Weighted multi-view clustering for handwritten numerals

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

Abstract

Many problems in educational data mining involve datasets that come from multiple different views or sources, which make the data mining task more challenging. However, most existing methods rely equally on every view, something lead to performance degradation in the case of incompatible views. In this work, we focus on a typical multi-view problem, the handwritten numerals clustering. In the proposed algorithm, each view is assigned a weight to express its importance and a simple yet efficient dynamical weight updating strategy is given.

Original languageEnglish
Title of host publicationProceedings of the 23rd International Conference on Computers in Education, ICCE 2015
EditorsYu-Ju Lan, Hsin-Yih Shyu, Chengjiu Yin, Hiroaki Ogata, Wenli Chen, Ming-Fong Jan, Sahana Murthy, Ying-Tien Wu, Siu Cheung Kong, Xiaoqing Gu, Beaumie Kim, Yongwu Miao, Niwat Srisawasdi, Yuping Wang, Chiu-Pin Lin, Carol H.C. Chu, Jari Laru, Miguel Nussbaum, Ma. Mercedes T. Rodrigo, Ju-Ling Shih, Amali Weerasinghe, Weiqin Chen, Feiyue Qiu, Vania Dimitrova, Ching-Kun Hsu, Lung-Hsiang Wong, Maiga Chang, Tore Hoel, Yen-Hui Audrey Li, Jon Mason, Hitoshi Sasaki, Li Zhang
PublisherAsia-Pacific Society for Computers in Education
Pages121-123
Number of pages3
ISBN (Electronic)9784990801458
StatePublished - 2015
Externally publishedYes
Event23rd International Conference on Computers in Education, ICCE 2015 - Hangzhou, China
Duration: 30 Nov 20154 Dec 2015

Publication series

NameProceedings of the 23rd International Conference on Computers in Education, ICCE 2015

Conference

Conference23rd International Conference on Computers in Education, ICCE 2015
Country/TerritoryChina
CityHangzhou
Period30/11/154/12/15

Keywords

  • Clustering
  • Educational data mining
  • Handwritten numerals clustering
  • Multi-view clustering

Fingerprint

Dive into the research topics of 'Weighted multi-view clustering for handwritten numerals'. Together they form a unique fingerprint.

Cite this