Scribd
Upload
12 views2 pages

SFA paper 4

The paper presents a systematic literature review on the use of machine learning techniques for predicting student performance and dropout rates from 2009 to 2021. It highlights the effectiveness of various ML methods in enhancing educational outcomes and emphasizes the need for improved predictive models and broader datasets. Future research should focus on optimizing feature selection, addressing data imbalance, and developing tailored solutions for different learning environments.

Uploaded by

sahilchakrabortty912
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
12 views2 pages

SFA paper 4

The paper presents a systematic literature review on the use of machine learning techniques for predicting student performance and dropout rates from 2009 to 2021. It highlights the effectiveness of various ML methods in enhancing educational outcomes and emphasizes the need for improved predictive models and broader datasets. Future research should focus on optimizing feature selection, addressing data imbalance, and developing tailored solutions for different learning environments.

Uploaded by

sahilchakrabortty912
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 2

#PAPER – 4 (ASystematic Literature Review of Student’ Performance

Prediction Using Machine Learning Techniques)


Introduction
The paper highlights how Educational Data Mining (EDM) has become critical in
understanding and improving student performance. Machine Learning (ML)
techniques are increasingly used to predict students at risk, identify dropouts,
and enhance learning strategies. Challenges like high dropout rates in e-
learning systems and lack of direct interaction between students and
instructors make predictive models essential for improving outcomes.

Abstract
The study provides a systematic literature review on student performance
prediction using ML techniques. It examines literature from 2009 to 2021,
emphasizing the use of EDM to analyze data from institutional and online
platforms. Findings suggest that ML methods like decision trees, support vector
machines, and neural networks significantly improve predictions of student
risks and dropout rates, aiding timely interventions and better outcomes.

Methodology
The authors used Okoli's guide for conducting a systematic literature review.
They sourced data from six academic databases (e.g., IEEE Xplore, SpringerLink)
and applied strict inclusion/exclusion criteria to filter relevant studies. The
review spans 78 papers published between 2009 and 2021, focusing on the use
of ML for student performance and dropout prediction.
Key research questions:
1. What problems exist in the field of student performance prediction?
2. What ML solutions address these problems?
3. How productive has research been in this domain?
Technology Used
1. Machine Learning Algorithms: Decision Trees, Neural Networks, Support
Vector Machines, Naive Bayes, Gradient Boosting.
2. Data Sources: Institutional databases and online learning platforms like
MOOCs, LMS, and Intelligent Tutoring Systems.
3. Software Tools: Tools like Weka for algorithm implementation and data
preprocessing.

Future Scope
The review suggests the need for:
1. Enhanced predictive models tailored to specific learning environments.
2. Broader datasets that combine academic and non-academic attributes
for deeper insights.
3. Integration of Auto ML to optimize feature selection and predictive
accuracy.
4. Solutions for handling data imbalance in e-learning environments and
better strategies to reduce dropout rates.
This summary provides a concise understanding of the paper's objectives,
findings, and implications for future research.

You might also like