University Selection Model Using Machine Learning Techniques

Mostafa, Lamiaa; Beshir, Sara

doi:10.1007/978-3-030-76346-6_60

Cited by 5 publications

(1 citation statement)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…After evaluating the classifiers on this data, the NB obtained the best classification result, reaching an accuracy of 92.1% compared to 90.37% for the Neural Networks (NNs) and 88.13% by the SVM. Similarly, Naïve Bayes had a higher accuracy than SVM in [16], at 93%. Artificial Neural Network (ANN) has been used in this research field.…”

Section: Literature Reviewmentioning

confidence: 93%

A Recommendation System for Selecting the Appropriate Undergraduate Program at Higher Education Institutions Using Graduate Student Data

2022

View full text Add to dashboard Cite

Selecting the appropriate undergraduate program is a critical decision for students. Many elements influence this choice for secondary students, including financial, social, demographic, and cultural factors. If a student makes a poor choice, it will have implications for their academic life as well as their professional life. These implications may include having to change their major, which will cause a delay in their graduation, having a low grade-point average (GPA) in their chosen major, which will cause difficulties in finding a job, or even dropping out of university. In this paper, various supervised machine learning techniques, including Decision Tree, Random Forest, and Support Vector Machine, were investigated to predict undergraduate majors. The input features were related to the student’s academic history and the job market. We were able to recommend the program that guarantees both a high academic degree and employment, depending on previous data and experience, for Master of Business Administration (MBA) students. This research was conducted based on a published research and using the same dataset and aimed to improve the results by applying hyper-tuning, which was absent in previous research. The obtained results showed that our work outperformed the work of the published research, where the random forest exceeded the other classification techniques and reached an accuracy of 97.70% compared to 75.00% on the published research. The importance of features was also investigated, and it was found that the degree percentage, MBA percentage, and entry test result were the top contributing features to the model.

show abstract

Section: Literature Reviewmentioning

confidence: 93%