Stock Trend Prediction using LSTM and Streamlit

Stock Trend Prediction using LSTM and Streamlit

Amitesh Srinivas
Department of Computer Science and Engineering
RV College of Engineering
Bengaluru, India
amiteshs.cs22@rvce.edu.in

Anand Patil
Department of Computer Science and Engineering (Data Science)
RV College of Engineering
Bengaluru, India
anandpatil.cd22@rvce.edu.in

Aditya G S
Department of Computer Science and Engineering
RV College of Engineering
Bengaluru, India
adityags.cs22@rvce.edu.in

Tarun HS
Department of Computer Science and Engineering (Data Science)
RV College of Engineering
Bengaluru, India
tarunhs.cd22@rvce.edu.in

Bhaskar M G
Department of Industrial Engineering and Management
RV College of Engineering
Bengaluru, India
bhaskarmg@rvce.edu.in

Abstract—In recent years, the field of stock trend prediction has witnessed a surge in the application of deep learning techniques, particularly Long Short-Term Memory (LSTM) networks, owing to their superior ability to handle time series data. This paper presents a web-based application developed using Python and Streamlit that predicts stock trends using an LSTM model trained on historical stock prices. Data is fetched using the yfinance API and preprocessed using pandas and NumPy. The model is developed with Keras and TensorFlow, and includes visualizations such as moving averages and actual vs. predicted price trends. This system aims to provide retail investors with accessible, interpretable, and real-time predictions to support informed investment decisions.

Keywords—Stock prediction, LSTM, Streamlit, deep learning, financial forecasting, time series analysis