Identifikasi Komplain di Lingkungan Perguruan Tinggi Menggunakan Algorithm Convolutional Long Short Term Memory dan FastText Embedding

Authors

  • khusnul khotimah Universitas Islam Negeri Maulana Malik Ibrahim, Malang
  • Cahyo Crysdian Universitas Islam Negeri Maulana Mallik Ibrahim, Malang
  • M.Imamudin Universitas Islam Negeri Maulana Mallik Ibrahim, Malang

DOI:

https://doi.org/10.30736/jti.v11i02.1752

Keywords:

Klasifikasi Keluhan, Convolutional LSTM, Pembelajaran Mendalam (Deep Learning), FastText, Perguruan Tinggi, Klasifikasi Teks

Abstract

Perguruan tinggi rutin menerima keluhan mahasiswa terkait akademik, dosen, tenaga kependidikan, pembelajaran, fasilitas, dan keuangan, yang jika diklasifikasikan secara manual memakan waktu dan rawan subjektivitas. Penelitian ini mengusulkan sistem klasifikasi keluhan otomatis berbasis Long Short-Term Memory (LSTM) yang dipadukan dengan representasi kata FastText. Sebanyak 835 komentar keluhan dari lima program studi dikumpulkan dan dibersihkan menjadi 449 komentar unik, dikelompokkan ke dalam enam kategori: akademik, keuangan, layanan dosen, layanan tenaga kependidikan, pembelajaran, dan fasilitas. Setelah case folding, tokenisasi, dan padding, teks direpresentasikan dengan embedding FastText berdimensi 300 yang dilatih khusus pada data latih tiap fold untuk mencegah kebocoran data. Delapan arsitektur berbasis LSTM (LSTM, Bi-LSTM, Attention-LSTM, Attention-Bi-LSTM, serta varian gabungannya dengan lapisan konvolusi) dikembangkan dan dibandingkan, dengan class weighting untuk mengatasi ketidakseimbangan data. Evaluasi dilakukan melalui 5x5 repeated cross-validation (25 pengulangan) dan data uji terpisah (80:20) menggunakan metrik akurasi, presisi, recall, dan F1-score. Model Convolutional Attention-LSTM (CNN-A-LSTM) menunjukkan kinerja paling konsisten, dengan akurasi data uji 71,11% dan macro F1-score 0,719 (rata-rata validasi silang 0,732±0,037), mengungguli model dasar LSTM (akurasi 66,67%, F1-score 0,687; validasi silang 0,655±0,047). Paired t-test pada 25 pasangan validasi silang membuktikan ketujuh varian arsitektur meningkat signifikan dibanding model dasar (p<0,05), bahkan setelah koreksi Holm-Bonferroni dan Benjamini-Hochberg. Temuan ini menegaskan bahwa penambahan konvolusi, konteks dua arah, dan atensi pada representasi FastText secara konsisten meningkatkan klasifikasi keluhan mahasiswa, mendukung pengelolaan keluhan yang lebih cepat, objektif, dan efisien di perguruan tinggi.

Downloads

Download data is not yet available.

References

[1] Z. Kastrati, F. Dalipi, A. S. Imran, K. P. Nuci, and M. A. Wani, “Sentiment Analysis of Students’ Feedback with NLP and Deep Learning: A Systematic Mapping Study,” Applied Sciences 2021, Vol. 11, Page 3986, vol. 11, no. 9, p. 3986, Apr. 2021, doi: 10.3390/APP11093986.

[2] A. Koufakou, “Deep learning for opinion mining and topic classification of course reviews,” Education and Information Technologies 2023 29:3, vol. 29, no. 3, pp. 2973–2997, Jun. 2023, doi: 10.1007/S10639-023-11736-2.

[3] M. Birjali, M. Kasri, and A. Beni-Hssane, “A comprehensive survey on sentiment analysis: Approaches, challenges and trends,” Knowl. Based. Syst., vol. 226, p. 107134, Aug. 2021, doi: 10.1016/J.KNOSYS.2021.107134.

[4] E. M. Mercha and H. Benbrahim, “Machine learning and deep learning for sentiment analysis across languages: A survey,” Neurocomputing, vol. 531, pp. 195–216, Apr. 2023, doi: 10.1016/J.NEUCOM.2023.02.015.

[5] S. Susandri, S. Defit, and M. Tajuddin, “Enhancing Text Sentiment Classification with Hybrid CNN-BiLSTM Model on WhatsApp Group”, doi: 10.12720/jait.15.3.355-363.

[6] C. H. Lin and U. Nuha, “Sentiment analysis of Indonesian datasets based on a hybrid deep-learning strategy,” Journal of Big Data 2023 10:1, vol. 10, no. 1, pp. 88-, May 2023, doi: 10.1186/S40537-023-00782-9.

[7] D. G. Mandhasiya, H. Murfi, and A. Bustamam, “The hybrid of BERT and deep learning models for Indonesian sentiment analysis,” Indonesian Journal of Electrical Engineering and Computer Science, vol. 33, no. 1, pp. 591–602, Jan. 2024, doi: 10.11591/IJEECS.V33.I1.PP591-602.

[8] G. Z. Nabiilah, I. N. Alam, E. S. Purwanto, and M. F. Hidayat, “Indonesian multilabel classification using IndoBERT embedding and MBERT classification,” International Journal of Electrical and Computer Engineering (IJECE), vol. 14, no. 1, pp. 1071–1078, Feb. 2024, doi: 10.11591/ijece.v14i1.pp1071-1078.

[9] Z. Anwar, H. Afzal, A. Ahsan, N. Iltaf, and A. Maqbool, “A novel hybrid CNN-LSTM approach for assessing StackOverflow post quality,” Journal of Intelligent Systems, vol. 32, no. 1, Jan. 2023, doi: 10.1515/JISYS-2023-0057/XML.

[10] Z. L. Zhai, X. Zhang, F. F. Fang, and L. Y. Yao, “Text classification of Chinese news based on multi-scale CNN and LSTM hybrid model,” Multimed. Tools Appl., vol. 82, no. 14, pp. 20975–20988, Jun. 2023, doi: 10.1007/S11042-023-14450-W.

[11] K. Kaur and P. Kaur, “Improving BERT model for requirements classification by bidirectional LSTM-CNN deep model,” Computers and Electrical Engineering, vol. 108, p. 108699, May 2023, doi: 10.1016/J.COMPELECENG.2023.108699.

[12] E. M. Dharma, F. Lumban Gaol, H. L. H. S. Warnars, and B. Soewito, “The accuracy comparison among word2vec, glove, and fasttext towards convolution neural network (CNN) text classification,” Journal of Theoretical and Applied Information Technology, vol. 100, no. 2, pp. 349–359, Jan. 2022.

[13] S. Lee, Y. Yang, I. Aiyanyo, M. Keith, M. Boussougou, and D.-J. Park, “Attention-Based 1D CNN-BiLSTM Hybrid Model Enhanced with FastText Word Embedding for Korean Voice Phishing Detection,” Mathematics 2023, Vol. 11, Page 3217, vol. 11, no. 14, p. 3217, Jul. 2023, doi: 10.3390/MATH11143217.

[14] Y. Liu et al., “RoBERTa: A Robustly Optimized BERT Pretraining Approach,” Jul. 2019, Accessed: Aug. 06, 2026. [Online]. Available: http://arxiv.org/abs/1907.11692

[15] A. HE and M. ABISADO, “Text Sentiment Analysis of Douban Film Short Comments Based on BERT-CNN-BiLSTM-Att Model,” IEEE, vol. 12, Mar. 2024, Accessed: Aug. 06, 2026. [Online]. Available: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=10478509

[16] C. P. Chai, “Comparison of text preprocessing methods,” Nat. Lang. Eng., vol. 29, no. 3, pp. 509–553, May 2023, doi: 10.1017/S1351324922000213.

[17] I. D. Mienye, T. G. Swart, and G. Obaido, “Recurrent Neural Networks: A Comprehensive Review of Architectures, Variants, and Applications,” Information 2024, Vol. 15, Page 517, vol. 15, no. 9, p. 517, Aug. 2024, doi: 10.3390/INFO15090517.

[18] S. F. Abdhood, N. Omar, and S. Tiun, “Data augmentation for Arabic text classification: a review of current methods, challenges and prospective directions,” PeerJ Comput. Sci., vol. 11, 2025, doi: 10.7717/PEERJ-CS.2685/.

[19] A. A. Khan, O. Chaudhari, and R. Chandra, “A review of ensemble learning and data augmentation models for class imbalanced problems: Combination, implementation and evaluation,” Expert Syst. Appl., vol. 244, Jun. 2024, doi: 10.1016/J.ESWA.2023.122778;PAGE:STRING:ARTICLE/CHAPTER.

[20] M. Arslan and C. Cruz, “Business text classification with imbalanced data and moderately large label spaces for digital transformation,” Applied Network Science 2024 9:1, vol. 9, no. 1, pp. 11-, Apr. 2024, doi: 10.1007/S41109-024-00623-5.

[21] J. Mutinda, W. Mwangi, and G. Okeyo, “Sentiment Analysis of Text Reviews Using Lexicon-Enhanced Bert Embedding (LeBERT) Model with Convolutional Neural Network,” Applied Sciences 2023, Vol. 13, Page 1445, vol. 13, no. 3, p. 1445, Jan. 2023, doi: 10.3390/APP13031445.

[22] J. Wang, Z. Li, and C. Ma, “Research on News Text Classification Based on BERT-BiLSTM-TextCNN-Attention,” ACM International Conference Proceeding Series, pp. 295–298, Mar. 2024, doi: 10.1145/3672919.3672973.

[23] M. R. R. Rana, A. Nawaz, T. Ali, A. M. El-Sherbeeny, and W. Ali, “A BiLSTM-CF and BiGRU-based Deep Sentiment Analysis Model to Explore Customer Reviews for Effective Recommendations,” Engineering, Technology & Applied Science Research, vol. 13, no. 5, pp. 11739–11746, Oct. 2023, doi: 10.48084/ETASR.6278.

[24] K. Saputra, K. E. Saputra, and R. Riccosan, “Indonesian news article authorship attribution multilabel multiclass classification using IndoBERT,” IAES International Journal of Artificial Intelligence (IJ-AI), vol. 13, no. 4, pp. 4688–4694, Dec. 2024, doi: 10.11591/ijai.v13.i4.pp4688-4694.

[25] O. Karakaya and Z. H. Kilimci, “An efficient consolidation of word embedding and deep learning techniques for classifying anticancer peptides: FastText+BiLSTM,” PeerJ Comput. Sci., vol. 10, p. e1831, 2024, doi: 10.7717/PEERJ-CS.1831/SUPP-8.

[26] T. Sabri, S. Bahassine, O. El Beggar, and M. Kissi, “An improved Arabic text classification method using word embedding,” International Journal of Electrical and Computer Engineering, vol. 14, no. 1, pp. 721–731, Feb. 2024, doi: 10.11591/IJECE.V14I1.PP721-731.

Downloads

Published

2026-09-17

How to Cite

khotimah, khusnul, Cahyo Crysdian, & M.Imamudin. (2026). Identifikasi Komplain di Lingkungan Perguruan Tinggi Menggunakan Algorithm Convolutional Long Short Term Memory dan FastText Embedding. Joutica, 11(02), 193–204. https://doi.org/10.30736/jti.v11i02.1752

Issue

Section

Articles

Similar Articles

1 2 3 4 5 > >> 

You may also start an advanced similarity search for this article.