep learning and technical analysis in cryptocurrency market,” Financ Res Lett, vol. 54, p. 103809, 2023, doi: https://doi.org/10.1016/j.frl.2023.103809.
[16] M. Schubert et al., “Identifying Label Errors in Object Detection Datasets by Loss Inspection.”