Rock-Classification-Using-Machine-Learning

This project involves applying various classification techniques to predict rock categories based on 11 rock features. The task is divided into several steps, starting with data preprocessing, where features and labels are extracted from two provided files. The dataset is split into training, validation, and testing subsets based on specified token numbers. The analysis begins with descriptive statistics and visualizations of the features, followed by exploring correlations between attributes and labels. Various machine learning models, including Multinomial Logistic Regression, Support Vector Machine, and Random Forest, are trained and fine-tuned by adjusting hyperparameters, and their performance is evaluated using accuracy, precision, recall, and F1 score. The models are then combined into an ensemble to improve performance. Lastly, human classification accuracy is compared with the best-performing model, and the relationship between human and model predictions is analyzed through scatter plots and correlation tests.

Accuracy is quite low. Room for development.