Automatic Question and Answering Supported by Abductive Learning and Its Application

doi:10.3778/j.issn.1002-8331.2306-0207

Abstract

Abstract: Automatic question and answering (QA) techniques can provide users with fast and accurate information retrieval and problem-solving services. However, the answers generated by current common methods are often inaccurate, incomplete and unnatural. To this end, an abductive learning-based automatic QA method is proposed, which employs knowledge graph-based QA to infer and optimize the generation-based QA. Furthermore, the overall abductive learning framework is employed to optimize entity recognition and relationship extraction methods. The proposed method is applied to self-learning of the “Data Structures” course. The results show that the abductive learning-based automatic QA method can overcome the shortcomings of both the generation-based QA and knowledge graph-based QA, and improve the accuracy of the QA system.

Key words: automatic question and answering (QA), abductive learning, knowledge graph QA, generative QA

摘要： 自动问答技术可以为用户提供快速且准确的信息检索和问题解答服务。然而，目前常见方法生成的答案存在不准确和不完整的问题，以及实体识别和关系抽取效果不准确，且答案不够自然。为此，提出基于反绎学习的自动问答方法，使用基于知识图谱的问答推理优化基于生成的问答，进一步从整体的反绎学习框架角度来优化实体识别和关系抽取方法，并将所提方法应用于《数据结构》课程的学习。结果表明，基于反绎学习的自动问答方法，可以改进基于生成的问答和基于知识图谱的问答两者的不足，提高问答系统的准确性。

关键词: 自动问答, 反绎学习, 知识图谱问答, 生成式问答

ZHANG Peng, HAO Guosheng, WANG Xia, XU Wenyang, ZHU Yi. Automatic Question and Answering Supported by Abductive Learning and Its Application[J]. Computer Engineering and Applications, 204, 60(17): 139-147.

张鹏, 郝国生, 王霞, 许文阳, 祝义. 反绎学习支持下的自动问答及其应用[J]. 计算机工程与应用, 204, 60(17): 139-147.

References

[1] MA X, HOVY E. End-to-end sequence labeling via bi-directional LSTM-CNNs-CRF[J]. arXiv:1603.01354, 2016.
[2] JI H, KE P, HUANG S, et al. Language generation with multi-hop reasoning on commonsense knowledge graph[J]. arXiv:2009.11692, 2020.
[3] LI C, BI B, YAN M, et al. Addressing semantic drift in generative question answering with auxiliary extraction[C]//Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing, 2021: 942-947.
[4] SHUM H Y, HE X, LI D. From Eliza to XiaoIce: challenges and opportunities with social chatbots[J]. Frontiers of Information Technology & Electronic Engineering, 2018, 19: 10-26.
[5] HAN F X, NIU D, LAI K, et al. Inferring search queries from web documents via a graph-augmented sequence to attention network[C]//Proceedings of the World Wide Web Conference, 2019: 2792-2798.
[6] DANON G, LAST M. A syntactic approach to domain-specific automatic question generation[J]. arXiv:1712.09827, 2017.
[7] BI X, NIE H, ZHANG X, et al. Unrestricted multi-hop reasoning network for interpretable question answering over knowledge graph[J]. Knowledge-Based Systems, 2022, 243: 108515.
[8] 郑泳智, 朱定局, 吴惠粦, 等. 知识图谱问答领域综述[J]. 计算机系统应用, 2022, 31(4): 1-13.
ZHENG Y Z, ZHU D J, WU H L, et al. Overview on knowledge graph question answering[J]. Computers Systems Applications, 2022, 31(4): 1-13.
[9] MOHIT B. Named entity recognition[J]. Natural language Processing of Semitic Languages, 2014: 221-245.
[10] JI G, LIU K, HE S, et al. Distant supervision for relation extraction with sentence-level attention and entity descriptions[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2017.
[11] KHOT T, SABHARWAL A, CLARK P. Answering complex questions using open information extraction[J]. arXiv:1704.
05572, 2017.
[12] LU W, NG H T, LEE W S, et al. A generative model for parsing natural language to meaning representations[C]//Proceedings of the 2008 Conference on Empirical Methods in Natural Language Processing, 2008: 783-792.
[13] ZHOU Z H. Abductive learning: towards bridging machine learning and logical reasoning[J]. Science China Information Sciences, 2019, 62: 1-3.
[14] CAI L W, DAI W Z, HUANG Y X, et al. Abductive learning with ground knowledge base[C]//Proceedings of the 13th International Joint Conference on Artificial Intelligence, 2021: 1815-1821.
[15] HUANG Y X, DAI W Z, CAI L W, et al. Fast abductive learning by similarity-based consistency optimization[J]. Proceedings of the Neural Information Processing Systems, 2021, 34: 26574-26584.
[16] 李锦烨, 黄瑞章, 秦永彬, 等. 基于反绎学习的裁判文书量刑情节识别[J]. 计算机应用, 2022, 42(6): 1802-1807.
LI J Y, HUANG R Z, QIN Y B, et al. Recognition of sentencing circumstances in adjudication documents based on abductive learning[J]. Journal of Computer Applications, 2022, 42(6): 1802-1807.
[17] FUNAHASHI K, NAKAMURA Y. Approximation of dynamical systems by continuous time recurrent neural networks[J]. Neural Networks, 1993, 6(6): 801-806.
[18] VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems, 2017: 6000-6010.
[19] SHAO L, GOUWS S, BRITZ D, et al. Generating high-quality and informative conversation responses with sequence-to-sequence models[J]. arXiv:1701.03185, 2017.
[20] SUTSKEVER I, VINYALS O, LE Q V. Sequence to sequence learning with neural networks[J]. arXiv:1409.3215, 2014.
[21] XIONG W, LI X L, IYER S, et al. Answering complex open-domain questions with multi-hop dense retrieval[J]. arXiv:2009.12756, 2020.
[22] YIN J, JIANG X, LU Z, et al. Neural generative question answering[J]. arXiv:1512.01337, 2015.
[23] ZHONG L, WU J, LI Q, et al. A comprehensive survey on automatic knowledge graph construction[J]. arXiv:2302.05019, 2023.
[24] CHENG J, REDDY S, SARASWAT V, et al. Learning structured natural language representations for semantic parsing[J]. arXiv:1704.08387, 2017.
[25] YE X, YAVUZ S, HASHIMOTO K, et al. RnG-KBQA: generation augmented iterative ranking for knowledge base question answering[J]. arXiv:2109.08678, 2021.
[26] JI S, PAN S, CAMBRIA E, et al. A survey on knowledge graphs: representation, acquisition, and applications[J]. IEEE Transactions on Neural Networks and Learning Systems, 2021, 33(2): 494-514.
[27] AMMAR A B. Query optimization techniques in graph Databases[J]. arXiv:1609.01893, 2016.
[28] LI X, FENG J, MENG Y, et al. A unified MRC framework for named entity recognition[J]. arXiv:1910.11476, 2019.
[29] XUE M G, YU B, LIU T, et al. Porous lattice transformer encoder for Chinese NER[C]//Proceedings of the 28th International Conference on Computational Linguistics, 2020: 3831-3841.
[30] LEE D H, KADAKIA A, TAN K, et al. Good examples make a faster learner: simple demonstration-based learning for low-resource NER[J]. arXiv:2110.08454, 2021.
[31] LIU J, CHEN S, WANG B, et al. Attention as relation: learning supervised multi-head self-attention for relation extraction[C]//Proceedings of the 29th International Conference on International Joint Conferences on Artificial Intelligence, 2021: 3787-3793.
[32] RATHORE V, BADOLA K, SINGLA P. PARE: a simple and strong baseline for monolingual and multilingual distantly supervised relation extraction[C]//Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics, 2022: 340-354.
[33] LI Z, FU L, WANG X, et al. RFBFN: a relation-first blank filling network for joint relational triple extraction[C]//Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics: Student Research Workshop, 2022: 10-20.
[34] BORDES A, USUNIER N, GARCIA-DURAN A, et al. Translating embeddings for modeling multi-relational data[C]//Proceedings of the 26th International Conference on Neural Information Processing Systems, 2013: 2787-2795.
[35] USBECK R, NGOMO A C N, HAARMANN B, et al. 7th open challenge on question answering over linked data (QALD-7)[C]//Proceedings of the Semantic Web Challenges: 4th SemWebEval Challenge at ESWC 2017, 2017: 59-69.
[36] BIKEL D M, MILLER S, SCHWARTZ R, et al. Nymble: a high-performance learning name-finder[C]//Proceedings of the 5th Conference on Applied Natural Language Processing, 1997: 194-201.
[37] MCCALLUM A, LI W. Early results for named entity recognition with conditional random fields, feature induction and web-enhanced lexicons[C]//Proceedings of the 7th Conference on Natural Language Learning at HLT-NAACL 2003, 2003: 188-191.
[38] ZENG Y, YANG H, FENG Y, et al. A convolution BiLSTM neural network model for Chinese event extraction[C]//Proceedings of the International Conference on Computer Processing of Oriental Languages, 2016: 275-287.
[39] HUANG Z, XU W, YU K. Bidirectional LSTM-CRF models for sequence tagging[J]. arXiv:1508.01991, 2015.
[40] DEVLIN J, CHANG M W, LEE K, et al. BERT: pre-training of deep bidirectional transformers for language understanding[J]. arXiv:1810.04805, 2018.
[41] RINK B, HARABAGIU S. UTD: classifying semantic relations by combining lexical and semantic resources[C]//Proceedings of the 5th International Workshop on Semantic Evaluation, 2010: 256-259.
[42] ZENG D, LIU K, LAI S, et al. Relation classification via convolutional deep neural network[C]//Proceedings of the 25th International Conference on Computational Linguistics, 2014: 2335-2344.
[43] ZHANG D, WANG D. Relation classification via recurrent neural network[J]. arXiv:1508.01006, 2015.
[44] SAXENA A, TRIPATHI A, TALUKDAR P. Improving multi-hop question answering over knowledge graphs using knowledge base embeddings[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020: 4498-4507.
[45] YE X, YAVUZ S, HASHIMOTO K, et al. RNG-KBQA: generation augmented iterative ranking for knowledge base question answering[C]//Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics, 2022: 6032-6043.
[46] DIEFENBACH D, SINGH K, MARET P. WDAqua-core0: a question answering component for the research community[M]//Semantic Web Evaluation Challenge. Cham: Springer, 2017: 84-89.
[47] ZOU L, HUANG R, WANG H, et al. Natural language question answering over RDF: a graph data driven approach[C]//Proceedings of the 2014 ACM SIGMOD International Conference on Management of Data, 2014: 313-324.
[48] KAPANIPATHI P, ABDELAZIZ I, RAVISHANKAR S, et al. Question answering over knowledge bases by leveraging semantic parsing and neuro-symbolic reasoning[J]. arXiv:2012.01707, 2020.