Building a Hindi-English customer support chatbot using pre-trained language models

Khatib, Mehwish Mohammed Hanif (2025) Building a Hindi-English customer support chatbot using pre-trained language models. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (548kB) | Preview

Abstract

Multilingual support is a significant challenge in linguistically diverse countries like India, where users may switch languages during conversations. This project presents a Hindi-English customer support chatbot designed to overcome language barriers using modern NLP, deep learning, and transformer-based models. Leveraging over 1,002,235 bilingual query-response pairs, the system employs BiLSTM for intent classification, BERT for response generation, and mBERT for multilingual multitask learning, optimized with TensorFlow and Keras. The chatbot, built with Tkinter, offers a user-friendly GUI allowing language choice and delivers high accuracy and fast responses, effectively handling code-switched inputs. Its performance metrics indicate readiness for real-world customer service deployment in Indian markets. Future enhancements include adding regional dialects, voice interfaces, and domain-specific fine-tuning, further improving bilingual user experience and contextual response accuracy.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Simiscuka, Anderson UNSPECIFIED
Subjects:	P Language and Literature > P Philology. Linguistics Q Science > QH Natural history > QH301 Biology > Methods of research. Technique. Experimental biology > Data processing. Bioinformatics > Artificial intelligence Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Artificial intelligence P Language and Literature > P Philology. Linguistics > Computational linguistics. Natural language processing
Divisions:	School of Computing > Master of Science in Artificial Intelligence
Depositing User:	Ciara O'Brien
Date Deposited:	28 May 2026 14:39
Last Modified:	28 May 2026 14:39
URI:	https://norma.ncirl.ie/id/eprint/9327

Actions (login required)

View Item