The UCI repository dataset of forest cover type was used to build a machine learning classification model using linear, logistic, LinearSVC, NearestCentroid, Random Forest and Decision tree classifiers.
- Various model iterations were done like changing cross-validation methods (StratifiedKfold, Kfold), using ensemble method like Bagging, tuning parameters using GridSearchCV, shuffling, changing random seeding and state of splits.
- The iterations were done to evaluate its effect on the classification model accuracy. Random Forest classifier achieved the highest accuracy of 88% among all the machine learning models.