Indonesians' Song Lyrics Topic Modelling Using Latent Dirichlet Allocation

Enrico Laoh, Isti Surjandari Prajitno, Limisgy Ramadhina Febirautami

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Citations (Scopus)

Abstract

Lyrics in songs have important roles to give identity and storyline of songs. Lyrics are also considered as one of influence factor for popularity of the songs. However, it is difficult to manually assess the topic from numerous songs especially Indonesian's songs which have high poetic level in the lyrics. Knowing the intrinsic value of lyrics from numerous songs becomes a challenge, especially the lyrics written in complicated and complex language like Bahasa. This paper aims to know and interpret the topics from Indonesian's song lyrics. Indonesian's songs were obtained from daily TOP 200 Spotify in January 2017 - January 2018 with 193 different songs using Bahasa in the lyrics. Latent Dirichlet Allocation (LDA) for the topic modeling was used in this paper. Using 10 topics based on perplexity results, LDA has proper way to interpret the topics in numerous songs by giving information about top words in every topic and topic probabilities for each document or song.

Original languageEnglish
Title of host publicationProceedings - 2018 5th International Conference on Information Science and Control Engineering, ICISCE 2018
EditorsYun Cheng, Shaozi Li, Ying Dai
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages270-274
Number of pages5
ISBN (Electronic)9781538655009
DOIs
Publication statusPublished - 14 Jan 2019
Event5th International Conference on Information Science and Control Engineering, ICISCE 2018 - Zhengzhou, Henan, China
Duration: 20 Jul 201822 Jul 2018

Publication series

NameProceedings - 2018 5th International Conference on Information Science and Control Engineering, ICISCE 2018

Conference

Conference5th International Conference on Information Science and Control Engineering, ICISCE 2018
CountryChina
CityZhengzhou, Henan
Period20/07/1822/07/18

Keywords

  • Indonesian
  • latent dirichlet allocation
  • lyrics
  • song
  • text mining
  • topic modelling

Fingerprint Dive into the research topics of 'Indonesians' Song Lyrics Topic Modelling Using Latent Dirichlet Allocation'. Together they form a unique fingerprint.

Cite this