Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Vol 3 No 6 (2019): Juni 2019

Klasifikasi Emosi Lirik Lagu menggunakan Improved K-Nearest Neighbor dengan Seleksi Fitur dan BM25

Febrina Sarito Sinaga (Fakultas Ilmu Komputer, Universitas Brawijaya)
Indriati Indriati (Fakultas Ilmu Komputer, Universitas Brawijaya)
Bayu Rahayudi (Fakultas Ilmu Komputer, Universitas Brawijaya)



Article Info

Publish Date
23 Jul 2019

Abstract

Emotions is a person's reaction or feeling into a situation. Emotion is temporary that can occurred by a stimulus because of some people around and the environment. One of example an environment that can trigger someone's emotion is from the song being listened to. Song lyrics are the parts that can build emotions. Choosing the right words for lyrics are very important because it will create the right emotion. In this case the emotional classification of song lyricis will be done classifying process using several methods are Improved K-Nearest Neighbor, BM25 and feature selection. The proses of classification have seome stages, which is the stage of pre-processing documents, stages of calculation the BM25 score and sorting document, and the classification stage with using the algorithm is Improved K-Nearest Neighbor. The testing for classifications was done uses 6 times K-fold and use the confusion matrix. This research is the amount of training data used by 100 documents, and testing data used by 20 testing documents. In the all the tests have done obtained the best average results when the value K = 55 with a result of f-measure is 0.6693, recall is 0.6582, and precision is 0.7427.

Copyrights © 2019






Journal Info

Abbrev

j-ptiik

Publisher

Subject

Computer Science & IT Control & Systems Engineering Education Electrical & Electronics Engineering Engineering

Description

Jurnal Pengembangan Teknlogi Informasi dan Ilmu Komputer (J-PTIIK) Universitas Brawijaya merupakan jurnal keilmuan dibidang komputer yang memuat tulisan ilmiah hasil dari penelitian mahasiswa-mahasiswa Fakultas Ilmu Komputer Universitas Brawijaya. Jurnal ini diharapkan dapat mengembangkan penelitian ...