机器阅读理解是自然语言处理(NLP)领域的一个研究热点,目前大部分的研究是针对答案简短的问题,而具有长答案的问题,如描述类问题是现实世界无法避免的,因此有必要对该类问题进行研究。该文采用QU-NNs模型对阅读理解中描述类问题的解答进行了探索,其框架为嵌入层、编码层、交互层、预测层和答案后处理层。由于该类问题语义概括程度高,所以对问题的理解尤为重要,该文在模型的嵌入层和交互层中分别融入了问题类型和问题主题、问题焦点这三种问题特征,其中问题类型通过卷积神经网络进行识别,问题主题和问题焦点通过句法分析获得,同时采用启发式方法对答案中的噪音和冗余信息进行了识别。在相关数据集上对QU-NNs(Question Understanding-Neural Networks)模型进行了实验,实验表明加入问题特征和删除无关信息可使结果提高2%～10%。
Abstract
This paper explores the solutions to the description problems in reading comprehension using QU-NNs model whose frameworks are the Embedding layer, the Encoding layer, the Interaction layer, the Prediction layer, and the answer Post-processing layer. To deal with the high degree of semantic generalization of the questions, we integrate three features of question (question type, question topic, question focus) in the Encoding layer and the Interaction layer of the model to better understand the question. Specifically, the question type is identified by a convolutional neural network, and the question topic and question focus are obtained through syntactic analysis. Further, a heuristic method is designed to identify the noise and redundant information in the answer. Experiments show that adding question features and removing redundant information increased the performance by 2%～10%.
关键词
阅读理解 /
描述类问题 /
问题理解 /
神经网络
{{custom_keyword}} /
Key words
reading comprehension /
description problems /
question understanding /
neural network
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] Hermann Karl Moritz, et al. Teaching machines to read and comprehend[C]//Proceedings of Advances in Neural Information Processing Systems, 2015: 1693-1701.
[2] Yiming Cui, et al. Consensus attention-based neural networks for Chinese reading comprehension[C]//Proceeding of the 26th International Conference on Computational Linguistics, Japan, 2016: 1777-1786.
[3] Matthew Richardson, Christopher J C Burges, Erin Renshaw. MCTest: A challenge dataset for the open-domain machine comprehension of text[C]//Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing, USA, 2013: 193-203.
[4] lvaro Rodrigo, et al. Overview of CLEF QA Entrance Exams task 2015[R]. Working Notes of CLEF 2015, 2015: 539-544.
[5] Anselmo Peas, et al. Overview of CLEF QA Entrance Exams task 2014[R]. Working Notes of CLEF2014, 2014: 300-306.
[6] Pranav Rajpurkar, et al. SQuAD: 100,000+ questions for machine comprehension of text[C]//Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing, Texas, 2016: 2383-2392.
[7] Tri Nguyen, Song, et al. MS MARCO: A Human Generated Machine Reading Comprehension Dataset[J]. arXiv preprint arXiv: 1611.09268, 2016.
[8] Wei He, et al. DuReader: a Chinese machine reading comprehension dataset from real-world applications[J]. arXiv preprint arXiv: 1711.05073v4, 2017.
[9] 山西大学中文信息处理组. 863项目子课题“语言问题求解和答案生成关键技术及系统”中期进展报告——阅读理解部分, 2016.
[10] Weissenborn D, Wiese G, Seiffe L. FastQA: A simple and efficient neural architecture for question answering[J]. arXiv preprint arXiv: 1703.04816v1, 2017.
[11] Zhang J, Zhu X, Chen Q, et al. Exploring Question Understanding and Adaptation in Neural-Network-Based Question Answering[J]. arXiv preprint arXiv: 1703.04617v2, 2017.
[12] Adams Wei Yu, et al. QANet: Combining local convolution with global self-attention for reading comprehension[J]. arXiv preprint arXiv: 1804.09541v1, 2018.
[13] Chen D, Bolton J, Manning C D. A thorough examination of the CNN/Daily Mail reading comprehension task[C]//Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics, 2016: 2358-2367.
[14] Kadlec Rudolf, et al. Text understanding with the attention sum reader network[C]//Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics, 2016: 908-918.
[15] Bhuwan Dhingra, et al. Gated-attention readers for text comprehension[C]//Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, 2016: 1832-1846.
[16] Yiming Cui, et al. Attention-over-attention neural networks for reading comprehension[C]//Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, 2016: 593-602.
[17] Minjoon Seo, et al. Bidirectional attention flow for machine comprehension[J]. arXiv preprint arXiv: 1611.01603v6, 2016.
[18] Microsoft Asia Natural Language Computing Group. R-net: Machine reading comprehension with self-matching networks. 2017.
[19] Shuohang Wang, Jing Jiang. Machine comprehension using match-LSTM and answer pointer[C]//Proceedings of the ICLR, 2016: 1-15.
[20] Caiming Xiong, Victor Zhong, Richard Socher. Dynamic coattention networks for question answering[J]. arXiv preprint arXiv: 1611.01604v4, 2018.
[21] Robin Jia, Percy Liang. Adversarial examples for evaluating reading comprehension systems[C]//Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, 2017: 2021-2031.
[22] Ulf Hermjakob. Parsing and question classification for question answering[C]//Proceedings of the Workshop on Open-Domain Question Answering at ACL, 2001: 1-6.
[23] Huizhong Duan, et al. Searching questions by identifying question topic and question focus[C]//Proceedings of the Meeting of the Association for Computational Linguistics, 2008: 156-164.
[24] Kishore Papineni, et al. BLEU: A method for automatic evaluation of machine translation[C]//Proceedings of the 40th Meeting of the Association for Computational Linguistics, 2002: 311-318.
[25] Lin, Chin Yew. ROUGE: A package for automatic evaluation of summaries[C]//Proceedings of the Workshop on Text Summarization Branches Out, 2004.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
国家自然科学基金(61673248,61806117);山西省研究生联合培养基地人才培养项目(2018JD02)
{{custom_fund}}
PDF(1455 KB)
