Cryptocurrency Price Analytics and Prediction Using a Layered Data Pipeline and Machine Learning

This work presents an end-to-end cryptocurrency price analytics and prediction pipeline implemented in Python. The system follows a Bronze–Silver–Gold data architecture and integrates real-world API-based data ingestion, time-series preprocessing, exploratory data analysis, feature engineering, and machine learning modeling. Linear Regression and Random Forest models are evaluated using standard regression metrics to assess predictive performance on cryptocurrency price data. The project emphasizes reproducibility, ethical use of publicly available data, and practical data engineering and machine learning workflows. Source code and experimental artifacts are openly available to support transparency and reuse.