Application of Artificial Neural Networks to Predict the Use of Mobile Learning by University Students

Abstract

The use of mobile devices has become pervasive in recent times, constituting an essential component of daily life. Mobile phones have enabled certain minorities to attain access to the Internet, news, and knowledge, thereby indicating their potential to reduce the digital divide experienced by ethnic groups and those from low socioeconomic backgrounds. This phenomenon has generated academic interest in the utilization of mobile devices to facilitate learning, as these devices merge the lines between computing and communications, giving access to both. The objective of this study is to ascertain the inclination of Peruvian higher education students to use mobile devices for learning. This will be achieved through the use of an anticipated model based on artificial neural networks (ANNs). ANNs are supervised machine learning techniques that imitate the organization and operation of the human brain to process data and render decisions. ANNs are computer systems that can learn from observation and experience, much like the human brain, and can subsequently use the acquired knowledge to recognize patterns and make predictions. The objective of this study is to assess the intention of Peruvian tertiary education students to employ mobile devices for learning by creating a predictive model that relies on ANNs. Among the main findings, it is evident that the ANN with optimal performance has 10 neurons within its hidden layer. Factors such as experience with virtual subjects, frequency of use, and coverage are crucial for the two intention variables. This enables directed prediction efforts towards the most significant variables identified by their importance.