Sunspot Numbers Prediction

Project Overview

This project aims to predict the number of sunspots using historical data (1818–2019). Several machine learning models were implemented, including Linear Regression, Ridge Regression, Decision Tree, and Random Forest, to achieve the best prediction accuracy. The Random Forest model provided the best performance with a Mean Absolute Error (MAE) of approximately 1.6.

---

Features

• Data Preprocessing:
  • Cleaned and transformed the dataset by handling missing values and dropping irrelevant columns.
  • Scaled features using StandardScaler for better model performance.
• Exploratory Data Analysis (EDA):
  • Analyzed correlations between features and the target variable.
  • Visualized sunspot trends using Matplotlib and Seaborn.
• Model Training:
  • Implemented Linear Regression, Ridge Regression, Decision Tree, and Random Forest models.
  • Tested polynomial features but found them ineffective.
• Hyperparameter Tuning:
  • Used GridSearchCV to optimize hyperparameters for Ridge and Random Forest models.
• Evaluation:
  • Achieved the best MAE of ~1.6 with the Random Forest model.
  • Validated the model on a test set for final performance metrics.

---

Tools and Libraries

• Programming Language: Python
• Libraries:
  • Data Processing: Pandas, NumPy
  • Visualization: Matplotlib, Seaborn
  • Machine Learning: Scikit-learn