Implementation of Naïve Bayes Classifier-based Machine Learning to Predict and Classify New Students at Matana University

Abstract

Data is growing fast, triggered by the wider use of information technology (IT) by the public or organizations. In line with data growth, data processing is also developing. One of them is data mining. Data mining is very necessary in business to obtain the information that needed to make business strategy such as marketing strategy. At Matana University, the Marketing Department needs to utilize data mining to optimize the achievement of its targets, particularly to predict and classify the prospective student data. Current data mining processing machines already have the high performances with large storage capacities. To improve the performance of data mining processing machines, machine learning is applied. The machine becomes intelligent and able to learn from the provided data. The application of machine learning to predict and classify new students is based on supervised learning by applying the Naïve Bayes Classifier (NBC) algorithm. The data used is the data of prospective students who have registered at Matana University. Machine learning is built, using the Python programming language. The result of the application has an accuracy of 0.73 (73%) and very helpful to the head of marketing in making marketing strategies. In the future, several developments can be done such as using other algorithms, accessed by smartphone, using dashboards in visualization, or adding data attributes such as parental income, religion, hobby, future goal and so on.