Klasifikasi Sentimen Multimodal Pada Media Sosial X Terkait Isu Kesehatan Mental Dengan Ekspansi Fitur FastText dan Model CNN-GRU
Abstrak
Media sosial kini berfungsi sebagai platform utama bagi orang untuk membagikan pengalaman dan pandangan, termasuk yang berkaitan dengan kesehatan mental. Namun, analisis data yang ada menghadapi beberapa tantangan, seperti perbedaan jenis seperti teks dan gambar dan pola ekspresi yang rumit. Penelitian ini bertujuan untuk merancang model analisis sentimen multimodal yang dapat mendeteksi masalah kesehatan mental di media sosial X dengan menggabungkan metode Convolutional Neural Network (CNN), Gated Recurrent Units (GRU), dan ekspansi fitur FastText. Metodologi yang diusulkan mencakup pengumpulan data secara real-time dari media sosial X, meliputi teks dan gambar. Data teks diolah dengan teknik praproses standar dan representasi fitur FastText, sedangkan data visual diambil menggunakan VGG-16 untuk mengenali pola visual yang relevan. 24.742 pasangan tweet gambar dikumpulkan dari platform Twitter dan dianotasi melalui sistem pemungutan suara mayoritas. Untuk membangun korpus kemiripan FastText, 63.512 data dari portal berita digital CNN dan Twitter digabungkan penggabungan modalitas dilakukan melalui lapisan integrasi untuk menghasilkan klasifikasi sentimen akhir (positif dan negatif). Hasil evaluasi pada dataset uji menunjukkan bahwa metode ini mampu meningkatkan akurasi deteksi sentimen hingga 0,12% dibandingkan metode yang hanya berbasis teks. Secara keseluruhan, akurasi yang diperoleh mencapai 87,89%. Dengan capaian ini, diharapkan penelitian ini dapat menjadi referensi dalam pemantauan isu kesehatan mental di media sosial X secara lebih efektif.
Keywords—Analisis Sentimen Multimodal, VGG-16, Hybrid CNN-GRU, TF-IDF, FastText
Referensi
O. : Adisty, W. Putri, B. Wibhawa, and A. S. Gutama, “41 KESEHATAN MENTAL MASYARAKAT INDONESIA (PENGETAHUAN, DAN KETERBUKAAN MASYARAKAT TERHADAP GANGGUAN KESEHATAN MENTAL)”.
S. Zahira Ardhania, F. Catur, and P. Lestari, “Triwikrama: Jurnal Ilmu Sosial PENGARUH MEDIA SOSIAL TERHADAP KESEHATAN MENTAL REMAJA,” vol. 4, no. 3, pp. 2024–91, 2023.
Kukuh Wijayanti and Qoniah Nur Wijayani, “PERANAN APLIKASI X ATAU X DALAM INTERAKSI KOMUNIKASI GUNA MEMBANTU PENYEIMBANGAN KESEHATAN MENTAL PADA REMAJA SAAT INI,” JOURNAL SAINS STUDENT RESEARCH, vol. 2, no. 1, pp. 07–15, Dec. 2023, doi: 10.61722/jssr.v2i1.469.
A. Rizki and Y. Sibaroni, “ANALISIS SENTIMEN UNTUK PENGUKURAN TINGKAT DEPRESI PENGGUNA X MENGGUNAKAN DEEP LEARNING.”
N. L. Lavenia and R. Permatasari, “Sentiment Analysis on X Social Media Regarding Depression Disorder Using the Naive Bayes Method,” CoreID Journal, vol. 1, no. 2, pp. 66–74, Jul. 2023, doi: 10.60005/coreid.v1i2.14.
A. Primadhani Tirtopangarsa and W. Maharani, “Sentiment Analysis of Depression Detection on X Social Media Users Using the K-Nearest Neighbor Method Analisis Sentimen Detexti Depresi pada Pengguna Media Sosial X dengan Menggunakan Metode K-Nearest Neighbor,” pp. 13–2021.
M. Daffa, A. Fahreza, A. Luthfiarta, M. Rafid, M. Indrawan, and A. Nugraha, “Analisis Sentimen: Pengaruh Jam Kerja Terhadap Kesehatan Mental Generasi Z,” JOURNAL OF APPLIED COMPUTER SCIENCE AND TECHNOLOGY (JACOST), vol. 5, no. 1, pp. 2723–1453, 2024, doi: 10.52158/jacost.715.
V. C. Chandu, Y. Marella, G. S. Panga, S. Pachava, and V. Vadapalli, “Measuring the Impact of COVID-19 on Mental Health: A Scoping Review of the Existing Scales,” Sep. 01, 2020, SAGE Publications Ltd. doi: 10.1177/0253717620946439.
M. Fachriza and H. Artikel, “Analisis Sentimen Kalimat Depresi Pada Pengguna X Dengan Naive Bayes, Support Vector Machine, Random Forest,” 2023. [Online]. Available: http://studentjournal.umpo.ac.id/index.php/komputek
S. Jalukar, A. Ratnaparkhi, P. Shinde, S. Kunkulol, and V. Kulkarni, “SENTIMENT ANALYSIS FOR DEPRESSION DETECTION,” International Journal Of Trendy Research In Engineering And Technology, vol. 06, no. 03, 2022, doi: 10.54473/ijtret.2022.6305.
J. Park, M.-H. Tsou, A. Nara, S. Cassels, and S. Dodge, “Developing a social sensing index for monitoring place-oriented mental health issues using social media (X) data,” Urban Informatics, vol. 3, no. 1, Jan. 2024, doi: 10.1007/s44212-023- 00033-5.
J. Zhang, “An Overview of the Application of Sentiment Analysis in Mental Well- being,” Applied and Computational Engineering, vol. 8, no. 1, pp. 354–359, Aug. 2023, doi: 10.54254/2755-2721/8/20230186.
A. Bhardwaj, S. Bharany, and S. K. Kim, “Fake social media news and distorted campaign detection framework using sentiment analysis & machine learning,” Heliyon, vol. 10, no. 16, Aug. 2024, doi: 10.1016/j.heliyon.2024.e36049.
J. Sharma and V. Tomer, “Depression detection using sentiment analysis of social media data,” AIP Conf Proc, vol. 2481, no. 1, p. 020044, Nov. 2022, doi: 10.1063/5.0104192.
S. Aslan, S. Kızıloluk, and E. Sert, “TSA-CNN-AOA: X sentiment analysis using CNN optimized via arithmetic optimization algorithm,” Neural Comput Appl, vol. 35, no. 14, pp. 10311–10328, May 2023, doi: 10.1007/s00521-023-08236-2.
S. Ghosal and A. Jain, “Depression and Suicide Risk Detection on Social Media using fastText Embedding and XGBoost Classifier,” in Procedia Computer Science, Elsevier B.V., 2022, pp. 1631–1639. doi: 10.1016/j.procs.2023.01.141.
D. Murthy, “Sociology of X/X: Trends, Challenges, and Future Research Directions,” Annu Rev Sociol, vol. 17, p. 5, 2024, doi: 10.1146/annurev-soc- 031021.
S. Cong and Y. Zhou, “A review of convolutional neural network architectures and their optimizations,” Artif Intell Rev, vol. 56, no. 3, pp. 1905–1969, 2023, doi: 10.1007/s10462-022-10213-5.
Z. Li, F. Liu, W. Yang, S. Peng, and J. Zhou, “A Survey of Convolutional Neural Networks: Analysis, Applications, and Prospects,” IEEE Trans Neural Netw Learn Syst, vol. 33, no. 12, pp. 6999–7019, 2022, doi: 10.1109/TNNLS.2021.3084827.
H. Alhakiem and E. Setiawan, “Aspect-Based Sentiment Analysis on X Using Logistic Regression with FastText Feature Expansion,” Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi), vol. 6, pp. 840–846, Nov. 2022, doi: 10.29207/resti.v6i5.4429.
Adam, A. Z. R., & Setiawan, “Social Media Sentiment Analysis Using Convolutional Neural Network (CNN) and Gated Recurrent Unit (GRU),” Jurnal Ilmiah Teknik Elektro Komputer Dan Informatika, vol. 9, no. 1, pp. 119–131, 2023, doi: 10.26555/jiteki.v9i1.25813
Ukaihongsar Watthana and Jitsakul, “Enhancing Sentiment Analysis Using Hybrid Deep Learning. In S. and J. W. and T. S. Meesad Phayung and Sodsee (Ed.),” Proceedings of the 18th International Conference on Computing and Information Technology (IC2IT 2022), pp. 183–193, 2022.
Palomino, M. A., & Aider, “Evaluating the Effectiveness of Text Pre-Processing in Sentiment Analysis,” Applied Sciences (Switzerland), vol. 12, no. 17, 2022, doi: 10.3390/app12178765
Harmandini, K. P., & Muslim, “Analysis of TF-IDF and TF-RF Feature Extraction on Product Review Sentiment,” Jurnal Dan Penelitian Teknik Informatika, vol. 8, no. 2, 2024, doi: 10.33395/v8i2.13376
Adelakun, N. O., & Lasisi, “Deep Learning-Based Sentiment Analysis In Financial Markets Using Gated Recurrent Unit,” Andalasian International Journal of Applied Science, Engineering and Technology, vol. 5, no. 1, pp. 27–38, 2025, doi: 10.25077/aijaset.v5i1.217
Nadeem, A., Aslam, N., Abid, M. K., & Fuzail, “Text-Based Sentiment Analysis Using CNN-GRU Deep Learning Model,” In J. inf. commun. technol. robot. appl (Vol. 14, Issue 1). Available: http://www.jictra.com.pk/index.php/jictra,pISSN:2523-5729,eISSN:2523-5739
Ren, J, “Multimodal Sentiment Analysis Based on BERT and ResNet,” 2024, Available: http://arxiv.org/abs/2412.03625
Li, H., Lu, Y., & Zhu, H., “Multi-Modal Sentiment Analysis Based on Image and Text Fusion Based on Cross-Attention Mechanism,” Electronics (Switzerland), vol. 13, no. 11, 2024, doi: 10.3390/electronics13112069
Al-Tameemi, I. K. S., Feizi-Derakhshi, M.-R., Pashazadeh, S., & Asadpour, M., “A Comprehensive Review of Visual-Textual Sentiment Analysis from Social Media Networks,” 2023, doi: 10.1007/s42001-024-00326-y
Malhotra, A., & Jindal, R., “Multimodal deep learning based framework for detecting depression and suicidal behaviour by affective analysis of social media post,” EAI Endorsed Transactions on Pervasive Health and Technology, vol. 6 no. 21, 2020, doi: 10.4108/eai.13-7-2018.164259
