Hybrid Chatbot for Customer Service Using Transfer Learning Algorithm on Embedding and Generative Models within a Retrieval-Augmented Generation (RAG) at PT Telekomunikasi Indonesia Tribe Agree

DOI

10.2991/978-94-6463-926-1_60How to use a DOI?

Keywords

Hybrid Chatbot; Generative Model; Embedding Model; Retrieval Augmented Generation; Algorithm Transfer Learning; Customer Service

Abstract

Conventional customer service concepts that rely on communication via email, WhatsApp, and mobile phones such as call centers have limitations in scalability, especially when the number of customers continues to grow without being matched by an increase in human resources. This is also experienced by Agree, a digital agribusiness platform, which currently serves more than 71,000 cultivators and 250 agribusiness companies with only three customer service personnel. Meanwhile, the use of chatbots as a service automation solution has limitations in answering questions that are beyond the scope of their training data or knowledge sources. To address these challenges, this research proposes a hybrid approach that combines conventional and modern methods. The conventional approach retains the call center function to handle complex cases that are outside of the training data and knowledge sources, while the modern method applies a generative language model-based chatbot supported by vector knowledge base technology. The generative chatbot in this study was built using the Retrieval-Augmented Generation (RAG) approach. The transfer learning algorithm is applied through a fine-tuning process on two main components, namely the embedding model used to transform documents into vector representations, and the generative model that is retrained to be able to generate more relevant and aligned responses with the knowledge source.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Muhammad Ilham Adi Pratama
AU  - Nana Ramadijanti
AU  - Nur Rosyid Mubtadai
PY  - 2025
DA  - 2025/12/31
TI  - Hybrid Chatbot for Customer Service Using Transfer Learning Algorithm on Embedding and Generative Models within a Retrieval-Augmented Generation (RAG) at PT Telekomunikasi Indonesia Tribe Agree
BT  - Proceedings of the International Conference on Applied Science and Technology on Engineering Science 2025 (iCAST-ES 2025)
PB  - Atlantis Press
SP  - 533
EP  - 550
SN  - 2352-5401
UR  - https://doi.org/10.2991/978-94-6463-926-1_60
DO  - 10.2991/978-94-6463-926-1_60
ID  - Pratama2025
ER  -