HYBRIDIZATION OF FASTTEXT-BLSTM AND BERT FOR ENHANCED SENTIMENT ANALYSIS ON SOCIAL MEDIA TEXTS
DOI:
https://doi.org/10.33480/jitk.v11i3.7488Kata Kunci:
BERT, BLSTM, FastText, Hybrid Model, Sentiment AnalysisAbstrak
The development of internet technology and social media has driven the increasing use of sentiment analysis to understand public opinion. This study aims to improve the classification performance of sentiment analysis by proposing a hybrid model that combines FastText-BLSTM and BERT. The dataset used consists of 900 Indonesian-language Netflix app user reviews obtained through crawling using Google Play Scraper. The research stages include text preprocessing, feature extraction using FastText and BERT, and classification using BLSTM, which are then combined in a concatenation layer to produce a richer feature representation. Experimental results show that the FastText-BLSTM-BERT hybrid model provides the best performance with an accuracy of 94.22%, a precision of 95.98%, a recall of 95.68%, and an F1-score of 95.83%. This achievement is superior to the single models of FastText-BLSTM and BERT. The main novelty of this research lies in the integration of contextual embeddings from BERT with subword-level semantic and sequential representations from FastText-BLSTM, which has not been extensively explored in prior studies on Indonesian sentiment analysis. This hybridization demonstrates significant improvement in model generalization and robustness for low-resource language texts
Unduhan
Referensi
[1] Y. Bilan, O. Oliinyk, H. Mishchuk, and M. Skare, “Impact of information and communications technology on the development and use of knowledge,” Technol. Forecast. Soc. Change, vol. 191, no. March, p. 122519, 2023, doi: 10.1016/j.techfore.2023.122519.
[2] R. Lozano-Blasco, M. Mira-Aladrén., and M. Gil-Lamata, “Social media influence on young people and children: Analysis on Instagram, Twitter and YouTube,” Comunicar, vol. 30, no. 74, pp. 117–128, 2023, doi: http://doi.org/10.3916/C74-2023-10.
[3] T. Mantoro, M. A. Permana, and M. A. Ayu, “Crime index based on text mining on social media using multi classifier neural-net algorithm,” Telkomnika (Telecommunication Comput. Electron. Control., vol. 20, no. 3, pp. 570–579, 2022, doi: https://doi.org/10.12928/TELKOMNIKA.v20i3.23321.
[4] S. Assegaff, E. Rasywir, and Y. Pratama, “Experimental of vectorizer and classifier for scrapped social media data,” Telkomnika (Telecommunication Comput. Electron. Control., vol. 21, no. 4, pp. 815–824, 2023, doi: https://doi.org/10.12928/TELKOMNIKA.v21i4.24180.
[5] J. R. Hanaysha, “Impact of social media marketing features on consumer’s purchase decision in the fast-food industry: Brand trust as a mediator,” Int. J. Inf. Manag. Data Insights, vol. 2, no. 2, p. 100102, 2022, doi: 10.1016/j.jjimei.2022.100102.
[6] H. R. Alhakiem and E. B. Setiawan, “Aspect-Bas1ed Sentiment Analysis on Twitter Using Logistic Regression with FastText Feature Expansion,” J. RESTI (Rekayasa Sist. dan Teknol. Informasi), vol. 6, no. 5, pp. 840–846, 2022, doi: https://doi.org/10.29207/resti.v6i5.4429.
[7] J. Jasmir, D. Z. Abidin, F. Fachruddin, and W. Riyadi, “Experimental of information gain and AdaBoost feature for machine learning classifier in media social data,” Indones. J. Electr. Eng. Comput. Sci., vol. 36, no. 2, pp. 1172–1181, 2024, doi: 10.11591/ijeecs.v36.i2.pp1172-1181.
[8] Z. A. Khan et al., “Developing Lexicons for Enhanced Sentiment Analysis in Software Engineering: An Innovative Multilingual Approach for Social Media Reviews,” Comput. Mater. Contin., vol. 79, no. 2, pp. 2771–2793, 2024, doi: https://doi.org/10.32604/cmc.2024.046897.
[9] Y. Li, “Deep Learning-Based Natural Language Processing Methods for Sentiment Analysis in Social Networks,” Math. Probl. Eng., vol. 2022, 2022, doi: 10.1155/2022/1390672.
[10] P. Ray and A. Chakrabarti, “A Mixed approach of Deep Learning method and Rule-Based method to improve Aspect Level Sentiment Analysis,” Appl. Comput. Informatics, vol. 18, no. 1–2, pp. 163–178, 2022, doi: 10.1016/j.aci.2019.02.002.
[11] W. Guo and A. Caliskan, “Detecting Emergent Intersectional Biases: Contextualized Word Embeddings Contain a Distribution of Human-like Biases,” AIES 2021 - Proc. 2021 AAAI/ACM Conf. AI, Ethics, Soc., pp. 122–133, 2021, doi: 10.1145/3461702.3462536.
[12] Z. Zhuang, Z. Liang, Y. Rao, H. Xie, and F. L. Wang, “Out-of-vocabulary word embedding learning based on reading comprehension mechanism,” Nat. Lang. Process. J., vol. 5, no. August, p. 100038, 2023, doi: 10.1016/j.nlp.2023.100038.
[13] J. Jasmir, W. Riyadi, S. R. Agustini, Y. Arvita, D. Meisak, and L. Aryani, “Bidirectional Long Short-Term Memory and Word Embedding Feature for,” J. RESTI (Rekayasa Sist. Dan Teknol. Informasi), vol. 6, no. 4, pp. 505–510, 2022, doi: https://doi.org/10.29207/resti.v6i4.4005.
[14] M. A. Hussain Sujon and H. Mustafa, “Comparative Study of Machine Learning Models on Multiple Breast Cancer Datasets,” Int. J. Adv. Sci. Comput. Eng., vol. 5, no. 1, pp. 15–24, 2023, doi: 10.62527/ijasce.5.1.105.
[15] S. Rapacz, P. Chołda, and M. Natkaniec, “A method for fast selection of machine-learning classifiers for spam filtering,” Electron., vol. 10, no. 17, 2021, doi: https://doi.org/10.3390/electronics10172083.
[16] F. N. N. H. Rossi Passarella, Siti Nurmaini, Muhammad Naufal Rachmatullah, Harumi Veny, “Development of a machine learning model for predicting abnormalities of commercial airplanes,” Res. Pract. Thromb. Haemost., p. 100137, 2023, doi: https://doi.org/10.1016/j.jsamd.2023.100613.
[17] I. A. Wisky, S. Defit, and G. W. Nurcahyo, “Development of Extraction Features for Detecting Adolescent Personality with Machine Learning Algorithms,” Int. J. INFORMATICS Vis., vol. 8, no. November, 2024, [Online]. Available: www.joiv.org/index.php/joiv
[18] H. Santoso, G. F. Soares, and C. M. Angelo, “Convolutional Neural Network - Based Recognition of Children ’ s Facial Expressions in Response to Gaming,” Int. J. Adv. Sci. Comput. Eng., vol. 6, no. 1, pp. 152–157, 2024.
[19] N. Ansar, M. S. Ansari, M. Sharique, A. Khatoon, M. A. Malik, and M. M. Siddiqui, “A Cutting-Edge Deep Learning Method For Enhancing IoT Security,” Int. J. Adv. Sci. Comput. Eng., vol. 6, no. December, pp. 98–103, 2024, [Online]. Available: http://arxiv.org/abs/2406.12400
[20] A. H. Mohammed and A. H. Ali, “Survey of BERT (Bidirectional Encoder Representation Transformer) types,” J. Phys. Conf. Ser., vol. 1963, no. 1, 2021, doi: 10.1088/1742-6596/1963/1/012173.
[21] N. M. Gardazi, A. Daud, M. K. Malik, A. Bukhari, T. Alsahfi, and B. Alshemaimri, “BERT applications in natural language processing: a review,” Artif. Intell. Rev., vol. 58, no. 6, 2025, doi: 10.1007/s10462-025-11162-5.
[22] Y. Hao, L. Dong, F. Wei, and K. Xu, “Self-Attention Attribution: Interpreting Information Interactions Inside Transformer,” 35th AAAI Conf. Artif. Intell. AAAI 2021, vol. 14B, pp. 12963–12971, 2021, doi: 10.1609/aaai.v35i14.17533.
[23] E. Jasmir, Jasmir; Rasywir, H. Yani, and A. Nugroho, “Comparison of word embedding features using deep learning in sentiment analysis,” TELKOMNIKA Telecommun. Comput. Electron. Control, vol. 23, no. 2, pp. 416–425, 2025, doi: 10.12928/TELKOMNIKA.v23i2.26223.
[24] A. Mahmoud and M. Zrigui, “BLSTM-API: Bi-LSTM Recurrent Neural Network-Based Approach for Arabic Paraphrase Identification,” Arab. J. Sci. Eng., vol. 46, no. 4, pp. 4163–4174, 2021, doi: 10.1007/s13369-020-05320-w.
[25] M. Umer et al., “Impact of convolutional neural network and FastText embedding on text classification,” Multimed. Tools Appl., vol. 82, no. 4, pp. 5569–5585, 2023, doi: 10.1007/s11042-022-13459-x.
[26] E. Hashmi, S. Y. Yayilgan, M. M. Yamin, S. Ali, and M. Abomhara, “Advancing Fake News Detection: Hybrid Deep Learning With FastText and Explainable AI,” IEEE Access, vol. 12, no. March, pp. 44462–44480, 2024, doi: 10.1109/ACCESS.2024.3381038.
[27] F. S. Utomo, “Information Retrieval Method for the Qur ’ an based on FastText and Latent Semantic Indexing,” J. Sist. Inf., vol. 14, pp. 1014–1024, 2025.
[28] N. Haffar and M. Zrigui, “A Synergistic Bidirectional LSTM and N-gram Multi-channel CNN Approach Based on BERT and FastText for Arabic Event Identification,” ACM Trans. Asian Low-Resource Lang. Inf. Process., vol. 22, no. 11, 2023, doi: 10.1145/3626568.
[29] N. Haffar and M. Zrigui, “BERT embedding model for arabic temporal relation classification using hybrid deep learning architectures and linguistic features,” Int. J. Knowledge-Based Intell. Eng. Syst., vol. 29, no. 1, pp. 32–54, 2025, doi: 10.3233/KES-230066.
[30] H. Gunduz, “Comparative analysis of BERT and FastText representations on crowdfunding campaign success prediction,” PeerJ Comput. Sci., vol. 10, 2024, doi: 10.7717/PEERJ-CS.2316.
[31] S. Shreyashree, P. Sunagar, S. Rajarajeswari, and A. Kanavalli, “BERT-Based Hybrid RNN Model for Multi-class Text Classification to Study the Effect of Pre-trained Word Embeddings,” Int. J. Adv. Comput. Sci. Appl., vol. 13, no. 9, pp. 667–674, 2022, doi: 10.14569/IJACSA.2022.0130979.
[32] N. K, K. M, S. V. Easwaramoorthy, D. C R, S. Yoo, and J. Cho, “Hybrid approach of deep feature extraction using BERT– OPCNN & FIAC with customized Bi-LSTM for rumor text classification,” Alexandria Eng. J., vol. 90, no. October 2023, pp. 65–75, 2024, doi: 10.1016/j.aej.2024.01.056
Unduhan
Diterbitkan
Terbitan
Bagian
Lisensi
Hak Cipta (c) 2026 Jasmir, Maria Rosario, Irawan Irawan, Agus Siswanto, Tiko Nur Annisa
Artikel ini berlisensi Creative Commons Attribution-NonCommercial 4.0 International License.
-a.jpg)
-b.jpg)
