alv1nz / comparing-classification-techniques-to-identify-the-best-classifier-in-r-programming

Comparing-Classification-Techniques-to-Identify-the-Best-Classifier

Exploration of classifiers in predicting the subscription of a term deposit exploring areas such as age, job, etc.

Testing 4 classifiers: CTree, J48, Linear Regression, kNN

Classifiers will be measured on these metrics: accuracy, error %, precisio, recall, F-score

Train and split test set

(Format: Accuracy, Error rate, precision, recall, F-Score] Results:

CTree: 87%, 13%, 58%, 50%, 54%

J48: 87%, 13%, 58%, 45%, 51%

Linear Regression: 86%, 14%, 64%, 9%, 15%

kNN: 86%, 14%, 54%, 25%, 34%

CTree Win