MoL-2018-35: Xing, Yujie (2018) Examining Personality Differences in Chit-Chat Sequence to Sequence Conversational Agents. [Report]
Preview |
Text
MoL-2018-35.text.pdf Download (6MB) | Preview |
Abstract
Personality inconsistency is one of the major problems for chit-chat sequence to sequence conversational agents. Works studying this problem have proposed models with the capability of generating personalized responses, but there is not an existing evaluation method for measuring the performance of these models on personality. This thesis develops a new evaluation method based on the psychological study of personality, in particular the Big Five personality traits. With the new evaluation method, the thesis examines if the responses generated by personalized chit-chat sequence to sequence conversational agents are distinguished for speakers with different personalities. The thesis also proposes a new model that generates distinguished responses based on given personalities. The results of our experiments in the thesis show that: for both the existing personalized model and the new model that we propose, the generated responses for speakers with different personalities are significantly more distinguished than a random baseline; specially for our new model, it has the capability of generating distinguished responses for different types of personalities measured by the Big Five personality traits.
| Item Type: | Report |
|---|---|
| Report Nr: | MoL-2018-35 |
| Series Name: | Master of Logic Thesis (MoL) Series |
| Year: | 2018 |
| Subjects: | Logic Philosophy |
| Depositing User: | Dr Marco Vervoort |
| Date Deposited: | 02 Aug 2018 15:19 |
| Last Modified: | 19 Mar 2019 13:06 |
| URI: | https://eprints.illc.uva.nl/id/eprint/1615 |
Actions (login required)
 |
View Item |
