NLP AND ML BASED CHATBOT FOR RAILWAYS

This project successfully designed, implemented, and tested a robust, intelligent Chatbot system for the Indian Railways leveraging modern NLP and ML technologies. The core objective of improving passenger service efficiency and quality was achieved, as evidenced by the high self-service rate and rapid response times. By moving away from rigid, rule-based systems, the proposed solution demonstrated superior ability to understand and process natural human language, offering a scalable and personalized conversational experience. The system's modular architecture, encompassing a state-of-the-art ML pipeline for intent and entity recognition, proved highly reliable in the production environment

Bender, E. M., & Koller, A. (2017). Climbing Mount Pnini: Characterizing the production of referential descriptions in generative models. Proceedings of the 15th European Workshop on Natural Language Generation. F. Chollet (2017). Python deep learning. A Manning Publication.  J.Devlin,M.W.Chang,K. Lee, & K. Toutanova (2019). BERT: Deep Bidirectional Transformer Language Understanding Pre-training. NAACL 2019 Conference Proceedings. Gunning, D. (2019). XAI: Explainable AI. Defense Advanced Research Projects Agency.