Artificial friends: A ChatGPT implementation in AlphaMini companionship robots

Spinei, Alexia (2024) Artificial friends: A ChatGPT implementation in AlphaMini companionship robots. Bachelor's Thesis, Artificial Intelligence.

Preview	Text Thesis.pdf Download (1MB) | Preview
	Text toestemming.pdf Restricted to Registered users only Download (130kB)

Abstract

Addressing the rise in loneliness, especially among emerging and older adults, this study investigates the potential of social robots as companions by enhancing their conversational abilities with ChatGPT-3.5, complemented by sentiment analysis. Utilizing the AlphaMini robot, equipped with Whisper for speech transcription and Amazon Comprehend for sentiment analysis, the research aims to improve the naturalness and empathy of robot-human interactions. Through an experiment involving 22 participants, the study compares the effectiveness of standard ChatGPT responses versus those augmented by Amazon Comprehend. The results indicate no significant difference in participants' perception of anthropomorphism, likeability, or general and emotional intelligence of the robot between groups. This suggests that ChatGPT-3.5 may already offer a level of empathetic engagement sufficient for the context of social robotics or that there is a potential need for advancements in complementary areas, such as integrating contextually relevant gestures, to enhance the overall interaction quality and user experience.

Item Type:	Thesis (Bachelor's Thesis)
Supervisor name:	Vogt, P.A.
Degree programme:	Artificial Intelligence
Thesis type:	Bachelor's Thesis
Language:	English
Date Deposited:	23 May 2024 14:30
Last Modified:	23 May 2024 14:30
URI:	https://fse.studenttheses.ub.rug.nl/id/eprint/32429

Actions (login required)

View Item