Implementing A Chatbot to Analyse Live Stocks Updates Using FLASK and NLU

Abstract

The future of stock market is a complicated activity because of the instability of the market, external economic variables, and price changes. This study introduces a smart stock forecast and advisory application that combines trend-based analytical methods with an experimental deep-learning platform to enable making a well-informed investment choice. The suggested system operates on a short-term and monthly trend analysis, which operates on moving averages, momentum indicators and weighted scoring algorithms to simulate stock movement as either rise, fall or neutral with confidence estimation espoused based on real-time and historical stock price data. Moreover, a Long Short-Term Memory (LSTM) architecture is constructed and trained on time-series price movements to examine the deep learning-based movement forecasts. Two main applications have been utilized in the implementation of the system: web-based application with the Flask framework that allows user authentication, watchlist management, prediction storage, and automated email notifications. A conversational assistant AI also increases the user engagement by offering market information and guide-based responses to queries. The suggested solution attests to the possible efficacy of using the analytical forecasting methods along with smart user-friendly elements to study the stock market in practice.