Predicting Cryptocurrency Prices: A Machine Learning Approach for Investment Strategies and Market Trend Analysis

Cryptocurrencies have emerged as significant players in the financial market, attracting substantial investments and interest. However, the market's high volatility and unpredictability pose challenges for investors in forecasting price movements and achieving profitable investments. This study aims to develop a model that predicts future cryptocurrency prices, providing valuable insights for investment strategies and market trend analysis. The proposed model utilizes advanced methods such as feature selection and a combination of Random Forest Regression and Linear Regression techniques to analyze historical price data and generate real-time forecasts. The study employs the CoinGecko API to retrieve cryptocurrency data and presents users with a Streamlit web application for interacting with the prediction system. Users can log in, select a cryptocurrency, and obtain future price predictions based on the chosen machine learning algorithm. The architecture of the program involves fetching data from the API, applying machine learning algorithms, and displaying the results through a user interface. By comparing the accuracy of Linear Regression and Random Forest Regressor, the study aims to identify the most effective approach for cryptocurrency price prediction. The proposed model seeks to provide a robust framework for analyzing data and generating actionable insights in the dynamic cryptocurrency market, empowering investors to make informed decisions and actively participate in the financial discovery journey.