Developing a Chatbot Using Machine Learning: A Case Study for US Stock Market

Abstract

Instead of conventional searches on websites, there is a trend of creating chatbots to ask and answer questions with customers, increasing interaction and customer experience. In this work, we present a chatbot system in the financial sector. Chatbots working with customers handle the website platform, customers can note the development of an exciting, great quality platform leading to more conversions and outstanding leads. This is especially true for businesses that need to attract many customers. Chatbots can provide instant information and answer users questions quickly, helping them access information effectively. Chatbots can create intuitive and enjoyable interactive experiences that help attract and maintain customer interest. This article presents the collaborative design between the Langchain framework and large language models (LLMs) such as ChatGPT, LLaMA....Large language models have demonstrated great potential in natural language processing tasks in the financial sector. Langchain framework is deployed on many projects, helping to significantly reduce time and costs. The chatbot system helps people access information faster. Normally, searching for financial information takes about 7-10 seconds, while the chatbot system extracting information takes about 3-5 seconds. Stockscan.io is a website in this field. In the financial sector, it is also a test case to test the system. The chatbot system can be applied to fields beyond finance and can be further improved.