RESIDENCE STATE AND COUNTRY PREDICTION OF STUDENT TOWARDS ICT FOR THE REAL-TIME
RESIDENCE STATE AND COUNTRY PREDICTION OF STUDENT TOWARDS ICT FOR THE REAL-TIME

Summary/Abstract: An experimental study was conducted to predict the residence state and country of students based on their response provided in the two different ICT survey held during the academic year 2014-2015 and the academic year 2017-2018. The first dataset consisted of 560 instances and 59 features and the second dataset was comprised of 331 instances and 46 features. The authors considered the state in the first dataset and country in the second dataset as the response variable and the rest of all are assumed as predictors after self-reduction few features. The datasets are trained and tested with the splitting (hold out) and k-fold Cross-Validation (CV) using three popular supervised machine learning classifiers named Artificial Neural Network (ANN), Support Vector Machine (SVM) and Random Forest (RF) in the Weka 3.8.1. In the state prediction, the RF classifier outperformed with the highest prediction accuracy of 82.11% the ANN and SMO at 560-folds of the CV method. The maximum accurate prediction count for the state class was 457 out of a total of 560. In the country prediction, the best fitting model was found SVM at 331 folds with a prediction accuracy of 89.73%. The highest right prediction counts for the country class was 297 out of a total of 331. Also, the highest association among features were calculated by the RF and the SVM. On one hand, in the state prediction, the statistical t-test found both algorithm SVM and ANN statistically significant and another hand, only the ANN is found significant in the country prediction. Further, these state and country predictive models may support the real-time prediction module of student's demography prediction towards technological awareness.

• Details
• Contents