Abstract

Abstract This study aims to use predictive models in order to figure out students that are in likelihood of attrition as well as determine the factors that may contribute to this attrition. Student attrition is the occurrence in which students leave the institution without finishing their studies or taking the degree. It includes students either willingly or involuntarily terminating their education and failing to graduate. The findings will enable relevant parties to establish successful strategies or approaches and activities to assist in reducing the number of students who leave. With the testing and analysis approaches, it was recognized that SVM 1 was the most effective algorithm holding a 90.928% accuracy rate with precision 0.921. Students attrition of classes for a variety of motives including personal or academic challenges that prevent them from becoming engaged students. The paper seeks to investigate the numerous factors that impact the decision of the students to attrite, as well as the optimal prediction model. Kaggle will be used as a source for the data and SPSS will be used for examining and preprocessing the chosen data. Also, through SPSS Modeler will additionally be used to extract visual insights using the provided dataset.

Library of Congress Subject Headings

College dropouts--Forecasting; Machine learning; Regression analysis; Principal components analysis; Support vector machines

Publication Date

5-21-2024

Document Type

Thesis

Student Type

Graduate

Degree Name

Professional Studies (MS)

Department, Program, or Center

Graduate Programs & Research

Advisor

Sanjay Modak

Advisor/Committee Member

Hammou Messatfa

Campus

RIT Dubai

Plan Codes

PROFST-MS

Download

Share

COinS