Using minimal generators for composite isolated point extraction and conceptual binary relation coverage: Application for extracting relevant textual features
详细信息 查看全文
- 作者：S. Elloumi ; elloumi.samir@gmail.com" class="auth_mail" title="E-mail the corresponding author ; samir.elloumi@fst.rnu.tn" class="auth_mail" title="E-mail the corresponding author ; F. Ferjani fethif@qu.edu.qa" class="auth_mail" title="E-mail the corresponding author ; A. Jaoua jaoua@qu.edu.qa" class="auth_mail" title="E-mail the corresponding author
- 关键词：Fringe relation ; Isolated labels ; Minimal generator ; Conceptual coverage ; Textual feature extraction
- 刊名：Information Sciences
- 出版年：2016
- 出版时间：1 April 2016
- 年：2016
- 卷：336
- 期：Complete
- 页码：129-144
- 全文大小：837 K
文摘
- •
We present a new approach called “MinGenCoverage” for conceptual binary relation covering.
- •
We start by locating the isolated points and we extract their corresponding mandatory concepts.
- •
In case of properties composition, we consider only the minimal generators as candidates for isolated point extraction.
- •
We applied our approach for textual data and we considered the concept’s labels associated to isolated points as the selected textual features.