情感分析在自然語言(NLP)領域中是熱門的研究主題，主要的研究都是圍繞情緒做分析，然而造成情緒的原因也值得深入分析。情緒與原因的組合提取，旨在於文章中提取描述情緒的子句和造成該情緒的原因子句，並配對兩者成為情緒原因組合。這樣的組合可以幫助人在分析情緒時更為全面，例如在使用者經驗和心理諮商等都有其應用價值。
　　本研究中，我們探討使用文字理解模型來解決此任務，在過去的論文中，大部分都將此情感原因的組合提取視為子句分類任務。我們主要的想法是將原先的子句分類任務轉換成問答任務來更深入理解文章，並且採用了兩階段的方法，分別是情緒與原因子句的提取以及篩選組合。此外，我們也提出資料增強策略來輔助原先訓練資料不足所帶來的準確率不佳問題。於標準資料集驗證實驗中，實驗結果展現我們所提出的方法在F1指標中為目前相關研究最佳的結果。

　　Sentiment analysis in Natural Language Processing (NLP) has been a popular topic and most studies pay more attention to the emotions analysis. However, in addition to analyzing emotion itself, the reasons that cause the emotion also deserve attention. Emotion-cause pair extraction (ECPE) aims to extract the emotion clauses and pair with corresponding causes in an excerpt to assist people understand human sentiment. The extracted causes are also valuable for many applications, such as user experiences, counseling psychology, etc.
　　In this thesis, we study the ECPE problem by employing machine reading comprehension model to extract emotions and causes in an excerpt. In the past, the ECPE was modeled as a clause-level classification task. In comparison, our idea is to transform the original clause-level classification task into the question answering task in the machine reading comprehension model. Based on such an idea, we propose a two-stage approach with specific data augmentation. The first stage is to leverage machine reading comprehension model to extract emotions and causes from a given excerpt and the second stage is to pair extracted emotion clauses and cause clauses for discovering final emotion-cause pair result. The evaluation based on benchmarking dataset demonstrate the effectiveness of the proposed approach; we push the state-of-the-art results from 61% to 65% in terms of F1 scores.

Acknowledgement--------------------------------1
摘要-------------------------------------------2
Abstract---------------------------------------3
1 Introduction------------------------------7
2 Related Work-----------------------------11
2.1 Emotion Cause Extraction---------------11
2.2 Emotion Cause Pair Extraction----------12
3 Preliminaries----------------------------14
3.1 Task Description-----------------------14
3.2 BERT Overview--------------------------14
4 Method-----------------------------------17
4.1 Emotion-Pair Extraction----------------17
4.2 Emotion / Cause Clauses Extraction-----18
4.2.1 Independent BERT-QA------------------18
4.2.2 Interactive BERT-QA------------------20
4.2.3 Data Augmentation--------------------21
4.3 Pairing and Filtering------------------22
5 Experiments------------------------------25
5.1 Dataset and Metrics--------------------25
5.2 Implementation Details-----------------26
5.3 Comparison-----------------------------27
5.4 Ablation Study-------------------------29
5.4.1 Interactive Structure----------------29
5.4.2 Task Transformation------------------31
5.4.3 Data Augmentation--------------------33
5.4.4 Effect of Filtering------------------34
6 Conclusion-------------------------------35
Reference-------------------------------------36

[1] Rui Xia and Zixiang Ding. 2019. Emotion-Cause Pair Extraction: A New Task to Emotion Analysis in Texts. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 1003-1012.
[2] Lin Gui, Jianna Hu, Yulan He, Ruifeng Xu, Qin Lu and Jiachen Du. 2017. A Question Answering Approach to Emotion Cause Extraction. In Empirical Methods in Natural Language Processing, pages 1639-1649.
[3] Lin Gui, Dongyin Wu, Ruifeng Xu, Qin Lu and Yu Zhou. 2016a. Event-driven Emotion Cause Extraction with Corpus Construction. In Empirical Methods in Natural Language Processing, pages 1639-1649.
[4] Sophia Yat Mei Lee, Ying Chen and Chu-Ren Huang. 2010. A Text-driven Rule-based System for Emotion Cause Detection. In Proceedings of the NAACL HLT 2010 Workshop on Computational Approaches to Analysis and Generation of Emotion in Text, pages 45-53.
[5] Irene Russo, Tommaso Caselli, Francesco Rubino, Ester Boldrini and Patricio Martinez-Barco. 2011. EMOCause: An Easy-agaptable Approach to Emotion Cause Contexts. In Proceedings of the 2nd Workshop on Computational Approaches to Subjectivity and Sentiment Analysis, ACL-HLT 2011, pages 153–160.
[6] Weiyuan Li and Hua Xu. 2014. Text-based Emotion Classification Using Emotion Cause Extraction. Expert Systems with Applications, 41(4): 1742-1749.
[7] Lin Gui, Li Yuan, Ruifeng Xu, Bin Liu, Qin Lu and Yu Zhou. 2014. Emotion Cause Detection with Linguistic Construction in Chinese Weibo Text. In Natural Language Processing and Chinese Computing, pages 457-464.
[8] Diman Ghazi, Diana Inkpen and Stan Szpakowicz. 2015. Detecting emotion stimuli in emotion-bearing sentences. In Computational Linguistics and Intelligent Text Processing, pages 152-165.
[9] Shuangyong Song and Yao Meng. 2015. Detecting Concept-level Emotion Cause in Microblogging. In Proceedings of the 24th International Conference on World Wide Web, pages 119-120.
[10] Lin Gui, Ruifeng Xu, Qin Lu, Dongyin Wu and Yu Zhou. 2016b. Emotion Cause Extraction, a Challenging Task with Corpus Construction. In Chinese National Conference on Social Media Processing, pages 98-109.
[11] Shuntaro Yada, Kazushi Ikeda, Keiichiro Hoashi and Kyo Kageura. 2017. A Bootstrap Method for Automatic Rule Acquisition on Emotion Cause Extraction. In IEEE International Conference on Data Mining Workshops, pages 414-421.
[12] Ying Chen, Wenjun Hou, Xiyao Cheng and Shoushan Li. 2018b. Joint Learning for Emotion Classification and Emotion Cause Detection. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 646-651.
[13] Xiangju Li, Kaisong Song, Shi Feng, Daling Wang and Yifei Zhang, 2018. A Co-attention Neural Network Model for Emotion Cause Analysis with Emotional Context Awareness. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 4752-4757.
[14] Rui Xia, Mengran Zhang and Zixiang Ding, 2019. RTHN: A RNN-Transformer Hierarchical Network for Emotion Cause Extraction. In Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence, pages 5285-5291.
[15] Ashish Vanwani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Lilon Jones, Aidan N.Gomez, Łukasz Kaiser and Illia Polosukhin. 2017. Attention Is All You Need. In Proceedings of the 31st International Conference on Neural Information Processing Systems, pages 6000-6010.
[16] Chuang Fan, Hongyu Yan, Jiachen Du, Lin Gui, Lidong Bing, Min Yang, Ruifeng Xu, Ruibin Mao. 2019. A Knowledge Regularized Hierarchical Approach for Emotion Cause Analysis. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, pages 5618–5628.
[17] Xiangju Li, Shi Feng, Daling Wang and Yifei Zhang. 2019. Context-aware emotion cause analysis with multi-attention-based neural network. In Knowledge-Based Systems, pages 205-218.
[18] Jiaxing Hu, Shumin Shi and Heyan Huang. 2019. Combining External Sentiment Knowledge for Emotion Cause Detection. In Natural Language Processing and Chinese Computing, pages 711-722.
[19] Jacob Devlin, Ming-Wei Chang, Kenton Lee and Kristina Toutanova. 2019. BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. In Proceedings of NAACL-HLT 2019, pages 4171–4186.
[20] Xin Zhao, Jing Jiang, Jing He, Yang Song, Palakorn Achanauparp, Ee-Peng Lim and Xiaoming Li. 2011. Topical Keyphrase Extraction from Twitter. In Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, pages 379-388.