Naïve Bayesian classifier algorithm and neural network time series for identification of lecturer publications in realizing internationalization of Universitas Negeri Semarang

Abstract

The Naïve Bayes Classifier is a classification method rooted in the Bayes theorem. The main characteristic of the Naïve Bayes Classifier is the very strong (naïf) assumption of the independence of each condition/event. If X is the vector that enters the feature and Y is the class label, Naïve Bayes is written with P (X | Y). This notation means that the probability of a class Y label is obtained after the X features are observed. This notation is also called the posterior probability for Y, while P (Y) is called the prior probability Y. One of the important factors included in the UNNES internationalization assessment is the results of lecturer publications both at the national and international levels. The artificial neural network as one of the information processing systems designed by imitating the workings of the human brain in completing a problem by doing the learning process through changes in synaptic weight can be used to analyze the productivity of the performance of UNNES lecturers and staff in realizing the year of reputation proclaimed in 2017 Research related to this theme has been carried out in 2017 and shows that the factors of knowledge and behavior of lecturers and education staff have the most influence on the achievement of the year of reputation at FMIPA UNNES in 2017.