Currently, an increasing number of governments have adopted question answering systems (QASs) in public service delivery. As some citizens with limited information literacy often express their questions vaguely when interacting with a chatbot, it is necessary to improve the contextual understanding and reasoning ability of government chatbots (G-chatbots). This goal can be achieved through the optimisation of the matching between question, answer and context. By incorporating the Relational Graph Convolutional Networks (R-GCNs) and fuzzy logic, this study proposes a multi-turn dialogue model that introduces a re-question mechanism and a subgraph matching algorithm. The experiment results show that the model can improve the contextual reasoning ability of G-chatbots by about 10% and generate answers in a more explainable way. This study innovatively integrates a question–answer–context matching approach, re-question mechanism into the MTRF-G-chatbot model, reducing barriers to citizens’ access to government services and enhancing contextual reasoning abilities.
AokiN. An experimental study of public trust in AI chatbots in the public sector. Gov Inform Q2020; 37(4): 101490.
2.
KimNHongS. Automatic classification of citizen requests for transportation using deep learning: case study from Boston city. Inform Process Manag2021; 58(1): 102410.
3.
United Nations Department of Economic and Social Affairs. The future of digital government. United Nat E-gov Surv2022; 6(10): 311.
4.
RoehlUBU. Automated decision-making and good administration: views from inside the government machinery. Gov Inf Q2023; 40(4): 101864.
5.
HerdPDeLeireTHarveyH, et al. Shifting administrative burden to the state: a case study of medicaid take-up. Public Admin Rev2013; 73: 69–81.
6.
MoynihanDPHerdPRibgyE. Policymaking by other means: do states use administrative barriers to limit access to Medicaid?Admin Soc2016; 48(4): 497–524.
BertotJEstevezEJanowskiT. Universal and contextualized public services: digital public service innovation framework. Gov Inform Q2016; 33(2): 211–222.
9.
YangZXuWChenR. A deep learning-based multi-turn conversation modeling for diagnostic Q&A document recommendation. Inform Process Manag2021; 58(3): 102485.
10.
YuanCZhouWLiM, et al. Multi-hop selector network for multi-turn response selection in retrieval-based chatbots. In: Proceedings of the 2019 conference on empirical methods in natural language processing and the 9th international joint conference on natural language processing, Hong Kong, China, 3–7 November 2019, pp. 111–120. East Stroudsburg, PA: Association for Computational Linguistics.
ZhangDYangZMengF, et al. TSAM: a two-stream attention model for causal emotion entailment. In: Proceedings of the 29th International Conference on computational linguistics, Gyeongju, Republic of Korea, 12–17 October 2022, pp. 6762–6772. Gyeongju, Republic of Korea: International Committee on Computational Linguistics.
13.
DingMZhouCChenQ, et al. Cognitive graph for multi-hop reading comprehension at scale. In: Proceedings of the 57th annual meeting of the association for computational linguistics, Florence, 28 July–2 August 2019, pp. 2694–2703. East Stroudsburg, PA: Association for Computational Linguistics.
14.
ZhangYSunSGalleyM, et al. DIALOGPT: large-scale generative pre-training for conversational response generation. In: Proceedings of the 58th annual meeting of the association for computational linguistics: system demonstrations, Online, 5–10 July 2020, pp. 270–278. East Stroudsburg, PA: Association for Computational Linguistics.
15.
PengWHuYXingL, et al. Control globally, understand locally: a global-to-local hierarchical graph network for emotional support conversation, https://arxiv.org/abs/2204.12749 (2022, accessed 03 June 2023).
16.
QinLXieTCheW, et al. A survey on spoken language understanding: recent advances and new frontiers, https://arxiv.org/abs/2103.03095 (2021, accessed 03 June 2023).
17.
ShangLLuZLiH. Neural responding machine for short-text conversation. In: Proceedings of the 53rd annual meeting of the association for computational linguistics and the 7th international joint conference on natural language processing, Beijing, China, 26–31 July 2015, pp. 1577–1586. East Stroudsburg, PA: Association for Computational Linguistics.
18.
Martinez-GilJ. A survey on legal question–answering systems. Computer Science Review2023; 48: 100552.
19.
SchwarzerMDüverJPlochD, et al. An interactive e-government question answering system. In: Lernen, wissen, daten, analysen, Potsdam, Germany, 12–14 September 2016, pp. 74–82.
20.
MilianiMPassaroLCLenciA. FRAQUE: a FRAme-based QUEstion-answering system for the Public Administration domain. In: Proceedings of the 1st workshop on language technologies for government and public administration (LT4Gov), Marseille, 11–16 May 2020, pp. 7–14. Paris: European Language Resources Association.
21.
AndroutsopoulouAKaracapilidisNLoukisE, et al. Transforming the communication between citizens and government through AI-guided chatbots. Gov Inform Q2019; 36(2): 358–367.
22.
ChenTGuoWGaoX, et al. AI-based self-service technology in public service delivery: user experience and influencing factors. Gov Inform Q2021; 38(4): 101520.
23.
JuJMengQSunF, et al. Citizen preferences and government chatbot social characteristics: evidence from a discrete choice experiment. Gov Inform Q2023; 40(3): 101785.
24.
MaZDuBShenJ, et al. An encoding mechanism for seq2seq based multi-turn sentimental dialogue generation model. Procedia Comput Sci2020; 174: 412–418.
25.
CambriaELiuQDecherchiS, et al. SenticNet 7: a commonsense-based neurosymbolic AI framework for explainable sentiment analysis. In: Proceedings of the thirteenth language resources and evaluation conference, Marseille, 20–25 June 2022, pp. 3829–3839. Paris: European Language Resources Association.
26.
BergnerASHildebrandCHäublG. Machine talk: how verbal embodiment in conversational AI shapes consumer–brand relationships. J Consum Res2023; 50(4): 742–764.
27.
YaoRHouLYangL, et al. Deep human answer understanding for natural reverse QA. Knowl-based Syst2022; 254: 109625.
28.
ZhouXWuOJiangC. Increasing naturalness of human–machine dialogue: the users’ choices inference of options in machine-raised questions. Knowl-based Syst2022; 243: 108485.
29.
YihSW-tChangM-WHeX, et al. Semantic parsing via staged query graph generation: question answering with knowledge base. In: Proceedings of the Joint Conference of the 53rd Annual Meeting of the ACL and the 7th international joint conference on natural language processing of the AFNLP, Beijing, China, 26–31 July 2015, East Stroudsburg, PA: Association for Computational Linguistics.
30.
ZhaoLXuWGaoS, et al. Utilizing graph neural networks to improving dialogue-based relation extraction. Neurocomputing2021; 456: 299–311.
31.
ZhouHYoungTHuangM, et al. Commonsense knowledge aware conversation generation with graph attention. In: IJCAI’18: Proceedings of the 27th international joint conference on artificial intelligence, Stockholm, 13-19 July 2018, pp. 4623–4629. Menlo Park, CA: AAAI Press.
32.
WuSTangYZhuY, et al. Session-based recommendation with graph neural networks. Proc AAAI Conf Artif Intell2019; 33(1): 346–353.
33.
SchlichtkrullMKipfTNBloemP, et al. Modeling relational data with graph convolutional networks. In: The semantic web, (eds GangemiANavigliRVidalM-E, et al.), Cham, 03 June 2018, pp. 593–607. Heidelberg: Springer International Publishing.
34.
WeiJWangXSchuurmansD, et al. Chain-of-thought prompting elicits reasoning in large language models. In: 36th conference on neural information processing systems (NeurIPS 2022), New Orleans, LA, 28 January 2022, pp. 24824–24837. Red Hook, NY: NeurIPS Foundation.
BestaMBlachNKubicekA, et al. Graph of thoughts: solving elaborate problems with large language models, https://arxiv.org/abs/2308.09687 (2023, accessed 26 July 2024).
37.
ZhaoXLiMLuW, et al. Enhancing zero-shot chain-of-thought reasoning in large language models through logic. arXiv:230913339, 2023.
38.
MorrisJXSinghCRushAM, et al. Tree prompting: efficient task adaptation without fine-tuning, https://arxiv.org/abs/2310.14034 (2023, accessed 26 July 2024).
39.
RuanJChenYZhangB, et al. TPTU: task planning and tool usage of large language model-based ai agents. arXiv:230803427, 2023.
40.
DziriNLuXSclarM, et al. Faith and fate: limits of transformers on compositionality, https://arxiv.org/abs/2305.18654 (2023, accessed 26 July 2024).
41.
DhuliawalaSKomeiliMXuJ, et al. Chain-of-verification reduces hallucination in large language models, https://arxiv.org/abs/2309.11495 (2023, accessed 26 July 2024).
42.
WebbTHolyoakKJLuH. Emergent analogical reasoning in large language models. Nat Hum Behav2023; 7(9): 1526–1541.
43.
ZhangYYangJYuanY, et al. Cumulative reasoning with large language models, https://arxiv.org/abs/2308.04371 (2023, accessed 26 July 2024).
44.
BerantJChouAFrostigR, et al. Semantic parsing on freebase from question-answer pairs. In: Proceedings of the 2013 conference on empirical methods in natural language processing, Seattle, WA, 18–21 October 2013, pp. 1533–1544. East Stroudsburg, PA: Association for Computational Linguistics.
45.
KwiatkowskiTZettlemoyerLGoldwaterS, et al. Lexical generalization in CCG grammar induction for semantic parsing. In: Proceedings of the 2011 conference on empirical methods in natural language processing, Edinburgh, 27–31 July 2011, pp. 1512–1523. East Stroudsburg, PA: Association for Computational Linguistics.
RenHHuWLeskovecJ. Query2box: reasoning over knowledge graphs in vector space using box embeddings, https://arxiv.org/abs/2002.05969 (2020, accessed 03 June 2023).
48.
BorgwardtSPeñalozaR. Consistency reasoning in lattice-based fuzzy description logics. Int J Approximate Reasoning2014; 55(9): 1917–1938.
49.
BorgwardtSPeñalozaR. Reasoning in fuzzy description logics using automata. Fuzzy Sets Syst2016; 298: 22–43.
50.
CantadorIViejo-TardíoJCortés-CedielME, et al. A chatbot for searching and exploring open data: implementation and evaluation in e-government. In: DGO2021: The 22Nd annual international conference on digital government research, Omaha, NE, 9–11 June 2021, pp. 168–179. New York: Association for Computing Machinery.
51.
PapineniKRoukosSWardT, et al. Bleu: a method for automatic evaluation of machine translation. In: Proceedings of the 40th annual meeting of the association for computational linguistics, Philadelphia, PA, 6–12 July 2002, pp. 311–318. East Stroudsburg, PA: Association for Computational Linguistics.
52.
LiJGalleyMBrockettC, et al. A diversity-promoting objective function for neural conversation models. In: Proceedings of the 2016 Conference of the North American chapter of the association for computational linguistics: human language technologies, San Diego, CA, 12–17 June 2016, pp. 110–119. East Stroudsburg, PA: Association for Computational Linguistics.
53.
LinC-Y. Rouge: a package for automatic evaluation of summaries. In: Text Summarization Branches Out, Barcelona, 25-26 July 2004, pp. 74–81. East Stroudsburg, PA: Association for Computational Linguistics.
54.
HuSZouLYuJX, et al. Answering natural language questions by subgraph matchingover knowledge graphs. IEEE Trans Knowl Data Eng2017; 30(5): 824–837.
55.
WangF. Social capital or non-human sources? A cross-context study on information source selection of migrant farmer workers. J Inf Sci2023; 49(5): 1358–1374.
