Smart Banking Chatbot

Abstract: This paper provides the development and implementation of a smart banking chatbot designed to enhance the user experience by providing automated responses to customer inquiries. With the increasing digitalization of banking services, the need for quick and efficient customer support has become crucial. The chatbot, developed using the RASA framework, leverages Natural Language Processing (NLP) to understand user intents and execute predefined actions. It offers services such as checking account balances, making transactions, and addressing customer queries in real-time, reducing the dependency on human customer support. By evaluating different machine learning algorithms, including Random forest, (KNN) k-Nearest neighbors, and Support Vector Machines(SVM), the chatbot ensures high accuracy in predicting user requests. RASA is an open-source (NLP) Natural Language Processing platform used to create virtual assistants and contextual chatbots. RASA has several main components: RASA Open Source and RASA Action Server. Rasa Open Source has Rasa NLU which is used to recognize the intent of communication and Rasa core to determine what to do next and how to continue the conversation. Rasa Action Server allows you to host python scripts to do certain custom tasks, like processing or modifying databases in the backend.The experimental results demonstrate that the chatbot significantly reduces response time and improves service efficiency, making it a valuable addition to modern banking systems