Attitude Prediction towards ICT and Mobile Technology for the Real-Time: An Experimental Study using Machine Learning
Attitude Prediction towards ICT and Mobile Technology for the Real-Time: An Experimental Study using Machine Learning

Summary/Abstract: An experimental study was conducted to predict the attitude of students towards Information and Communication Technology (ICT) and Mobile Technology (MT) in Indian and Hungarian university. We structured five levels of attitude such as Very High, High, Moderate, Low and Very Low to measure the attitude of students. A primary dataset was gathered from two popular universities belongs to India and Hungary in the year 2018. The dataset consists of 331 instances and 46 features belong to the 4 major ICT parameters named attitude, development and availability, educational benefits and usability of modern ICT resources and MT. This paper focuses on the prediction of the students’ attitude towards ICT and MT in higher education of both countries individually and collectively. Experiments were conducted using four machine learning classifiers multilayer perceptron (ANN), Support vector machine (SVM), K-nearest neighbours (KNN) and Discriminant (DISC). The multiclassification problem was solved using test, train and validated datasets separately and jointly in SPSS IBM Modeler version 18.1. It is also revealed that the future awareness level for attitude will be Very High or Moderate in both countries. The results of the paper reveal that the feature aggregation with train-test-validation technique improves the SVM's prediction accuracy for attitude as compared to others. Further, the attitude prediction accuracy for Hungarian students was found 98.1% (by SVM) which is greater than the accuracy of 96.6% (by SVM) for Indian students. It was also concluded that SVM classifier outperformed others in prediction of attitude. Further, we recommend presented predictive models to be implemented as real-time awareness level prediction of the university's students in future research.

• Details
• Contents