Predicting Student Degree Completion Using Random Forest

Cardona, Tatiana; Cudney, Elizabeth A.; Snyder, Jennifer L.; Hoerl, Roger W.

doi:10.18260/1-2--35074

Cited by 2 publications

(2 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Analisis Data Eksplorasi (EDA) mencakup serangkaian teknik dan metode analisis awal yang ditujukan untuk memperoleh pemahaman yang mendalam tentang struktur dan properti suatu kumpulan data [15] Salah satu tujuan utama EDA adalah menemukan pola-pola penting dalam data. Ini mencakup identifikasi distribusi, hubungan antarvariabel, outlier, dan tren yang dapat memberikan wawasan penting terkait dengan fenomena yang sedang diamati.…”

Section: Exploratory Data Analysis (Eda)unclassified

Prediksi Kelulusan Tepat Waktu Siswa SMK Teknik Komputer Menggunakan Algoritma Random Forest

Fatunnisa,

Marcos

2024

JAMIKA

View full text Add to dashboard Cite

School performance can be measured through student completion rates, which is a key indicator. Low graduation rates indicate problems in the education and learning system, which require timely intervention to prevent students from not completing their education. Therefore, predicting graduation rates is crucial for schools in order to determine students who are likely to not complete their education, so as to provide early assistance to improve their academic performance. This research is urgent because by understanding and predicting student completion, schools can allocate resources more effectively to support at-risk students, with the ultimate goal of improving graduation rates and overall school performance. This research uses random forest algorithm with graduation dataset. The distribution of training and test data selection uses stratified random sampling method to ensure a balanced representation of each class generated. The Random Forest model was successfully obtained through training and model evaluation using test data, showing an accuracy of 1.0 or equivalent to 100%. The use of the random forest algorithm on student graduation datasets can be an effective approach in supporting timely graduation prediction due to its high accuracy and the model's ability to recognize both graduating and non-passing students.

show abstract

Section: Exploratory Data Analysis (Eda)unclassified

Prediksi Kelulusan Tepat Waktu Siswa SMK Teknik Komputer Menggunakan Algoritma Random Forest

Fatunnisa,

Marcos

2024

JAMIKA

View full text Add to dashboard Cite

show abstract

“…Determining the factors influencing student retention and completion rates provides insight into opportunities for intentional student advising, better planning, and development of retention strategies based on student needs (Slim et al, 2014). In recent years machine learning techniques have been applied to analyze student data, which aligns with the focus of improving information processing through data mining (Cardona et al, 2019a) using methods such as neural networks (NN) (Cardona et al, 2019b) and support vector machines (SVM) (Cardona & Cudney, 2019). According to the literature, these techniques offer student dropout predictions with high confidence (Pereira & Zambrano, 2017).…”

mentioning

confidence: 99%

Data Mining and Machine Learning Retention Models in Higher Education

Cardona

Cudney

Hoerl

et al. 2020

Journal of College Student Retention: Research, Theory & Pr

View full text Add to dashboard Cite

This study presents a systematic review of the literature on the predicting student retention in higher education through machine learning algorithms based on measures such as dropout risk, attrition risk, and completion risk. A systematic review methodology was employed comprised of review protocol, requirements for study selection, and analysis of paper classification. The review aims to answer the following research questions: (1) what techniques are currently used to predict student retention rates, (2) which techniques have shown better performance under specific contexts?, (3) which factors influence the prediction of completion rates in higher education?, and (4) what are the challenges with predicting student retention? Increasing student retention in higher education is critical in order to increase graduation rates. Further, predicting student retention provides insight into opportunities for intentional student advising. The review provides a research perspective related to predicting student retention using machine learning through several key findings such as the identification of the factors utilized in past studies and methodologies used for prediction. These findings can be used to develop more comprehensive studies to further increase the prediction capability and; therefore, develop strategies to improve student retention.

show abstract

Predicting Student Degree Completion Using Random Forest

Abstract: Her research interests include statistical modeling, Operations research and Data Science. She has served as a head teaching assistant for four semesters in operations management and project management in the MS&T.

Cited by 2 publications

References 7 publications

Prediksi Kelulusan Tepat Waktu Siswa SMK Teknik Komputer Menggunakan Algoritma Random Forest

Prediksi Kelulusan Tepat Waktu Siswa SMK Teknik Komputer Menggunakan Algoritma Random Forest

Data Mining and Machine Learning Retention Models in Higher Education

Contact Info

Product

Resources

About