Personality Based Text Generation

Personality Based Text Generation

The rise of chatbots in applications like customer service and companionship highlights the need for personalized experiences. Current limitations often cause user frustration and disengagement. This work proposes a novel approach to enhance chatbot interactions by dynamically adapting text generati...

Full description

Saved in:
Bibliographic Details
Published in:2024 15th International Conference on Computing Communication and Networking Technologies (ICCCNT) pp. 1 - 6
Main Authors: Priyanka, S, Santhiya, S, Keerthika, S, Divyadharshini, B, Madan, P, Yaswin, Ms
Format: Conference Proceeding
Language:English
Published: IEEE 24-06-2024
Subjects:
Adaptation models
Agreeableness
Analytical models
Artificial intelligence
Chatbots
Conscientiousness
CTRL
EleutherAI's GPT-Neo 2.7B
Extroversion
GPT-3
Neural networks
Neuroticism
Openness
Privacy
Prototypes
Training
Transformers
Virtual assistants
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The rise of chatbots in applications like customer service and companionship highlights the need for personalized experiences. Current limitations often cause user frustration and disengagement. This work proposes a novel approach to enhance chatbot interactions by dynamically adapting text generation to the user's personality using the Big Five personality traits (Openness, Conscientiousness, Extroversion, Agreeableness, Neuroticism). The proposed approach involves multiple stages, starting with data acquisition and preprocessing. A large text dataset annotated for the Big Five traits forms the foundation for training a personality analysis model. Neural network models such as GPT-3, CTRL, and GPT-Neo 2.7B are evaluated for personality-based text generation. Among these, GPT-Neo 2.7B shows the best performance, with a BLEU score of 0.65, in capturing personality nuances. Evaluating the system's effectiveness involves human assessment of the generated text to ensure accurate translation of personality indicators into appropriate communication styles. Additionally, user studies will measure satisfaction and engagement during interactions with the chatbot. The primary goal is to develop a functional chatbot prototype that personalizes responses based on the user's personality, leading to improved engagement and satisfaction, thereby significantly advancing the field of human-computer interaction.
ISSN:2473-7674
DOI:10.1109/ICCCNT61001.2024.10724424