Credit Evaluation Model For Banks Using Data Mining Techniques

4296 Words18 Pages
Credit Evaluation Model for Banks Using Data Mining Techniques
By Sharan Brahmanapally, Bachelor of Technology

A Project submitted in Partial
Fulfillment of the Requirements
For the Degree of
Master of Science
In the field of Industrial Engineering

Advisory Committee:
Dr. Hoo Sang Ko

Graduate School
Southern Illinois University Edwardsville
August, 2015

TABLE OF CONTENTS
TABLE OF CONTENTS ii
LIST OF FIGURES iii
LIST OF TABLES iii
ABSTRACT iv
CHAPTER 1 1
INTRODUCTION 1
1.1 Introduction 1
1.1.1 Decision Process for Credit Evaluation 3
1.2 Problem Statement 4
1.3 Aim of the Project 4
1.4 Objectives of the Project 5
CHAPTER 2 6
LITERATURE REVIEW 6
2.1 Introduction 6
2.2 Theoretical Background 6
2.2.1 Decision Trees 6
2.2.2
…show more content…
It is a process of analyzing the relationship among the data from various perspectives and summarizing it into valuable information. It also assists the banks to look for hidden patterns in a group and discovers unknown relationships in the data. These data mining techniques facilitate useful data interpretations for the banking sector to avoid customer attrition. An accurate prediction on the credit approval is important to prospective homeowners, developers, investors, appraisers, tax assessors and other real estate market participants without fraudulence. People who are looking to buy a new place or thing, tend to be more conservative with their budget and acquiring loans from financial institutions. The credit functionality is prime for any banking system over the tentative market conditions. The lack of general credit review system & precise methods in banks are the important reasons, why an expert support system is necessary.
This project aims to evaluate the performance and accuracy of classification models for credit evaluation. The classification models are developed based on decision trees (J48 & CART), Support Vector Machine (SVM) and Logistic Regression along with Ensemble Methods. We used a credit approval dataset from UCI repository to compare the accuracies of the various data mining techniques. All the developed models achieved more than 85% accuracy, and
Get Access