Masterarbeit
User Experience in Chatbots: A Comparative Analysis of Chatbot Interaction Pattern Implementations
Completion
2025/06
Research Area
Students
Akshata Anil Mithari
Advisers
Verena Traubinger M.Sc.
Dr.-Ing. Sebastian Heil
Description
The advent of chatbots has led to their increased integration in diverse sectors such as e-commerce, customer service, and healthcare. While the research on natural language-based interactions in chatbots is well-developed, the exploration of graphical interface interaction is less advanced. Existing studies have identified chatbot interaction patterns, but a comparison of their effect on the user experience is still missing. This research aims to bridge this gap by analysing these patterns and their implementations, identifying new ones, and assessing their impact on user satisfaction, engagement, and efficiency. The goal is to provide insights to improve the chatbot interface design.
Initially, the thesis will conduct a comprehensive analysis of various e-commerce chatbots, using existing studies on interaction patterns as a basis, to understand common user interactions in areas such as customer support, product recommendations, and purchasing. It will then identify and select key chatbot interaction patterns and their implementation variations. Following this, the research will focus on implementing multiple chatbot versions, each incorporating different pattern variations in a simulated e-commerce environment. The final phase involves conducting an evaluation of their effect on user experience by assessing metrics like user satisfaction, efficiency, and engagement. This data can then help determine which chatbot design elements are most effective in enhancing specific user interaction.
The objective of this thesis is the creation of a solution or the combination of existing approaches to solve the problem of how different implementations of chatbot interaction patterns influence the user experience as described above. This comprises the following parts. An analysis of the state of the art of already known chatbot interaction patterns, chatbot user interfaces, current technologies for chatbot implementations, and other relevant work. A demonstration of the solution should present prototypical chatbots which use different implementation of chatbot interaction patterns. A suitable evaluation based both on the conceptualized experiment settings above, and the compliance with elicited requirements based on the literature research should provide insights into the effect of different pattern implementations on the user experience.
