Machine Learning based Prediction of Dropout Students from the Education University using SMOTE

Revathy, M.; Kamalakkannan, S.; Kavitha, P.

doi:10.1109/icssit53264.2022.9716450

Cited by 5 publications

(4 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Effective techniques to improve the performance of classifiers in the presence of a class imbalance in training data [42] include data-level approaches such as undersampling, oversampling, and their combinations [43,44]. In the context of this study, the majority class consists of successful students, and this asymmetry can lead a classifier to primarily predict that students do not drop out of the MOOC, which is an undesirable situation as students at risk are incorrectly identified as non-dropouts, resulting in them not receiving the necessary support intervention.…”

Section: Handling Imbalanced Classes In Dropout Predictionmentioning

confidence: 99%

“…As a variation of the SMOTE technique, He et al [48] proposed ADASYN (adaptive synthetic sampling method), which generates more synthetic data for the instances that are more difficult to learn [49]. Improved predictive performance of classification models through the combined use of PCA and resampling techniques has been reported (e.g., [42]).…”

Section: Handling Imbalanced Classes In Dropout Predictionmentioning

confidence: 99%

“…Based on the above theoretical background, this study supports the idea that it is interesting to investigate the early prediction of dropout in the MOOC "Programming in Python for Non-Programmers", taking into consideration primarily the SRL (self-regulated learning) factor, after using various oversampling techniques. For this purpose, we employ five classic educational models [50] commonly used in the literature for comparing oversampling techniques when dealing with class imbalance [36,37,42,46] predictive models (NB, LR, SVM, DT, kNN), and we consider data from student questionnaires regarding their SRL, as well as other typical data, easily provided for this purpose, aiming to identify at-risk students as early as possible and support them with appropriate interventions to complete the course.…”

Section: Rationale and Research Questionsmentioning

confidence: 99%

See 2 more Smart Citations

Predictive Modeling of Student Dropout in MOOCs and Self-Regulated Learning

Psathas,

Chatzidaki,

Demetriadis

2023

Computers

View full text Add to dashboard Cite

The primary objective of this study is to examine the factors that contribute to the early prediction of Massive Open Online Courses (MOOCs) dropouts in order to identify and support at-risk students. We utilize MOOC data of specific duration, with a guided study pace. The dataset exhibits class imbalance, and we apply oversampling techniques to ensure data balancing and unbiased prediction. We examine the predictive performance of five classic classification machine learning (ML) algorithms under four different oversampling techniques and various evaluation metrics. Additionally, we explore the influence of self-reported self-regulated learning (SRL) data provided by students and various other prominent features of MOOCs as potential indicators of early stage dropout prediction. The research questions focus on (1) the performance of the classic classification ML models using various evaluation metrics before and after different methods of oversampling, (2) which self-reported data may constitute crucial predictors for dropout propensity, and (3) the effect of the SRL factor on the dropout prediction performance. The main conclusions are: (1) prominent predictors, including employment status, frequency of chat tool usage, prior subject-related experiences, gender, education, and willingness to participate, exhibit remarkable efficacy in achieving high to excellent recall performance, particularly when specific combinations of algorithms and oversampling methods are applied, (2) self-reported SRL factor, combined with easily provided/self-reported features, performed well as a predictor in terms of recall when LR and SVM algorithms were employed, (3) it is crucial to test diverse machine learning algorithms and oversampling methods in predictive modeling.

show abstract

Section: Handling Imbalanced Classes In Dropout Predictionmentioning

confidence: 99%

Section: Handling Imbalanced Classes In Dropout Predictionmentioning

confidence: 99%

Section: Rationale and Research Questionsmentioning

confidence: 99%

See 1 more Smart Citation

Predictive Modeling of Student Dropout in MOOCs and Self-Regulated Learning

Psathas,

Chatzidaki,

Demetriadis

2023

Computers

View full text Add to dashboard Cite

show abstract

“…The two main approaches to deal with imbalanced target variables is to use an oversampling technique which creates additional minority classes (student records with a course failure) or undersampling techniques to randomly delete majority classes (student records without a course failure). We selected to turn our dataset into a balanced dataset by using an oversampling method called synthetic minority oversampling technique (SMOTE) which is a common method for dealing with models predicting student success in higher education [ 45 , 46 , 47 ]. After performing SMOTE, our dataset will result in students who failed a course being equal to the number of students who did not fail a course.…”

Section: Simulation Of Dataset and Creation Of A Random Forest Machin...mentioning

confidence: 99%

Using Machine Learning in Veterinary Medical Education: An Introduction for Veterinary Medicine Educators

Hooper,

Hecker,

Artemiou

2023

Veterinary Sciences

View full text Add to dashboard Cite

Machine learning (ML) offers potential opportunities to enhance the learning, teaching, and assessments within veterinary medical education including but not limited to assisting with admissions processes as well as student progress evaluations. The purpose of this primer is to assist veterinary educators in appraising and potentially adopting these rapid upcoming advances in data science and technology. In the first section, we introduce ML concepts and highlight similarities/differences between ML and classical statistics. In the second section, we provide a step-by-step worked example using simulated veterinary student data to answer a hypothesis-driven question. Python syntax with explanations is provided within the text to create a random forest ML prediction model, a model composed of decision trees with each decision tree being composed of nodes and leaves. Within each step of the model creation, specific considerations such as how to manage incomplete student records are highlighted when applying ML algorithms within the veterinary education field. The results from the simulated data demonstrate how decisions by the veterinary educator during ML model creation may impact the most important features contributing to the model. These results highlight the need for the veterinary educator to be fully transparent during the creation of ML models and future research is needed to establish guidelines for handling data not missing at random in medical education, and preferred methods for model evaluation.

show abstract

Using Machine Learning in Veterinary Medical Education: An Introduction for Veterinary Medicine Educators

Hooper¹,

Hecker²,

Artemiou³

2023

Preprint

View full text Add to dashboard Cite

Machine learning (ML) offers potential opportunities to enhance the learning, teaching and assessments within veterinary medical education including but not limited to assisting with admissions processes as well as student progress evaluations. The purpose of this primer is to assist veterinary educators in appraising and potentially adopting these rapid upcoming advances in data science and technology. In the first section, we introduce ML concepts and highlight similarities/differences between ML and classical statistics. In the second section, we provide a step-by-step worked example using simulated veterinary student data to answer a hypothesis driven question. Python syntax with explanations is provided within the text to create a random forest ML prediction model and within each step, specific considerations such as how to manage incomplete student records are highlighted when applying ML algorithms within the veterinary education field. The results from the simulated data demonstrate how decisions by the veterinary educator during ML model creation may impact the most important features contributing to the model. These results highlight the need for the veterinary educator to be fully transparent during the creation of ML models and future research is needed to establish guidelines for handling data not missing at random in medical education, and preferred methods for model evaluation.

show abstract

Machine Learning based Prediction of Dropout Students from the Education University using SMOTE

Cited by 5 publications

References 26 publications

Predictive Modeling of Student Dropout in MOOCs and Self-Regulated Learning

Predictive Modeling of Student Dropout in MOOCs and Self-Regulated Learning

Using Machine Learning in Veterinary Medical Education: An Introduction for Veterinary Medicine Educators

Using Machine Learning in Veterinary Medical Education: An Introduction for Veterinary Medicine Educators

Contact Info

Product

Resources

About