Application of Prediction Time of Graduation Using the Naïve Bayes Algorithm With the Python Program

Abstract

Accreditation is a process to ensure the quality of a university and study program. There are several factors that determine the quality standard of accreditation. One of them is the time of graduation. However, there is no means that can be used to predict early student graduation time. Therefore, this study aims to create a means that can predict early graduation time. In this study, data mining methods were used, namely the Naïve Bayes algorithm. After that, data processing and application development will be carried out using the Python program. The data used in the data mining process is three years of historical data and the data used for the trial are active student data for the second and third years. There are 5 types of patterns with an accuracy value of 81%, 87%, 92%, 92%, and 95%.

Improving Cyber Situational Awareness via Data mining and Predictive Analytic Techniques

“…One of the most common applications of Naïve Bayes is spam filtering and text categorization. (Murphy, 2006) Naïve Bayes assumes that all variables in the dataset are independent from each other and calculate the probability of different events based on the variables, however, some authors such as Ray (2015) consider this as a disadvantage because it is almost impossible to obtain data set that its features are completely irrelevant in real life. One of the main advantages of this algorithm is that it only needs a small number of training set in order to build a classifier and according to Murphy (2006), Naïve Bayes is much more efficient, accurate and faster compared to some other algorithms such as decision trees in some classification problems.…”