The singular value decomposition-based anchor word selection method for separable nonnegative matrix factorization

Delano Novrilianto, Hendri Murfi, Arie Wibowo

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

2 Citations (Scopus)

Abstract

One of the recent methods for the topic modeling is separable nonnegative matrix factorization (SNMF). In general, SNMF consists of three main steps, which are, generating a word co-occurrence matrix, selecting anchor words, and recovering a topic matrix. The anchor words strongly influence the interpretability of extracted topics. In this paper, we propose a new method for selecting the anchor words by using singular value decomposition (SVD). We assume that the most dominant words in each latent semantics created by SVD are the potential candidates for the anchor words. Our simulations show that the SVD-based anchor word selection method can reach better interpretability scores of extracted topics than the common convex hull-based method on two of three datasets.

Original languageEnglish
Title of host publicationProceedings of the 2017 International Conference on Asian Language Processing, IALP 2017
EditorsRong Tong, Minghui Dong, Yanfeng Lu, Yue Zhang
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages289-292
Number of pages4
ISBN (Electronic)9781538619803
DOIs
Publication statusPublished - 21 Feb 2018
Event21st International Conference on Asian Language Processing, IALP 2017 - Singapore, Singapore
Duration: 5 Dec 20177 Dec 2017

Publication series

NameProceedings of the 2017 International Conference on Asian Language Processing, IALP 2017
Volume2018-January

Conference

Conference21st International Conference on Asian Language Processing, IALP 2017
Country/TerritorySingapore
CitySingapore
Period5/12/177/12/17

Keywords

  • anchor words
  • separable nonnegative matrix factorization
  • singular value decomposition
  • topic modeling

Fingerprint

Dive into the research topics of 'The singular value decomposition-based anchor word selection method for separable nonnegative matrix factorization'. Together they form a unique fingerprint.

Cite this