Analisis Kinerja Algoritma Deep Learning pada Pengolahan Data Kompleks
--- DOI:
https://doi.org/10.64803/joeer.v1i3.20Keywords:
Deep Learning, CNN–RNN Hybrid, Data Kompleks, Pembelajaran Mendalam, Analisis TemporalAbstract
Perkembangan pesat algoritma deep learning telah memberikan kontribusi signifikan dalam pengolahan data kompleks yang memiliki karakteristik spasial dan temporal. Namun, penerapan model deep learning tunggal sering menghadapi keterbatasan dalam menangkap pola data secara menyeluruh, khususnya pada data berdimensi tinggi dan deret waktu. Penelitian ini bertujuan untuk menganalisis efektivitas penggunaan model deep learning hybrid yang mengombinasikan Convolutional Neural Network (CNN) dan Recurrent Neural Network (RNN) dalam meningkatkan kinerja pengolahan data kompleks. Metode penelitian yang digunakan adalah pendekatan kuantitatif eksperimental dengan membandingkan performa model CNN tunggal, RNN tunggal, dan model hybrid CNN–RNN. Dataset yang digunakan merupakan data sekunder dengan karakteristik multivariat dan temporal, yang diproses melalui tahapan pra-pemrosesan, pelatihan, dan evaluasi model. Hasil penelitian menunjukkan bahwa model hybrid CNN–RNN memberikan performa terbaik dibandingkan model tunggal, ditunjukkan oleh peningkatan akurasi, presisi, recall, dan F1-score secara signifikan. Selain itu, analisis kurva loss pelatihan dan validasi menunjukkan proses pembelajaran yang stabil dan kemampuan generalisasi yang baik. Penerapan teknik regularisasi dan attention mechanism juga terbukti mampu mengurangi overfitting serta meningkatkan interpretabilitas model. Dengan demikian, model deep learning hybrid CNN–RNN memiliki potensi besar untuk diterapkan dalam berbagai domain pengolahan data kompleks, seperti analisis sinyal medis, sistem keamanan berbasis IoT, dan analisis aktivitas manusia.
References
[1] F. Zhang et al., “Predicting soil moisture content over partially vegetation covered surfaces from hyperspectral data with deep learning,” Soil Sci. Soc. Am. J., vol. 85, no. 4, pp. 989–1001, 2021, doi: 10.1002/saj2.20193.
[2] A. K. -, N. G. -, and A. S. -, “A Critical Review on Sentiment Analysis Based on Deep Learning Techniques,” Int. J. Multidiscip. Res., vol. 6, no. 5, 2024, doi: 10.36948/ijfmr.2024.v06i05.28572.
[3] H. Liu et al., “Combined CNN and RNN Neural Networks for GPR Detection of Railway Subgrade Diseases,” Sensors, vol. 23, no. 12, p. 5383, 2023, doi: 10.3390/s23125383.
[4] J. Zhang et al., “Sensor-Driven Real-Time Recognition of Basketball Goal States Using IMU and Deep Learning,” Sensors, vol. 25, no. 12, p. 3709, 2025, doi: 10.3390/s25123709.
[5] J. Wang, P. Wang, H. Tian, K. Tansey, J. Liu, and W. Quan, “A deep learning framework combining CNN and GRU for improving wheat yield estimates using time series remotely sensed multi-variables,” Comput. Electron. Agric., vol. 206, p. 107705, 2023, doi: 10.1016/j.compag.2023.107705.
[6] S. Mekruksavanich and A. Jitpattanakul, “Deep convolutional neural network with rnns for complex activity recognition using wrist-worn wearable sensor data,” Electron., vol. 10, no. 14, p. 1685, 2021, doi: 10.3390/electronics10141685.
[7] Y. Wang et al., “Computer-aided intracranial eeg signal identification method based on a multi-branch deep learning fusion model and clinical validation,” Brain Sci., vol. 11, no. 5, p. 615, 2021, doi: 10.3390/brainsci11050615.
[8] V. Prakash and A. K. Shukla, “An AI-Enabled Privacy-Preserving Federated Learning Framework of Hybrid bi-LSTM-RNN-CNN for Deep Intelligent Intrusion Detection in Fog Computing in the IoT Domain,” Concurr. Comput. Pract. Exp., vol. 37, no. 27–28, 2025, doi: 10.1002/cpe.70291.
[9] A. Meliboev, J. Alikhanov, and W. Kim, “Performance Evaluation of Deep Learning Based Network Intrusion Detection System across Multiple Balanced and Imbalanced Datasets,” Electron., vol. 11, no. 4, p. 515, 2022, doi: 10.3390/electronics11040515.
[10] A. S. Gaafar, J. M. Dahr, and A. K. Hamoud, “Comparative Analysis of Performance of Deep Learning Classification Approach based on LSTM-RNN for Textual and Image Datasets,” Inform., vol. 46, no. 5, pp. 21–28, 2022, doi: 10.31449/inf.v46i5.3872.
[11] D. T. Kim Chi, N. T. Mai Trang, T. B. Minh Son, N. Ngoc Thao, and T. Q. Nguyen, “Hybrid deep learning framework for robust time-series classification: Integrating inception modules with residual networks,” J. Algorithms Comput. Technol., vol. 19, 2025, doi: 10.1177/17483026251348851.
[12] M. Kamyab, G. Liu, and M. Adjeisah, “Attention-Based CNN and Bi-LSTM Model Based on TF-IDF and GloVe Word Embedding for Sentiment Analysis,” Appl. Sci., vol. 11, no. 23, p. 11255, 2021, doi: 10.3390/app112311255.
[13] T. Zengeya and J. Vincent Fonou-Dombeu, “A Review of State of the Art Deep Learning Models for Ontology Construction,” IEEE Access, vol. 12, pp. 82354–82383, 2024, doi: 10.1109/ACCESS.2024.3406426.
[14] W. U. R. Qamar, M. H. Lee, and B. Abibullaev, “Deep learning in intracranial EEG for seizure detection: advances, challenges, and clinical applications,” Front. Neurosci., vol. 19, 2025, doi: 10.3389/fnins.2025.1677898.
[15] S. Yamamoto et al., “Data-driven electrophysiological feature based on deep learning to detect epileptic seizures,” J. Neural Eng., vol. 18, no. 5, p. 56040, 2021, doi: 10.1088/1741-2552/ac23bf.
[16] E. Ahmadzadeh, H. Kim, O. Jeong, N. Kim, and I. Moon, “A Deep Bidirectional LSTM-GRU Network Model for Automated Ciphertext Classification,” IEEE Access, vol. 10, pp. 3228–3237, 2022, doi: 10.1109/ACCESS.2022.3140342.
[17] V. Laitsos, G. Vontzos, D. Bargiotas, A. Daskalopulu, and L. H. Tsoukalas, “Enhanced Automated Deep Learning Application for Short-Term Load Forecasting,” Mathematics, vol. 11, no. 13, p. 2912, 2023, doi: 10.3390/math11132912.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Siti Khodijah, Cindy Atika Rizki (Author)
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
