University of Twente Student Theses


Adjusting Clarifying Questions to User’s Profile

Neta, Liran (2023) Adjusting Clarifying Questions to User’s Profile.

[img] PDF
Abstract:Literature reports different needs of users with high domain knowledge (experts) or little domain knowledge (novices). User-centered systems accommodate these differences. User-centered product search, particularly clarifying questions (CQs), can improve user performance and satisfaction in conversational product-search systems. Therefore, this research aims to enhance a conversational product search system with user-centered CQs. The research looked to figure out if providing CQs that match the user’s level of knowledge results in a better search experience for the user. Our findings show that users’ perceptions of the system’s effectiveness can be slightly influenced by matching the type of CQs to their level of domain competence. For this research, we decided upon two types of CQs - Advanced and Basic. Advanced CQs include specific measurements, such as “12 to 18 inches”, while Basic CQs include vague sizes, such as “small, medium, big”. Overall, experts preferred the Advanced chatbot, while no significant distinction was found between the Basic and Advanced chatbots among novices. However, further elaboration on unfamiliar terms can be helpful for novices. Recommendation, given by the chatbot, on various capacities (such as RAM size) based on the user’s purpose of usage seem to be valuable as well for the users.
Item Type:Essay (Bachelor)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:54 computer science
Programme:Computer Science BSc (56964)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page