Abstract

Aspect-based sentiment analysis (ABSA) of student comments about universities presents a significant challenge in the field of natural language processing (NLP). These comments often exhibit complex structures, addressing multiple aspects such as facilities, teaching quality, campus environment, and administrative services, and are frequently accompanied by diverse sentiment expressions. This study aims to develop an index generation algorithm leveraging a transformer model—specifically bidirectional encoder representations from transformers (BERT)—to enhance the efficiency and accuracy of sentiment interpretation related to these aspects. The proposed approach employs semantic indexing to more effectively capture the relationship between specific aspects and the sentiments expressed in student reviews. Semantic indexing is particularly designed to address critical challenges such as contextual ambiguity and variability in sentiment expressions often found in unstructured textual data. This study also compares the effectiveness of the transformer-based approach with traditional methods to highlight its advantages in terms of analytical accuracy and contextual comprehension. The expected outcomes include the development of a model capable of providing in-depth insights into student sentiment across various dimensions of university experience. The application of this model is anticipated to support university administrators in making data-driven decisions to enhance service quality and student satisfaction, while also contributing to the theoretical development of ABSA in the context of higher education.

Keywords

Aspect-based sentiment analysis, Transformer models, BERT, Semantic indexing, Student feedback analysis.

References

[1] Marutho D, Rustad S. Optimizing aspect-based sentiment analysis using sentence embedding transformer, bayesian search clustering, and sparse attention mechanism. Journal of Open Innovation: Technology, Market, and Complexity. 2024; 10(1):1-19.

[2] Ahmadian H, Abidin TF, Riza H, Muchtar K. Hybrid models for emotion classification and sentiment analysis in Indonesian language. Applied Computational Intelligence and Soft Computing. 2024; 2024(1):1-17.

[3] Daqiqil I, Saputra H, Syamsudhuha KR, Andriyani Y. Sentiment analysis of student evaluation feedback using transformer-based language models. Indonesian Journal of Electrical Engineering and Computer Science. 2024; 36:1127-39.

[4] Zhu X, Zhu Y, Zhang L, Chen Y. A BERT-based multi-semantic learning model with aspect-aware enhancement for aspect polarity classification. Applied Intelligence. 2023; 53(4):4609-23.

[5] Mahmud T, Ptaszynski M, Masui F. Exhaustive study into machine learning and deep learning methods for multilingual cyberbullying detection in Bangla and Chittagonian texts. Electronics. 2024; 13(9):1-36.

[6] Fernandes R, Willison J, Boyle C. Characteristics of facilitated critical thinking when students listen to and speak English as an additional language in Indonesia. Thinking Skills and Creativity. 2024; 52:1-17.

[7] Kristiyanti DA, Sanjaya SA, Tjokro VC, Suhali J. Dealing imbalance dataset problem in sentiment analysis of recession in Indonesia. IAES International Journal of Artificial Intelligence (IJ-AI). 2024; 13(2):2060-72.

[8] Bhopale AP, Tiwari A. Transformer based contextual text representation framework for intelligent information retrieval. Expert Systems with Applications. 2024; 238:121629.

[9] Chauhan GS, Nahta R, Meena YK, Gopalani D. Aspect based sentiment analysis using deep learning approaches: a survey. Computer Science Review. 2023; 49:100576.

[10] Wang A, Qi Y, Baiyila D. C-BERT: a mongolian reverse dictionary based on fused lexical semantic clustering and BERT. Alexandria Engineering Journal. 2025; 111:385-95.

[11] Jazuli A, Widowati W, Kusumaningrum R. Aspect-based sentiment analysis on student reviews using the indo-bert base model. In E3S web of conferences 2023. EDP Sciences.

[12] Wu H, Zhou D, Sun C, Zhang Z, Ding Y, Chen Y. LSOIT: lexicon and syntax enhanced opinion induction tree for aspect-based sentiment analysis. Expert Systems with Applications. 2024; 235:121137.

[13] Glenn A, Lacasse P, Cox B. Emotion classification of Indonesian tweets using bidirectional LSTM. Neural Computing and Applications. 2023; 35(13):9567-78.

[14] Wu Y. The construction of higher education evaluation index system based on vector analysis. Journal of Electrical Systems. 2024; 20(3):448–58.

[15] Ma X. Artificial intelligence-driven education evaluation and scoring: comparative exploration of machine learning algorithms. Journal of Intelligent Systems. 2024; 33(1):1-19.

[16] Antoine S, Grisoni L, Tomasello R, Pulvermüller F. The prediction potential indexes the meaning and communicative function of upcoming utterances. Cortex. 2024; 177:346-62.

[17] Marzi G, Balzano M, Marchiori D. K-Alpha calculator–krippendorff's alpha calculator: a user-friendly tool for computing Krippendorff’s alpha inter-rater reliability coefficient. MethodsX. 2024; 12:102545.

[18] Souza FC, Nogueira RF, Lotufo RA. BERT models for Brazilian Portuguese: pretraining, evaluation and tokenization analysis. Applied Soft Computing. 2023; 149:110901.

[19] Ramasamy M, Elangovan M. Optimized neural attention mechanism for aspect-based sentiment analysis framework with optimal polarity-based weighted features. Knowledge and Information Systems. 2024; 66(4):2501-35.

[20] Laksana EA, Suryana A, Rosita A, Heryono H. Evaluation of E-learning activity effectiveness in higher education through sentiment analysis by using naïve bayes classifier. Sisforma. 2018; 5(1):22-7.

[21] Chakraborty K, Bhattacharyya S, Bag R, Mršić L. Sentiment analysis on labeled and unlabeled datasets using BERT architecture. Soft Computing. 2024; 28(15):8623-40.

[22] Zhao C, Feng R, Sun X, Shen L, Gao J, Wang Y. Enhancing aspect-based sentiment analysis with BERT-driven context generation and quality filtering. Natural Language Processing Journal. 2024; 7:100077.

[23] Narayan V, Ganapathisamy S. Learning analytics with correlation-based SAN-LSTM mechanism for formative evaluation and improved online learning. International Journal of Advanced Technology and Engineering Exploration. 2024; 11(112):373-87.

[24] Zhang J, Yan W, Zhang Q, Li Z, Liang L, Zuo M, et al. Umami-BERT: an interpretable BERT-based model for umami peptides prediction. Food Research International. 2023; 172:113142.

[25] Alnabhan MQ, Branco P. BERTGuard: two-tiered multi-domain fake news detection with class imbalance mitigation. Big Data and Cognitive Computing. 2024; 8(8):1-26.

[26] Zhou P, Jiang X, Zhao S. Unsupervised technical phrase extraction by incorporating structure and position information. Expert Systems with Applications. 2024; 245:123140.

[27] Zhou X, Zhang T, Cheng C, Song S. Dynamic multichannel fusion mechanism based on a graph attention network and BERT for aspect-based sentiment classification. Applied Intelligence. 2023; 53(6):6800-13.

[28] Liu W, Sun Y, Yu B, Wang H, Peng Q, Hou M, et al. Automatic text summarization method based on improved textrank algorithm and k-means clustering. Knowledge-Based Systems. 2024; 287:111447.

[29] Zou A, Hao W, Jin D, Zou S, Zheng Y, Sun F, et al. An efficient corpus indexer for dynamic corpora retrieval. Expert Systems with Applications. 2024; 254:124306.

[30] Ma Y, Huang R, Yan M, Li G, Wang T. Attention-based local mean k-nearest centroid neighbor classifier. Expert Systems with Applications. 2022; 201:117159.

Cite this article