CNN-LSTM Architecture for Multi-Task Sentiment and Emotion Classification on Large-Scale Indonesian TikTok Application Reviews

Wahyu Fajar Setiawan; Afif Amirullah; Ilham Putra Ariatama; Ratih Nur Esti Anggraini

Authors

Wahyu Fajar Setiawan Sepuluh Nopember Institute of Technology (ITS)
Afif Amirullah Sepuluh Nopember Institute of Technology (ITS)
Ilham Putra Ariatama Sepuluh Nopember Institute of Technology (ITS)
Ratih Nur Esti Anggraini Sepuluh Nopember Institute of Technology (ITS)

Keywords:

Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM), Multi-Task Learning, Sentiment Analysis, Emotion Classification, Deep Learning

Abstract

Sentiment and emotion analysis of mobile application reviews has attracted significant attention as a means to understand users’ perceptions and experiences. The research proposes a novel Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) model for multi-task sentiment and emotion classification on Indonesian TikTok application reviews. A large-scale corpus consisting of 500,000 reviews is collected from the Google Play Store and preprocessed through cleaning, normalization, tokenization, stopword removal, and stemming. Sentiment labels (positive, negative, and neutral) are assigned using a lexicon-based approach, while emotion labels are annotated through emoji analysis and word matching based on five basic emotions: anger, fear, happiness, love, and sadness. The proposed CNN-LSTM model is evaluated against a hybrid Bidirectional Encoder Representations from Transformers – Convolutional Neural Network (BERT-CNN) architecture. Experimental results show that the CNN-LSTM model outperforms the BERT-CNN model, achieving an accuracy of 91.30% for sentiment classification and 99.15% for emotion classification, compared to 42.43% and 72.85%, respectively, obtained by the BERT-CNN model. These findings indicate that the CNN-LSTM architecture is more effective in capturing sequential patterns and contextual features in Indonesian review texts, particularly in a multi-task learning setting. Despite its strong performance, the research is limited by its focus on a single platform and the use of lexicon-based automatic labeling, suggesting future work on cross-domain evaluation and manual annotation refinement.

Dimensions

Author Biographies

Wahyu Fajar Setiawan, Sepuluh Nopember Institute of Technology (ITS)

Department of Informatics, Faculty of Intelligent Electrical and Informatics Technology

Afif Amirullah, Sepuluh Nopember Institute of Technology (ITS)

Department of Informatics, Faculty of Intelligent Electrical and Informatics Technology

Ilham Putra Ariatama, Sepuluh Nopember Institute of Technology (ITS)

Department of Informatics, Faculty of Intelligent Electrical and Informatics Technology

Ratih Nur Esti Anggraini, Sepuluh Nopember Institute of Technology (ITS)

Department of Informatics, Faculty of Intelligent Electrical and Informatics Technology

References

[1] N. Khamphakdee and P. Seresangtakul, “An efficient deep learning for Thai sentiment analysis,” Data, vol. 8, no. 5, pp. 1–22, 2023.

[2] N. A. S. Abdullah and N. I. A. Rusli, “Multilingual sentiment analysis: A systematic literature review,” Pertanika Journal of Science & Technology, vol. 29, no. 1, pp. 445–470, 2021.

[3] L. Chen, S. Shang, and Y. Wang, “Crosslingual sentiment analysis with MultiEmo: Exploring language-agnostic models for emotion recognition,” 2024. [Online]. Available: https://doi.org/10.20944/preprints202408.1639.v1

[4] S. H. Park, K.-M. Kim, O. J. Lee, Y. Kang, J. Lee, S. M. Lee, and S. Lee, ““why do I feel offended?”-Korean dataset for offensive language identification,” in Findings of the Association for Computational Linguistics: EACL 2023. Dubrovnik, Croatia: Association for Computational Linguistics, 2023, pp. 1142–1153.

[5] M. Garouani and J. Kharroubi, “MAC: An open and free Moroccan Arabic corpus for sentiment analysis,” in The Proceedings of the International Conference on Smart City Applications. Safranbolu, T¨urkiye: Springer, Oct. 27–29, 2021, pp. 849–858.

[6] H. Fouadi, H. El Moubtahij, H. Lamtougui, and A. Yahyaouy, “Sentiment analysis of Arabic comments using machine learning and deep learning model,” Indian Journal of Computer Science and Engineering (IJCSE), vol. 13, no. 3, pp. 598–606, 2022.

[7] K. M. Awlla, H. Veisi, and A. A. Abdullah, “KuBERT: Central Kurdish BERT model and its application for sentiment analysis,” 2024. [Online]. Available: https://doi.org/10.21203/rs.3.rs-4552724/v1

[8] V. Dhananjaya, S. Ranathunga, and S. Jayasena, “Lexicon-based fine-tuning of multilingual language models for low-resource language sentiment analysis,” CAAI Transactions on Intelligence Technology, vol. 9, no. 5, pp. 1116–1125, 2024.

[9] M. S. Akhtar, D. Ghosal, A. Ekbal, P. Bhattacharyya, and S. Kurohashi, “All-in-one: Emotion, sentiment and intensity prediction using a multi-task ensemble framework,” IEEE Transactions on Affective Computing, vol. 13, no. 1, pp. 285–297, 2019.

[10] G. Thakkar, N. M. Preradovi´c, and M. Tadi´c, “Transferring sentiment cross-lingually within and across same-family languages,” Applied Sciences, vol. 14, no. 13, pp. 1–21, 2024.

[11] M. Sangeetha and K. Nimala, “Sentiment analysis on code-mixed Tamil-English corpus: A comprehensive study of transformer-based models,” 2023. [Online]. Available: https://doi.org/10.21203/rs.3.rs-3418283/v1

[12] A. Kniele and M. Beloucif, “Uppsala University at SemEval-2023 task12: Zero-shot sentiment classification for Nigerian Pidgin tweets,” in Proceedings of the 17th International Workshop on Semantic Evaluation (SemEval-2023). Toronto, Canada: Association for Computational Linguistics, 2023, pp. 1491–1497.

[13] L. Khan, A. Amjad, N. Ashraf, H. T. Chang, and A. Gelbukh, “Urdu sentiment analysis with deep learning methods,” IEEE Access, vol. 9, pp. 97 803–97 812, 2021.

[14] K. Cortis, K. Verma, and B. Davis, “Fine-tuning neural language models for multidimensional opinion mining of English-Maltese social data,” in Proceedings of the International Conference on Recent Advances in Natural Language Processing (RANLP 2021). INCOMA Ltd., 2021, pp. 309–314.

[15] C. Kumaresan and P. Thangaraju, “Sentiment analysis in multiple languages: A review of current approaches and challenges,” REST Journal on Data Analytics and Artificial Intelligence, vol. 2, no. 1, pp. 8–15, 2023.

[16] D. Homskiy and N. Maloyan, “DN at SemEval-2023 task 12: Low-resource language text classification via multilingual pretrained language model fine-tuning,” in Proceedings of the 17th International Workshop on Semantic Evaluation (SemEval-2023). Toronto, Canada: Association for Computational Linguistics, 2023, pp. 1537–1541.

[17] T. D. Purnomo and J. Sutopo, “Comparison of pre-trained BERT-based transformer models for regional language text sentiment analysis in indonesia,” International Journal Science and Technology, vol. 3, no. 3, pp. 11–21, 2024.

[18] F. Koto and G. Y. Rahmaningtyas, “Inset lexicon: Evaluation of a word list for Indonesian sentiment analysis in microblogs,” in 2017 International Conference on Asian Language Processing (IALP). Singapore: IEEE, Dec. 5–7, 2017, pp. 391–394.

[19] K. Ronny Mabokela, M. Primus, and T. Celik, “Advancing sentiment analysis for low-resourced african languages using pre-trained language models,” PLOS One, vol. 20, no. 6, pp. 1–37, 2025.

[20] Z. Kastrati, L. Ahmedi, A. Kurti, F. Kadriu, D. Murtezaj, and F. Gashi, “A deep learning sentiment analyser for social media comments in low-resource languages,” Electronics, vol. 10, no. 10, pp. 1–19, 2021.

[21] U. I. Shabrina, R. Sarno, R. N. E. Anggraini, A. T. Haryono, and A. F. Septiyanto, “Sentiment analysis of presidential candidate debates from Youtube videos,” in 2024 IEEE International Conference on Artificial Intelligence and Mechatronics Systems (AIMS). Bandung, Indonesia: IEEE, Feb. 21–23, 2024, pp. 1–6.

[22] T. De Melo and P. Merialdo, “SentiLexIT: Advancing Italian sentiment analysis through automated lexicon generation,” 2024. [Online]. Available: https://doi.org/10.21203/rs.3.rs-4630348/v1

[23] A. T. Haryono, R. Sarno, R. N. E. Anggraini, and K. R. Sungkono, “Permuted temporal Kolmogorov-Arnold networks for stock price forecasting using generative aspect-based sentiment analysis,” IEEE Access, vol. 12, pp. 178 672–178 689, 2024.

[24] W. Yu, L. Yin, C. Zhang, Y. Chen, and A. X. Liu, “Application of quantum recurrent neural network in low-resource language text classification,” IEEE Transactions on Quantum Engineering, vol. 5, pp. 1–13, 2024.

[25] M. Alali, N. Mohd Sharef, M. A. Azmi Murad, H. Hamdan, and N. A. Husin, “Multitasking learning model based on hierarchical attention network for Arabic sentiment analysis classification,” Electronics, vol. 11, no. 8, pp. 1–23, 2022.

[26] K. R. Mabokela, T. Celik, and M. Raborife, “Multilingual sentiment analysis for under-resourced languages: A systematic review of the landscape,” IEEE Access, vol. 11, pp. 15 996–16 020, 2022.

[27] V. K. Agbesi, W. Chen, C. C. Ukwuoma, N. A. Kuadey, C. C. M. Agbesi, C. J. Ejiyi, E. S. A. Gyarteng, G. W. Muoka, and A. M. Kuadey, “Multichannel 2D-CNN attention-based BiLSTM method for low-resource Ewe sentiment analysis,” Journal of Data Science and Intelligent Systems, vol. 3, no. 1, pp. 67–77, 2025.

[28] M. R. Ashraf, Y. Jana, Q. Umer, M. A. Jaffar, S. Chung, and W. Y. Ramay, “BERT-based sentiment analysis for low-resourced languages: A case study of Urdu language,” IEEE Access, vol. 11, pp. 110 245–110 259, 2023.

[29] S. Kaddoura, M. Itani, and C. Roast, “Analyzing the effect of negation in sentiment polarity of Facebook dialectal Arabic text,” Applied Sciences, vol. 11, no. 11, pp. 1–13, 2021.

[30] R. Mabokela, M. Roborife, and T. Celik, “Investigating sentiment-bearing words-and emoji-based distant supervision approaches for sentiment analysis,” in Proceedings of the Fourth Workshop on Resources for African Indigenous Languages (RAIL 2023). Dubrovnik, Croatia: Association for Computational Linguistics, 2023, pp. 115–125.

[31] S. Wang, Y. Dai, J. Shen, and J. Xuan, “Research on expansion and classification of imbalanced data based on SMOTE algorithm,” Scientific reports, vol. 11, no. 1, pp. 1–11, 2021.