Novelty detection with constructive probabilistic neural networks

详细信息	查看全文 | 推荐本文 |

• 作者：Adriano L.I. Oliveira ; Flavio R.G. Costa ; Clovis O.S. Filho
• 关键词：Novelty detection ; Anomaly detection ; Probabilistic neural networks ; Constructive training ; Dynamic decay adjustment
• 刊名：Neurocomputing
• 出版年：2008
• 期刊代码：95_09252312
• 类别：cp
• 出版时间：January 2008
• 卷：71
• 期：4-6
• 页码：1046-1053
• 文件大小：601 K

摘要

This paper investigates the use of probabilistic neural networks trained with the dynamic decay adjustment algorithm (PNN–DDA) for novelty detection tasks. PNN–DDA is a fast, constructive neural model originally developed and investigated for standard classification tasks. The training algorithm is controlled by two parameters, θ⁺ and θ^-. Simulations employing four data sets from the UCI machine learning repository are reported. The results show that parameter θ^- considerably influences the performance of PNN–DDA for novelty detection, and furthermore, that PNN–DDA achieves performance comparable to NNDD with the advantage of producing much smaller classifiers.