Coherence disruption in human-chatbot communication

• Název práce v češtině: Narušení koherence v interakci člověka s chatbotem
• Název v anglickém jazyce: Coherence disruption in human-chatbot communication
• Klíčová slova: konverzace|koreference|narušení koherence|anafora|chatbot|umělá inteligence
• Klíčová slova anglicky: conversation|coreference|coherence disruption|anaphora|chatbot|artificial intelligence
• Akademický rok vypsání: 2023/2024
• Typ práce: diplomová práce
• Jazyk práce: angličtina
• Ústav: Ústav českého jazyka a teorie komunikace (21-UCJTK)
• Vedoucí / školitel: Mgr. Anna Nedoluzhko, Ph.D.
• Řešitel: skrytý - zadáno a potvrzeno stud. odd.
• Datum přihlášení: 05.03.2024
• Datum zadání: 12.03.2024
• Schválení administrátorem: bylo schváleno
• Datum potvrzení stud. oddělením: 15.05.2024
• Datum a čas obhajoby: 16.06.2025 09:00
• Datum odevzdání elektronické podoby: 07.05.2025
• Datum proběhlé obhajoby: 16.06.2025
• Odevzdaná/finalizovaná: odevzdaná studentem a finalizovaná
• Oponenti: doc. Mgr. Radek Šimík, Ph.D.

Zásady pro vypracování

The aim of the study is to utilize a chatbot to research the disruption of conversation coherence. A chatbot can serve this purpose because 1) it allows indirect influence on the conversation and 2) the reactions of speakers to coherence disruption when communicating with a machine are more easily exposed due to the absence of intersubjectivity (Schegloff 1992). The main task is to obtain data from transcripts of conversations between humans and the chatbot, which contain various types of such disruption. To repeatedly initiate coherence disruption of a similar nature, the direction of the conversation is partially controlled by conversational artificial intelligence systems (Hudeček, Dušek, 2023). The observed types of disruption stem from interconnected areas of unsatisfactory or unassignable anaphora and abrupt, unexpected topic shifts. The obtained data undergo qualitative analysis, which, using the framework of centering theory (Poesio et al 2004), allows for a closer description of various types of coherence disruption and reactions. This is followed by quantitative analysis to verify whether there is a correlation between specific types of coherence disruption and specific types of reactions to them in relation to the severity of their impact on overall conversation coherence.

Seznam odborné literatury

Givón, T. (2020). Coherence. John Benjamins. https://doi.org/10.1075/z.230

Hirst, G., McRoy, S., Heeman, P., Edmonds, P., & Horton, D. (1994). Repairing conversational misunderstandings and non-understandings. Speech Communication, 15(3-4), 213-229. https://doi.org/10.1016/0167-6393(94)90073-6

Hudeček, V., & Dušek, O. (2023). Are Large Language Models All You Need for Task-Oriented Dialogue? In S. Stoyanchev, S. Joty, D. Schlangen, O. Dušek, C. Kennington, & M. Alikhani (Eds.), Proceedings of the 24th Annual Meeting of the Special Interest Group on Discourse and Dialogue (pp. 216-228). Prague, Czechia: Association for Computational Linguistics. https://doi.org/10.18653/v1/2023.sigdial-1.21

Jurafsky, D., & Martin, J. H. (2023). Speech and Language Processing. Draft of January 7, 2023. Chapter 27: Discourse Coherence.

Nedoluzhko, Anna. Rozšířená textová koreference a asociační anafora (koncepce anotace českých dat v pražském závislostním korpusu). Dizertační práce, vedoucí Uličný, Oldřich. Univerzita Karlova, Filozofická fakulta, Ústav českého jazyka a teorie komunikace, 2010.

Poesio, M., Stevenson, R., Di Eugenio, B., & Hitzeman, J. (2004). Centering: A Parametric Theory and Its Instantiations. Computational Linguistics, 30(3), 309-363. https://doi.org/10.1162/0891201041850911

Schegloff, E. A. (1992). Repair After Next Turn: The Last Structurally Provided Defense of Intersubjectivity in Conversation. American Journal of Sociology, 97(5), 1295–1345. http://www.jstor.org/stable/2781417

Uryupina, O., Kabadjov, M., Poesio, M. (2016). Detecting Non-reference and Non-anaphoricity. In: Poesio, M., Stuckardt, R., Versley, Y. (Eds.) Anaphora Resolution. Theory and Applications of Natural Language Processing. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-662-47909-4_13