EDUCATION DATA MINING FOR PREDICTING STUDENTS’ PERFORMANCE

By Dr. GEETHA N DATA SCIENTIST, BENGALURU, Dr.PIYUSH KUMAR PAREEK NITTE MEENAKSHI INSTITUTE OF TECHNOLOGY, BENGALURU and Mr.MANJU D SRI VENKATESWARA COLLEGE OF ENGINEERING. BENGALURU

In the research field, Educational Data Mining (EDM) will be used in various important approaches such as Data Mining (DM), Machine Learning (ML), predicting and statistical analysis in the educational sector. The main goal of mining in the field of education is to develop better mechanisms for analyzing student performance by using the suitable algorithms and techniques. The previous academic performances, in-course performance, student’s demographic status, and knowledge test can be used for the prediction of future performances. This analysis helps the students, organizations, mentors, teachers to improve the students’ performance in their respective field of education and for successful student life. This research is intended to predict the performance of the Engineering students by analyzing various results obtained from different machine learning algorithms and other analytical tools.

• Language: English
• Publisher: Lulu.com
• Release date: Feb 7, 2022
• ISBN: 9781678107680

Book preview

EDUCATION DATA MINING FOR PREDICTING STUDENTS’ PERFORMANCE

Dr. GEETHA N

DATA SCIENTIST, BENGALURU

Dr.PIYUSH KUMAR PAREEK

NITTE MEENAKSHI INSTITUTE OF TECHNOLOGY, BENGALURU

Mr.MANJU D

SRI VENKATESWARA COLLEGE OF ENGINEERING. BENGALURU

ABSTRACT

In the research field, Educational Data Mining (EDM) will be used in various important approaches such as Data Mining (DM), Machine Learning (ML), predicting and statistical analysis in the educational sector. The main goal of mining in the field of education is to develop better mechanisms for analyzing student performance by using the suitable algorithms and techniques. The previous academic performances, in-course performance, student’s demographic status, and knowledge test can be used for the prediction of future performances. This analysis helps the students, organizations, mentors, teachers to improve the students’ performance in their respective field of education and for successful student life. This research is intended to predict the performance of the Engineering students by analyzing various results obtained from different machine learning algorithms and other analytical tools. The main contribution to this research is to identify the weaken areas of the students who are from different backgrounds like urban/rural, state/cbsc and improving the student’s performance by incorporating different training methods. This project provides the different approaches and techniques like collection of the student data, analyzing the data using machine learning techniques and incorporating different training methods for performance improvements of the students. This research also provides the right framework for the different set of students from various different backgrounds, which helps the stakeholders to take the necessary steps to achieve student’s desired results at the university level. Choosing the right metric nonetheless of the student’s academic or demographic (non-academic) is one of the major important factor in this research. Finally, based on the results obtained and analyzing the results a framework has been framed for the performance improvement in the engineering colleges irrespective of their backgrounds.

Table of Contents

1      INTRODUCTION

1.1      NEED

1.1.1      EDUCATIONAL DATA MINING

1.1.2      EDM in Technical Education

1.2      PREDICTION PROCESS

1.3      STUDENT’S DATA GATHERING

1.4      APPLICATIONS

2      INTRODUCTION

3      LITERATURE SURVEY

4      OBJECTIVES

5      METHODOLOGY

5.1      Data Collection:

5.2      Data Pre-processing

5.2.1      Data Cleaning

5.2.2      Feature Selection

5.2.3      Data Transformation:

5.3      Machine Learning Algorithms in Prediction- EDM

5.3.1      ID3 Algorithm

5.3.2      Naïve Bayes:

5.3.3      Logistic Regression

5.3.4      Linear Regression

5.3.5      K-NN Algorithm

5.4      IBM SPSS Modeller

6      RESULTS:

6.1      DM Algorithms- Results:

6.1.1      ID3 Algorithm:

6.1.2      Naïve Bayes:

6.1.3      Logistic Regression:

6.1.4      Linear Regression:

6.1.5      KNN Algorithm:

6.2      Survey Results

6.2.1      Interpersonal skills:

6.2.2      Bridge course:

6.2.3      Correlation between different Attributes that helped in the research

7      Framework for Improvement of Students Performance

7.1      Framework Proposed:

7.1.1      Analytical Skills Training:

7.1.2      Bridge course Training

7.1.3      Interpersonal skills Training

7.1.4      Mentoring/Counselling

FUTURE STUDY:

CONCLUSION:

APPENDIX- B

LIST OF TABLES

LIST OF FIGURES