KDDM Final Project

Datasets

Dataset URL

-The dataset involves financial and personal attributes of customers over several months.

-The aim is to classify the person’s credit score into good, bad or standard category.

-The data consists of person’s Annual Income, Number of bank accounts, Number of Credit cards, Outstanding Debt, Payment Behaviour, etc.

Methodology utilized for Data Cleaning

• Mode & Mean to replace the incorrect or flase and missing values in the Data.

• Outlier Removal.

• Removal of NaN Values after the above to method.

Models Used

Model	Accuracy
KNN	0.91
Naive Bayes	0.71
Logistice Regression	0.68
Random Forest	0.99
Decision Tree	0.96
K-Means Clusterin	0.6067
SVM	0.6992