Fuzzy C-means in lower dimensional space for topics detection on indonesian online news

Praditya Nugraha, Muhammad Rifky Yusdiansyah, Hendri Murfi

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

2 Citations (Scopus)

Abstract

One of the automated methods for textual data analysis is topic detection. Fuzzy C-Means is a soft clustering-based method for topic detection. Textual data usually has a high dimensional data, which make Fuzzy C-Means fails for topic detection. An approach to overcome the problem is transforming the textual data into lower dimensional space to identify the memberships of the textual data in clusters and use these memberships to generate topics from the high dimensional textual data in the original space. In this paper, we apply the Fuzzy C-Means in lower dimensional space for topic detection on Indonesian online news. Our simulations show that the Fuzzy C-Means gives comparable accuracies than nonnegative matrix factorization and better accuracies than latent Dirichlet allocation regarding topic interpretation in the form of coherence values.

Original languageEnglish
Title of host publicationData Mining and Big Data - 4th International Conference, DMBD 2019, Proceedings
EditorsYuhui Shi, Ying Tan
PublisherSpringer Verlag
Pages269-276
Number of pages8
ISBN (Print)9789813295629
DOIs
Publication statusPublished - 1 Jan 2019
Event4th International Conference on Data Mining and Big Data, DMBD 2019 - Chiang Mai, Thailand
Duration: 26 Jul 201930 Jul 2019

Publication series

NameCommunications in Computer and Information Science
Volume1071
ISSN (Print)1865-0929

Conference

Conference4th International Conference on Data Mining and Big Data, DMBD 2019
Country/TerritoryThailand
CityChiang Mai
Period26/07/1930/07/19

Keywords

  • Clustering
  • Fuzzy c-means
  • Low dimension space
  • Online news
  • Topic detection

Fingerprint

Dive into the research topics of 'Fuzzy C-means in lower dimensional space for topics detection on indonesian online news'. Together they form a unique fingerprint.

Cite this