Implementation of Naïve Bayes Classifier Algorithm to Categorize Indonesian Song Lyrics Based on Age

Closed

Moch. Fadli Shadiqin Thirafi, Faisal Rahutomo

2018 3rd International Conference on Sustainable Information Engineering and Technology, SIET 2018 - Proceedings Conference paper Cited by 11 Quartile

Abstract

The rapid development of technology accelerates the process of distributing the song in Indonesia. But that songs have not been followed by the age label of the listener, such as the age label on the movie. It certainly makes people difficult to choose songs that appropriate to their age, so people are free to enjoy a variety of songs and probably will earn the bad effect to the mental development. Therefore, there need to classify the lyrics of Indonesian language songs according to the age of the community. Several types of research classify song by its genre like pop, rock, jazz, i.e., classify by emotion of the song, and region of the song. In this study, develops a system that can categorize the lyrics into four age groups of the listener, namely children, adolescents, adults and all ages using naïve bayes classifier (NBC) algorithm with the number of datasets are 400 titles obtained through crawling on lirik.kapanlagi.com website. The testing algorithm using training data of 60, 70, 80, and 90 percent from the overall dataset with equitable distribution resulted accuracies are 62.5%, 65%, 67.5%, and 67.5% respectively. © 2018 IEEE.

Affiliations

Information Technology Department, State Polytechnic of Malang, Malang, Indonesia