LLM Based Embedded Value Trained ChatBot Framework for Personalized Academic Learning

Chandan Satwani, Vrashabh Patil, Kavya Morab, Santosh Pattar, Prema T. Akkasaligar

2025

Abstract

Latest developments in natural language processing and machine learning techniques have enabled the development of chatbots. However, these chatbots are generalized and not tuned for specific requirements of a particular domain. In an academic setting, both learners and teachers today require a specialized chatbot for their personalized teaching learning experience. In this regard, we propose a novel approach using Large Language Model (LLM), fine-tuned with embedded value training. It leverages contextual embeddings and semantic representation to provide tailored educational content. The experimental results demonstrate an improvement of 10%, 20%, and 30% for BERT F1, ROUGE, and BLEU scores respectively, when compared to generic ChatGPT 3.5 and Gemini AI chat applications. These results suggests effectiveness in use of academic specific chatbot in improving student engagement, comprehension and retention though personalized learning experience.

Paper Citation

in Harvard Style

Satwani C., Patil V., Morab K., Pattar S. and Akkasaligar P. (2025). LLM Based Embedded Value Trained ChatBot Framework for Personalized Academic Learning. In Proceedings of the 3rd International Conference on Futuristic Technology - Volume 2: INCOFT; ISBN 978-989-758-763-4, SciTePress, pages 669-675. DOI: 10.5220/0013599800004664

in Bibtex Style

@conference{incoft25,
author={Chandan Satwani and Vrashabh Patil and Kavya Morab and Santosh Pattar and Prema Akkasaligar},
title={LLM Based Embedded Value Trained ChatBot Framework for Personalized Academic Learning},
booktitle={Proceedings of the 3rd International Conference on Futuristic Technology - Volume 2: INCOFT},
year={2025},
pages={669-675},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013599800004664},
isbn={978-989-758-763-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 3rd International Conference on Futuristic Technology - Volume 2: INCOFT
TI - LLM Based Embedded Value Trained ChatBot Framework for Personalized Academic Learning
SN - 978-989-758-763-4
AU - Satwani C.
AU - Patil V.
AU - Morab K.
AU - Pattar S.
AU - Akkasaligar P.
PY - 2025
SP - 669
EP - 675
DO - 10.5220/0013599800004664
PB - SciTePress