Analysis of Time Series Novelty Detection Strategies for Synthetic and Real Data

被引：0
作者：
André Paoliello Modenesi
Antônio Pádua Braga
机构：
[1] Federal University of Minas Gerais,Department of Electronic Engineering
来源：
Neural Processing Letters | 2009年 / 30卷
关键词：
Novelty detection; Neural networks; Signal processing; Time series; Principal component analysis; Reduced coulomb energy networks; Density probability estimation;
D O I：
暂无
中图分类号：
学科分类号：
摘要：
Novelty detection is inspired by animal behavior and can be applied to a variety of practical situations. By automatically bringing attention to anomalous events, without the need of previously defining those events, it can perform certain tasks that would otherwise require human scrutiny. This paper reviews some approaches to the problem when applied to the time series analysis problem.
引用
收藏
相关论文
共 21 条
[1]  
Bishop C(1994)Novelty detection and neural network validation IEEE Proc Vis, Image Signal Process 141 217-222
[2]  
Chen T(1998)A unified algorithm for principal and minor components extraction Neural Netw 11 385-390
[3]  
Amari S-I(2005)Using AUC and accuracy in evaluating learning algorithms IEEE Trans Knowl Data Eng 17 299-310
[4]  
Lin Q(1972)Correlation matrix memories IEEE Trans Comput 21 353-359
[5]  
Huang J(1976)Fast adaptive formation of orthogonalizing filters and associative memory in recurrent networks of neuron-like elements Biol Cybern 25 85-95
[6]  
Ling CX(2005)SOM-based novelty detection using novel data Intelligent Data Engineering and Automated Learning—IDEAL 3578 359-366
[7]  
Kohonen T(2003)Time series novelty detection using one-class support vector machines Proc Int Jt Conf Neural Netw 3 1741-1745
[8]  
Kohonen T(2003)Novelty etection in learning systems Neural Comput Surv 3 157-195
[9]  
Oja E(1992)Principal components, minor components, and linear neural networks Neural Netw 5 927-935
[10]  
Lee H-J(1962)On the estimation of a probability density function and mode Ann Math Stat 33 1065-1076
123