TY - CHAP
T1 - Smart School Selection with Supervised Machine Learning
AU - Kumar, Deepak
AU - Verma, Chaman
AU - Stoffová, Veronika
AU - Illes, Zoltán
AU - Gupta, Anish
AU - Bakariya, Brijesh
AU - Singh, Pradeep Kumar
N1 - Publisher Copyright:
© 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.
PY - 2022/11/3
Y1 - 2022/11/3
N2 - In today’s competitive academic environment, parents and students usually face the school selection problem for a decade. Keeping the question in mind, we proposed to seek the select significant features (academic, social, demographic, etc.) with the help of machine learning algorithms (Support Vector Machine (SVM), Extreme Gradient Boosting (XGB), and Random Forest (RF)). These features will be helpful for guardians/parents, schools, and teachers in deciding the students the best school for their education. We used a statistical approach (one-way ANOVA) to investigate the impact of school selection reasons towards student’s grades. The standard open data set of Portuguese secondary school student was used here for analysis. A Synthetic Minority Over-sampling Technique-Nominal Continuous (SMOTE-NC) technique was used for resampling the imbalanced Reason target class. The proposed automatic school selection recommender might be helpful in every academic community and intelligent education. We found school selection reasons have a statistically significant impact on the final grade. The RF comes out as a strong predictor among all proposed models with an accuracy of 71%. The final grade, going out with friends, parents’ job, and activities are the essential features for Smart School Selection.
AB - In today’s competitive academic environment, parents and students usually face the school selection problem for a decade. Keeping the question in mind, we proposed to seek the select significant features (academic, social, demographic, etc.) with the help of machine learning algorithms (Support Vector Machine (SVM), Extreme Gradient Boosting (XGB), and Random Forest (RF)). These features will be helpful for guardians/parents, schools, and teachers in deciding the students the best school for their education. We used a statistical approach (one-way ANOVA) to investigate the impact of school selection reasons towards student’s grades. The standard open data set of Portuguese secondary school student was used here for analysis. A Synthetic Minority Over-sampling Technique-Nominal Continuous (SMOTE-NC) technique was used for resampling the imbalanced Reason target class. The proposed automatic school selection recommender might be helpful in every academic community and intelligent education. We found school selection reasons have a statistically significant impact on the final grade. The RF comes out as a strong predictor among all proposed models with an accuracy of 71%. The final grade, going out with friends, parents’ job, and activities are the essential features for Smart School Selection.
KW - One-way-ANOVA
KW - RF
KW - SMOTE-NC
KW - SVM
KW - Supervised learning
KW - XGB
UR - http://www.scopus.com/inward/record.url?scp=85141874753&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-08815-5_13
DO - 10.1007/978-3-031-08815-5_13
M3 - Chapter
AN - SCOPUS:85141874753
T3 - Studies in Computational Intelligence
SP - 221
EP - 235
BT - Studies in Computational Intelligence
PB - Springer Science and Business Media Deutschland GmbH
ER -