Academic Warning for College Students for Predicting Student Dropout Rate using Dipper Throated Optimization Algorithm

From past few years, educational institutions record student details for monitoring and analysing individual student performance based on their courses. The academic warning for college students who are logged into the system, the details accessed at the time of learning stage and number of times th...

Full description

Saved in:
Bibliographic Details
Published in:2024 First International Conference on Software, Systems and Information Technology (SSITCON) pp. 1 - 5
Main Author: Wang, Meilin
Format: Conference Proceeding
Language:English
Published: IEEE 18.10.2024
Subjects:
academic warning for college students
Accuracy
Computer aided instruction
dipper throated optimization
dropout
Electronic learning
Feature extraction
massive open online courses
Optimization
prediction
Predictive models
Scalability
Software systems
support vector machine
Support vector machines
Vectors
z-score normalization
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:From past few years, educational institutions record student details for monitoring and analysing individual student performance based on their courses. The academic warning for college students who are logged into the system, the details accessed at the time of learning stage and number of times the particular student logged in particular course. Predicting student dropout rate face an issue in class imbalance such as majority of students remained with the enrolled and huge dropouts. Additionally, the presence of missing data in student's records gives difficulty in predicting dropout rate. To overcome these challenges, Dipper Throated Optimization (DTO) is proposed for predicting optimal features from Massive Open Online Courses (MOOC) dataset. The student's data are taken from the MOOC dataset which undergoes pre-processing by using Z-score normalization technique to reduce feature dominance. Then, the normalized features are further processed into feature selection with DTO method to select optimal features. After that, the selected features are further processed with Support Vector Machine (SVM) classifier to classify student dropout rate as dropout or persist. The proposed DTO method gives better results than existing Logistic Regression (LR) model in terms of accuracy (0.95), precision (0.96), recall (0.98) and F1 score (0.97) respectively.
DOI:10.1109/SSITCON62437.2024.10797187