Change points detection in crime-related time series: An on-line fuzzy approach based on a shape space representation
详细信息 查看全文
- 作者：Fabrizio Albertettia ; fabrizio.albertetti@unine.ch" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Lionel Grossriederb ; lionel.grossrieder@unil.ch" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Olivier Ribauxb ; olivier.ribaux@unil.ch" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Kilian Stoffela ; kilian.stoffel@unine.ch" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae
- 关键词：Change points detection ; Qualitative description of data ; Time series analysis ; Fuzzy logic ; Crime analysis
- 刊名：Applied Soft Computing
- 出版年：2016
- 出版时间：March 2016
- 年：2016
- 卷：40
- 期：Complete
- 页码：441-454
- 全文大小：2859 K
文摘
- •
The method provides both a dynamic detection and fuzzy querying of change points in crime-related time series.
- •
The approach is flexible and intuitive.
- •
Geometric properties expressed with linguistic variables can be used to query change points.
- •
The method can be used on-line and does not rely on training sets.
- •
The computational costs are very low and depend only on the degree of the approximating polynomials.