Perbandingan Naive Bayes dan Gated Recurrent Unit untuk Klasifikasi Keluhan Publik di Kabupaten Sleman
DOI:
https://doi.org/10.26418/jp.v10i3.81687Keywords:
machine learning, deep learning, Naive Bayes, Gated Recurrent Unit, klasifikasi teksAbstract
Pemerintah Kabupaten Sleman, sejak 15 Mei 2016, telah meluncurkan aplikasi bernama 'Lapor Sleman', yang berfungsi sebagai platform untuk mengajukan keluhan masyarakat. Warga menginput data keluhan ke dalam aplikasi 'Lapor Sleman', termasuk kategori keluhan, judul laporan, rincian, serta foto dan koordinat. Seiring waktu, penggunaan media sosial seperti Twitter telah memberikan masyarakat berbagai opsi untuk mengekspresikan keluhannya. Klasifikasi manual dari banyak postingan Twitter oleh manusia telah menjadi tugas rutin yang seharusnya dapat dilakukan secara otomatis oleh mesin. Selain itu, memfasilitasi masyarakat dalam mengajukan keluhan, termasuk klasifikasi keluhan otomatis, adalah sesuatu yang perlu diimplementasikan. Tujuan utama penelitian ini adalah mengevaluasi efektivitas dan akurasi metode Machine Learning (Naive Bayes Classifier) dan Deep Learning (Gated Recurrent Unit) dalam mengkategorikan keluhan masyarakat di Kabupaten Sleman. Tujuan penelitian secara khusus meliputi: (1) Menganalisis dampak hyperparameter Naive Bayes Classifier (NBC) terhadap akurasi selama pelatihan dan implementasi berkas pickle; (2) Menganalisis hyperparameter Gated Recurrent Unit (GRU) terhadap akurasi selama pelatihan dan penggunaan berkas h5 dan pth; (3) Membandingkan kinerja antara metode NBC dan GRU. Dataset pelatihan terdiri dari 5.308 keluhan masyarakat yang dikumpulkan dari aplikasi 'Lapor Sleman'. Hasil menunjukkan keunggulan NBC. Namun, NBC sedikit kurang dalam mencapai akurasi dataset keseluruhan dibandingkan dengan GRU PyTorch. Sebaliknya, GRU di TensorFlow menunjukkan kinerja yang relatif kurang baik, menempati peringkat ketiga dalam sebagian besar aspek. Selain itu, GRU di TensorFlow menunjukkan tanda-tanda overfitting, terlihat melalui penurunan akurasi selama pengujian dengan dataset lengkap.
References
H. Dai, X. Peng, X. Shi, L. He, Q. Xiong, and H. Jin, "Reveal training performance mystery between TensorFlow and PyTorch in the single GPU environment," Science China Information Sciences, vol. 65, no. 1, 2021/12/16 2021, doi: 10.1007/s11432-020-3182-1.
R. Elshawi, A. Wahab, A. Barnawi, and S. Sakr, "DLBench: a comprehensive experimental evaluation of deep learning frameworks," Cluster Computing, vol. 24, no. 3, pp. 2017-2038, 2021/09/01 2021, doi: 10.1007/s10586-021-03240-4.
M. Li et al., "The Deep Learning Compiler: A Comprehensive Survey," IEEE Transactions on Parallel and Distributed Systems, vol. 32, no. 3, pp. 708-727, 2021/3 2021, doi: 10.1109/tpds.2020.3030548.
L. Alzubaidi et al., "A survey on deep learning tools dealing with data scarcity: definitions, challenges, solutions, tips, and applications," Journal of Big Data, vol. 10, no. 1, p. 46, 2023/04/14 2023, doi: 10.1186/s40537-023-00727-2.
S. Georgiou, M. Kechagia, T. Sharma, F. Sarro, and Y. Zou, "Green AI: Do Deep Learning Frameworks Have Different Costs?," presented at the International Conference on Software Engineering (ICSE), Pittsburgh, USA, 2022. [Online]. Available: http://dx.doi.org/10.1145/3510003.3510221.
R. Munjal, S. Arif, F. Wendler, and O. Kanoun, "Comparative Study of Machine-Learning Frameworks for the Elaboration of Feed-Forward Neural Networks by Varying the Complexity of Impedimetric Datasets Synthesized Using Eddy Current Sensors for the Characterization of Bi-Metallic Coins," Sensors, vol. 22, no. 4, p. 1312, 2022, doi: 10.3390/s22041312.
O.-C. Novac et al., "Analysis of the Application Efficiency of TensorFlow and PyTorch in Convolutional Neural Network," Sensors, vol. 22, no. 22, p. 8872, 2022, doi: 10.3390/s22228872.
Zulqarnain, R. Ghazali, Y. M. Mohmad Hassim, and M. Rehan, "Text Classification Based on Gated Recurrent Unit Combines With Support Vector Machine," International Journal of Electrical and Computer Engineering (IJECE), vol. 10, no. 4, p. 3734, 08/01 2020, doi: 10.11591/ijece.v10i4.pp3734-3742.
L. Deng et al., "News Text Classification Method Based on the GRU_CNN Model," International Transactions on Electrical Energy Systems, vol. 2022, pp. 1-11, 08/31 2022, doi: 10.1155/2022/1197534.
S. Mondal, M. Barman, and A. Nag, "Question classification task based on deep learning models with self-attention mechanism," Multimedia Tools and Applications, 2024/04/26 2024, doi: 10.1007/s11042-024-19239-z.
X. Zhang, Z. Wu, K. Liu, Z. Zhao, J. Wang, and C. Wu, "Text Sentiment Classification Based on BERT Embedding and Sliced Multi-Head Self-Attention Bi-GRU," Sensors, vol. 23, no. 3, p. 1481, 01/28 2023, doi: 10.3390/s23031481.
S. M. Intani, B. I. Nasution, M. E. Aminanto, Y. Nugraha, N. Muchtar, and J. I. Kanggrawan, "Automating Public Complaint Classification Through JakLapor Channel: A Case Study of Jakarta, Indonesia," presented at the 2022 IEEE International Smart Cities Conference (ISC2), 26-29 Sept. 2022, 2022. [Online]. Available: http://dx.doi.org/10.1109/ISC255366.2022.9922346.
W. Keszya, N. Oky Dwi, and Farikhin, "Application of The Naïve Bayes Classifier Algorithm to Classify Community Complaints," Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi), vol. 6, no. 5, 11/02 2022, doi: 10.29207/resti.v6i5.4498.
X. Luo, "Efficient English text classification using selected Machine Learning Techniques," Alexandria Engineering Journal, vol. 60, no. 3, pp. 3401-3409, 2021/06/01/ 2021, doi: https://doi.org/10.1016/j.aej.2021.02.009.
E. Lia, R. Yova, P. Lelianto Eko, and H. Andro, "Sentiment Analysis of Twitter Users to the PeduliLindungi Using Naïve Bayes Algorithm," Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi), vol. 7, no. 2, 03/28 2023, doi: 10.29207/resti.v7i2.4684.
V. Merve, V. Erkan, and B. İhsan Ömür, "Performance Comparison between Naive Bayes and Machine Learning Algorithms for News Classification," in Bayesian Inference, B. İhsan Ömür Ed. Rijeka: IntechOpen, 2024, p. Ch. 5.
K. E. ArunKumar, D. Kalaga, C. Kumar, M. Kawaji, and T. Brenza, "Comparative analysis of Gated Recurrent Units (GRU), long Short-Term memory (LSTM) cells, autoregressive Integrated moving average (ARIMA), seasonal autoregressive Integrated moving average (SARIMA) for forecasting COVID-19 trends," Alexandria Engineering Journal, vol. 61, no. 10, 01/06 2022, doi: 10.1016/j.aej.2022.01.011.
J. Noh, H.-J. Park, J. S. Kim, and S.-J. Hwang, "Gated Recurrent Unit with Genetic Algorithm for Product Demand Forecasting in Supply Chain Management," Mathematics, vol. 8, no. 4, p. 565, 2020, doi: 10.3390/math8040565.
B. Hussain, M. K. Afzal, S. Ahmad, and A. M. Mostafa, "Intelligent Traffic Flow Prediction Using Optimized GRU Model," IEEE Access, vol. 9, pp. 100736-100746, 2021, doi: 10.1109/ACCESS.2021.3097141.
R. Kumar, B. Krishna Goswami, S. Motiram Mhatre, and S. Agrawal, "Naive Bayes in focus: A thorough examination of its algorithmic foundations and use cases," Int. J. Innov. Sci. Res. Technol., vol. 9, no. 5, pp. 2078-2081, 2024/6/7 2024, doi: 10.38124/ijisrt/ijisrt24may1438.
E. Elgeldawi, A. Sayed, A. R. Galal, and A. M. Zaki, "Hyperparameter Tuning for Machine Learning Algorithms Used for Arabic Sentiment Analysis," Informatics, vol. 8, no. 4, p. 79, 2021, doi: 10.3390/informatics8040079.
Y. Fauziah, B. Yuwono, and A. Sasmito, "Lexicon Based Sentiment Analysis in Indonesia Languages : A Systematic Literature Review," RSF Conference Series: Engineering and Technology, vol. 1, no. 1, pp. 363-367, 12/23 2021, doi: 10.31098/cset.v1i1.397.
