Protein-Sequence-Classification

Protein sequence multiclassification Machine Learning Project

• Dataset: https://www.kaggle.com/shahir/protein-data-set
• Kaggle Notebook: https://www.kaggle.com/ammarhelali/protein-sequence-eda-multiclassification

Accuracy Model

• 48% GaussianNB
• 91% KNN
• 91.7% KNN with SMOTE
• 94% RandomForest (60 estimators)
• 98.5% RandomForest (225 estimators)
• 98.4% RandomForest (225) with SMOTE
• 98.1% RandomForest (150) with Feature Selection
• 75.6% XGBoost with SMOTE(300 estimator)
• 81.2% XGBoost with SMOTE(600 estimator)