Domination dependency analysis of sales marketing based on multi-label classification using label ordering and cycle chain classification

Boonyarit Soonsiripanichkul, Tomohiro Murata

研究成果: Conference contribution

2 被引用数 (Scopus)

抄録

Multi-label classification (MLC) is the technique to solve conditional problem where decisions is a set of labels. Classification is the learning task by using historical examples to make model of conditions to make decision of unseen example. To improve decision in MLC. We can use advantage of domination analysis or dependency relations between the classes by using enhanced Bayesian Chain Classifier (BCC). We introduce an approach for chaining classifier primary order by its individual label accuracy priority (LPC-CC). Our method considers the dependencies among label based on label accuracy priority ordering. Thus, Binary Relevance (BR) theoretical is used for label sequencing priority, and cycle classifiers chain using nave Bayes is for finding domination. The model have been tested on 2 well-known benchmark datasets named Yeast and Emotions and a collection of car sell records from a Thailand Automotive Company.

本文言語English
ホスト出版物のタイトルProceedings - 2016 5th IIAI International Congress on Advanced Applied Informatics, IIAI-AAI 2016
出版社Institute of Electrical and Electronics Engineers Inc.
ページ1048-1053
ページ数6
ISBN(電子版)9781467389853
DOI
出版ステータスPublished - 2016 8 31
イベント5th IIAI International Congress on Advanced Applied Informatics, IIAI-AAI 2016 - Kumamoto, Japan
継続期間: 2016 7 102016 7 14

Other

Other5th IIAI International Congress on Advanced Applied Informatics, IIAI-AAI 2016
国/地域Japan
CityKumamoto
Period16/7/1016/7/14

ASJC Scopus subject areas

  • 情報システム
  • コンピュータ ネットワークおよび通信
  • コンピュータ サイエンスの応用
  • コンピュータ ビジョンおよびパターン認識

フィンガープリント

「Domination dependency analysis of sales marketing based on multi-label classification using label ordering and cycle chain classification」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル