The Chatbot Repair Tango

Placeholder Show Content

Abstract/Contents

Abstract
Miscommunication occurs often in conversation. Humans employ various repair strategies to resolve miscommunications that occur in human-to-human speech, in what is called “conversational repair”. However, most chatbots currently in use either do not engage in repair at all, or conduct repair in ways that are, to a human speaker, clumsy and unnatural. This unnatural repair may lead to negative user experiences. This thesis draws on research into users’ experience of bots and conversational repair to explore the types of miscommunication, and possible repair strategies, that may occur between a task-oriented survey chatbot and a user. I find that errors in misunderstanding and repetition are common for medical survey chatbots, but that the first difficult key step for the chatbot is to notice that an error has occurred. From there, I identify repair strategies that are most ‘natural’ in conversation, including previously researched techniques such as “Confirmation”, along with new techniques such as “Simplification” and “Sense-Making”, in the hopes of building chatbots that will be able to have more positive interactions with users. This naturalistic repair, where a patient and the chatbot are able to continue a conversation naturally and fluidly in a way that is metaphorically similar to a tango, is key to improving user experience with chatbots.

Description

Type of resource text
Date created June 2021

Creators/Contributors

Author Yang, Anna Christy
Degree granting institution Stanford University, Department of Symbolic Systems
Primary advisor Bernstein, Michael
Advisor Shrager, Jeff

Subjects

Subject Artificial Intelligence
Subject Chatbot
Subject Conversational Agent
Subject Conversational Repair
Subject Human Computer Interaction
Subject Intelligent Assistant
Subject Natural Language Interfaces
Subject Pragmatics
Subject Symbolic Systems
Subject SymSys
Genre Thesis

Bibliographic information

Access conditions

Use and reproduction
User agrees that, where applicable, content will not be used to identify or to otherwise infringe the privacy or confidentiality rights of individuals. Content distributed via the Stanford Digital Repository may be subject to additional license and use restrictions applied by the depositor.
License
This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).

Preferred citation

Preferred Citation
Yang, Anna Christy. (2021). The Chatbot Repair Tango. Stanford Digital Repository. Available at: https://purl.stanford.edu/nv755gw7179

Collection

Master's Theses, Symbolic Systems Program, Stanford University

View other items in this collection in SearchWorks

Contact information

Also listed in

Loading usage metrics...