International Conference on Communication, Computing and Information Technology |
Foundation of Computer Science USA |
ICCCMIT2014 - Number 3 |
March 2015 |
Authors: Lakshmi Devasena C |
d034c088-f735-44fc-9aa0-3a5afdb74617 |
Lakshmi Devasena C . Comparative Analysis of Random Forest, REP Tree and J48 Classifiers for Credit Risk Prediction. International Conference on Communication, Computing and Information Technology. ICCCMIT2014, 3 (March 2015), 30-36.
Envisaging the Credit nonpayer is a risky task of Financial Industries like Banks. find out the defaulter before giving loan is a noteworthy and conflict-ridden task of the Bankers. Classification techniques are the superior choice for predictive analysis like finding the claimant, whether he/she is a modest customer or a cheat. Defining the excellent classifier is a tough assignment for any industrialist like a banker. This gives consent to computer science researchers to drill down efficient research works through evaluating different classifiers and finding out the best classifier for such predictive problems. This research work scrutinizes the efficiency of different Tree Based Classifiers (Random Forest, REP Tree and J48 Classifiers) for the credit risk prediction and compares their robustness through various measures. German credit dataset has been taken and used to envisage the credit risk with the help of open source machine learning tool.