本碩士論文旨在發展能夠帶給使用者情緒支持系統，專為失智症患者家屬設計。目的是利用人工智慧系統和自然語言預處理技術，進行初步的失智症情緒描述分類。
失智症患者逐年增加，家屬的挑戰亦然。失智症患者可能隨著時間遺忘回憶及痛苦但家屬不會，家屬需陪伴患者直至患者行為功能逐漸喪失。本研究發現，家屬於網路上的提問、發文等並非全部都不知道答案，有部分是希望有人理解它的處境是如此辛苦、難受，藉由表達、傾聽之後心情得以平復，又可以繼續堅持。
正因失智症患者家屬的情緒負擔很重，本研究所開發的失智症文章情緒描述分類工具，提供失智症患者家屬一個不用借助他人也可以獲得情緒反饋的工具。失智症情緒描述分類系統的系統建立搭配人工智慧的應用，能夠確保分類結果的信度與效度。
本研究使用新北市政府所出版的《失智100問》失智症相關問答集，搭配作者於PTT及Dcard所收集的失智症相關文章，進行情緒相關描述的抽取並將其類別標註及訓練。並針對語料集進行預訓練模型的建置，並將其分為訓練、測試兩組語料，並使用資料增強技術加強模型訓練以提高評分效果，探討情緒描述用於失智症患者家屬描述分類之可行性，以支持未來擁有情緒支持功能之系統研究的開發。
研究數據顯示，本研究所訓練的情緒描述分類預訓練模型，用於本研究的語料集可成功訓練及進行分類，顯示基於當代自然語言處理技術來發展失智症情緒描述分類工具之研究之可行性。

This master’s thesis aims to develop an emotional support system designed specifically for family caregivers of dementia patients. The purpose is to utilize artificial intelligence (AI) systems and natural language preprocessing techniques to conduct a preliminary classification of emotional descriptions related to dementia.
With the growing number of dementia patients each year, the challenges faced by their families have also increased. While dementia patients may gradually forget their memories and suffering over time, their family caregivers do not. They must accompany the patients as their behavioral functions gradually deteriorate. This study finds that not all questions and posts shared online by caregivers lack answers; rather, some are expressions of their emotional struggles, seeking understanding and relief through sharing and listening, enabling them to continue their caregiving journey.
Given the heavy emotional burden on dementia caregivers, this study has developed a tool for classifying emotional descriptions in dementia-related texts. This tool provides caregivers with a means of receiving emotional feedback without relying on others. By leveraging AI applications, the system ensures the reliability and validity of classification results.
The research utilizes the Dementia 100 Questions, a dementia-related Q&A collection published by the New Taipei City Government, along with dementia-related posts gathered from PTT and Dcard. Emotional descriptions were extracted, categorized, and annotated for training. A pretrained model was constructed using this dataset, which was divided into training and testing subsets. Data augmentation techniques were employed to enhance model training and improve performance scores. The study explores the feasibility of classifying emotional descriptions for dementia caregivers, supporting the development of future systems with emotional support functionalities.
Research data indicate that the pretrained emotional description classification model successfully performs classification tasks on the research dataset, demonstrating the feasibility of developing a dementia-related emotional description classification tool using contemporary natural language processing techniques.

Acheampong, F. A., Nunoo-Mensah, H., & Chen, W. (2021). Transformer models for text-based emotion detection: a review of BERT-based approaches. Artificial Intelligence Review, 54(8), 5789-5829.
Aithal, S. G., Rao, A. B., & Singh, S. (2021). Automatic question-answer pairs generation and question similarity mechanism in question answering system. Applied Intelligence, 1-14.
Biswas, P., Sharan, A., & Kumar, R. (2014). Question Classification using syntactic and rule based approach. 2014 International Conference on Advances in Computing, Communications and Informatics (ICACCI), pp. 1033-1038, doi: 10.1109/ICACCI.2014.6968434.
Chen, X., Cong, P., & Lv, S. (2022). A long-text classification method of Chinese news based on BERT and CNN. IEEE Access, 10, 34046-34057.
Choi, E., He, H., Iyyer, M., Yatskar, M., Yih, W. T., Choi, Y., ... & Zettlemoyer, L. (2018). QuAC: Question answering in context. arXiv preprint arXiv:1808.07036.
Devlin, J., Chang, M. W., Lee, K., & Toutanova, K. (2018). Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805.
Do, P., Phan, T. H., & Gupta, B. B. (2021). Developing a Vietnamese tourism question answering system using knowledge graph and deep learning. Transactions on Asian and Low-Resource Language Information Processing, 20(5), 1-18.
Dong, K., Liu, Y., Xu, F., & Liu, P. (2023). DCAT: Combining Multi-semantic Dual-channel Attention Fusion for Text Classification. IEEE Intelligent Systems.
Elkaim, L. M., Niazi, F., Levett, J. J., Bokhari, R., Gorodetsky, C., Breitbart, S., ... & Ibrahim, G. M. (2022). Deep brain stimulation in children and youth: perspectives of patients and caregivers gleaned through Twitter. Neurosurgical Focus, 53(4), E11.
Farrar, M., Lundt, L., Franey, E., & Yonan, C. (2021). Patient perspective of tardive dyskinesia: results from a social media listening study. BMC psychiatry, 21, 1-8.
Guo, Q., Cao, S., & Yi, Z. (2022). A medical question answering system using large language models and knowledge graphs. International Journal of Intelligent Systems, 37(11), 8548-8564.
Harb, J. G., Ebeling, R., & Becker, K. (2020). A framework to analyze the emotional reactions to mass violent events on Twitter and influential factors. Information Processing & Management, 57(6), 102372.
Hendrycks, D., Liu, X., Wallace, E., Dziedzic, A., Krishnan, R., & Song, D. (2020). Pretrained transformers improve out-of-distribution robustness. arXiv preprint arXiv:2004.06100.
Hotchkiss, J., Ridderman, E., & Buftin, W. (2024). Overall US hospice quality according to decedent caregivers—Natural Language Processing and sentiment analysis of 3389 online caregiver reviews. American Journal of Hospice and Palliative Medicine®, 41(5), 527-544.
Howard, J., & Ruder, S. (2018). Universal language model fine-tuning for text classification. arXiv preprint arXiv:1801.06146.
Huang, Z., Xu, S., Hu, M., Wang, X., Qiu, J., Fu, Y., ... & Wang, C. (2020). Recent trends in deep learning based open-domain textual question answering systems. IEEE Access, 8, 94341-94356.
Kalyan, K. S., Rajasekharan, A., & Sangeetha, S. (2021). Ammus: A survey of transformer-based pretrained models in natural language processing. arXiv preprint arXiv:2108.05542.
Karn, A. L., Karna, R. K., Kondamudi, B. R., Bagale, G., Pustokhin, D. A., Pustokhina, I. V., & Sengan, S. (2023). RETRACTED ARTICLE: Customer centric hybrid recommendation system for E-Commerce applications by integrating hybrid sentiment analysis. Electronic commerce research, 23(1), 279-314.
Kastrati, Z., Ahmedi, L., Kurti, A., Kadriu, F., Murtezaj, D., & Gashi, F. (2021). A deep learning sentiment analyser for social media comments in low-resource languages. Electronics, 10(10), 1133.
Lee, C., Han, D., Han, K., & Yi, M. (2022). Improving graph-based movie recommender system using cinematic experience. Applied Sciences, 12(3), 1493.
Lei, Z., Haq, A. U., Zeb, A., Suzauddola, M., & Zhang, D. (2021). Is the suggested food your desired?: Multi-modal recipe recommendation with demand-based knowledge graph. Expert Systems with Applications, 186, 115708.
Lin, S. Y., Kung, Y. C., & Leu, F. Y. (2022). Predictive intelligence in harmful news identification by BERT-based ensemble learning model with text sentiment analysis. Information Processing & Management, 59(2), 102872.
Mallikarjuna, C., & Sivanesan, S. (2022). Question Classification Using Limited Labelled Data. Information Processing & Management, 59(6), 103094.
Mazza, M., Piperis, M., Aasaithambi, S., Chauhan, J., Sagkriotis, A., & Vieira, C. (2022). Social media listening to understand the lived experience of individuals in Europe with metastatic breast cancer: a systematic search and content analysis study. Frontiers in Oncology, 12, 863641.
Ménard, A., O’Sullivan, T., Mulvey, M., Belanger, C., & Fraser, S. (2024). Perceptions of hospital care for persons with dementia during the COVID-19 pandemic: a social media sentiment analysis. The Gerontologist, 64(7), gnad155.
Mikolov, T., Chen, K., Corrado, G., & Dean, J. (2013). Efficient estimation of word representations in vector space. arXiv preprint arXiv:1301.3781.
Mohasseb, A., Bader-El-Den, M., & Cocea, M. (2018). Question categorization and classification using grammar based approach. Information Processing & Management, 54(6), 1228-1243.
Pota, M., Esposito, M., De Pietro, G., & Fujita, H. (2020). Best practices of convolutional neural networks for question classification. Applied Sciences, 10(14), 4710.
Qiao, C., Huang, B., Niu, G., Li, D., Dong, D., He, W., ... & Wu, H. (2018). A New Method of Region Embedding for Text Classification. In ICLR (Poster).
Raposo, G., Ribeiro, R., Martins, B., & Coheur, L. (2022, April). Question rewriting? Assessing its importance for conversational question answering. In European Conference on Information Retrieval (pp. 199-206). Cham: Springer International Publishing.
Ray, B., Garain, A., & Sarkar, R. (2021). An ensemble-based hotel recommender system using sentiment analysis and aspect categorization of hotel reviews. Applied Soft Computing, 98, 106935.
Ray, S. K., Singh, S., & Joshi, B. P. (2010). A semantic approach for question classification using WordNet and Wikipedia. Pattern recognition letters, 31(13), 1935-1943.
Saint-Dizier, P. (2014). Advanced Question-Answering and Discourse Semantics. In Computational Linguistics: Concepts, Methodologies, Tools, and Applications (pp. 598-616). IGI Global.
Sarrouti, M., & El Alaoui, S. O. (2020). SemBioNLQA: A semantic biomedical question answering system for retrieving exact and ideal answers to natural language questions. Artificial intelligence in medicine, 102, 101767.
Savini, E., & Caragea, C. (2022). Intermediate-task transfer learning with BERT for sarcasm detection. Mathematics, 10(5), 844.
Sazon, H., Catapan, S. D. C., Rahimi, A., Canfell, O. J., & Kelly, J. (2024). How do Twitter users feel about telehealth? A mixed‐methods analysis of experiences, perceptions and expectations. Health Expectations, 27(1), e13927.
Sun, C., Qiu, X., Xu, Y., & Huang, X. (2019). How to fine-tune bert for text classification?. In Chinese Computational Linguistics: 18th China National Conference, CCL 2019, Kunming, China, October 18–20, 2019, Proceedings 18 (pp. 194-206). Springer International Publishing.
Sutskever, I., Vinyals, O., & Le, Q. V. (2014). Sequence to sequence learning with neural networks. Advances in neural information processing systems, 27.
Tanana, M. J., Soma, C. S., Kuo, P. B., Bertagnolli, N. M., Dembe, A., Pace, B. T., ... & Imel, Z. E. (2021). How do you feel? Using natural language processing to automatically rate emotion in psychotherapy. Behavior research methods, 1-14.
Tofighi, B., El Shahawy, O., Segoshi, A., Moreno, K. P., Badiei, B., Sarker, A., & Krawczyk, N. (2020). Assessing perceptions about medications for opioid use disorder and Naloxone on Twitter. Journal of Addictive Diseases, 39(1), 37-45.
Tong, J., Wang, Z., & Rui, X. (2022). A Multimodel-Based Deep Learning Framework for Short Text Multiclass Classification with the Imbalanced and Extremely Small Data Set. Computational Intelligence and Neuroscience, 2022.
Van-Tu, N., & Anh-Cuong, L. (2016). Improving question classification by feature extraction and selection. Indian Journal of Science and Technology, 9(17), 1-8.
Wolf, T., Debut, L., Sanh, V., Chaumond, J., Delangue, C., Moi, A., ... & Rush, A. M. (2020). Transformers: State-of-the-art natural language processing. In Proceedings of the 2020 conference on empirical methods in natural language processing: system demonstrations (pp. 38-45).
Yazdinejad, A., Rabieinejad, E., Hasani, T., & Srivastava, G. (2023). A bert-based recommender system for secure blockchain-based cyber physical drug supply chain management. Cluster Computing, 26(6), 3389-3403.
Zhang, D., Li, J., Xie, Y., & Wulamu, A. (2023). Research on performance variations of classifiers with the influence of pre-processing methods for Chinese short text classification. Plos one, 18(10), e0292582.
Zhang, H., Shan, Y., Jiang, P., & Cai, X. (2022). A Text Classification Method Based on BERT-Att-TextCNN Model. In 2022 IEEE 5th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC) (Vol. 5, pp. 1731-1735). IEEE.
Zhao, A., & Yu, Y. (2021). Knowledge-enabled BERT for aspect-based sentiment analysis. Knowledge-Based Systems, 227, 107220.
Zheng, S., Li, Y., Chen, S., Xu, J., & Yang, Y. (2020). Predicting drug–protein interaction using quasi-visual question answering system. Nature Machine Intelligence, 2(2), 134-140.