/AdultSalaryDataUS

Beginner ML project on the UCI Adult Dataset

Primary LanguageJupyter NotebookGNU General Public License v3.0GPL-3.0

AdultSalaryDataUS

Beginner ML project on the UCI Adult Dataset here

A total of 4 classifiers were used, Namely -

  1. Random Forest Classifer.
  2. AdaBoost Classifier.
  3. KNN Classifier.
  4. SVM Classifier.

The details is given in the jupyter notebook of each classifier in the Classifiers Folder.

The result and peroformance on the training set:

  1. Random Forest gave the best accuracy and performed well on the negative label, but couldn't acheive a good recall or f1-score.
  2. AdaBoost had an average accuracy but the best ROC AUC score and the best recall and f1-score.
  3. KNN performed worse than Random Forst or AdaBoost, but better than SVM.
  4. SVM classifier was not able to do a good job, maye because of Randomised Search Cross Validation