Abstract: This project proposes the use of stacked LSTM (Long Short-Term Memory) networks for developing an app that can provide users with accurate and reliable predictions of future stock prices based on historical data. Stacked LSTMs are a type of artificial neural network that are designed to analyse sequential data, such as time series data, and are well-suited to stock price prediction because they can effectively analyse the relationships between past and present price movements over long periods of time. The app will be developed using Python programming language and will utilize various libraries such as TensorFlow, Keras, and Pandas for data analysis and visualization. The user interface will be designed to be user-friendly, allowing users to easily access and view the stock data and predictions. In addition to the predictive model itself, this project aims to demonstrate the effectiveness of stacked LSTM networks for predicting stock prices. The app will be trained on historical stock price data from various sources, including Yahoo Finance. The results of this project may have significant implications for investors who rely on accurate stock price predictions for making informed investment decisions. By demonstrating the effectiveness of stacked LSTM networks for predicting stock prices, this project may help investors make more informed decisions about which stocks to buy or sell. Overall, this project aims to contribute to the growing body of research on machine learning algorithms for stock price prediction and provide users with an easy-to-use app for making informed investment decisions based on accurate predictions.

Keywords: LSTM (Long Short-Term Memory), Time Series Data, Min-Max Scaler, Sequential Data

| DOI: 10.17148/IARJSET.2023.105100