孤立点检测改进径向基神经网络动态预测模型
|
摘要
提出一种基于凝聚层次聚类消除孤立点的新方法,借助聚类树识别孤立点。去除孤立点后,利用RBF网络建立动态预测模型,实验结果表明,网络的训练和泛化性能较消除孤立点前有明显提高。说明凝聚层次聚类方法用在孤立点检测方面是有效可行的,消除孤立点后建立的模型收敛速度快,泛化能力更优。
Propose a new agglomerative hierarchical clustering based method to eliminate outliers,with clustering tree to identify outliers.After removing the outliers,build a dynamic prediction model by RBF network,and the experimental results show that the training and generalization performance are markedly improved,which means the agglomerative hierarchical clustering method is effective and workable for outlier detection.After the elimination of outliers,the model shows faster converging speed and higher generalization ability.
Propose a new agglomerative hierarchical clustering based method to eliminate outliers,with clustering tree to identify outliers.After removing the outliers,build a dynamic prediction model by RBF network,and the experimental results show that the training and generalization performance are markedly improved,which means the agglomerative hierarchical clustering method is effective and workable for outlier detection.After the elimination of outliers,the model shows faster converging speed and higher generalization ability.
引文
[1]Whitehead B A.Cooperative-competitive genetic evolution of radial basis function centers and widths for time series prediction[J].IEEE Transactions on Neural Networks,1996,7(4):869-880.
[2]Lu Y W,Sundararajan N,Saratchandran P.Performance evaluation of sequential mininal radial basis function(RBF)neural network learning algorithm[J].IEEE Transactions on Neural Networks,1998,9(6):308-317.
[3]Han Jia-wei,Kamber M.Data mining:Concepts and techniques[M].2nd ed.Beijing:China Machine Press,2006.
[4]Knorr E,Ng R.Algorithms for mining distance-based outliers in large data sets[C]//Gupta A,Shmueli O,Widom J.Proc of the VLDB Conf.New York:Morgan Kaufmann Publishers,1998:392-403.
[5]Ramaswamy S,Rastogi R,Shim K.Efficient algorithms for mining outliers from large data sets[C]//Chen W D,Naughton J F,Bern-stein P A.Proc of the ACM SIGMOD Conf.Dallas:ACM Press,2000:427-438.
[6]Breunig M M,Kriegel H P,Ng R,et al.LOF:Identifying density-based local outliers[C]//Chen W D,Naughton J F,Bernstein P A.Proc of the ACM SIGMOD Conf.Dallas:ACM Press,2000:94-104.
[7]Arning A,Agrawal R,Raghavan P.A linear method for deviation detection in large databases[C]//Simoudis E,Han J W,Fayyad U M.Proc of the KDD Conf.Portland:AAAI Press,1996:164-169.
[8]Knorr E M,Ng R T,Tucakov V.Distance-based outliers:Algorithms and applications[J].VLDB Journal:Very Large Databases,2000:237-253.
[9]石剑飞,闫怀志,牛占云.基于凝聚的层次聚类算法的改进[J].北京理工大学学报,2008,28(1):66-69.
[10]王海霞,周晓山.预测人员震害损失的神经网络模型[J].世界地震工程,2007,23(4):194-198.
[2]Lu Y W,Sundararajan N,Saratchandran P.Performance evaluation of sequential mininal radial basis function(RBF)neural network learning algorithm[J].IEEE Transactions on Neural Networks,1998,9(6):308-317.
[3]Han Jia-wei,Kamber M.Data mining:Concepts and techniques[M].2nd ed.Beijing:China Machine Press,2006.
[4]Knorr E,Ng R.Algorithms for mining distance-based outliers in large data sets[C]//Gupta A,Shmueli O,Widom J.Proc of the VLDB Conf.New York:Morgan Kaufmann Publishers,1998:392-403.
[5]Ramaswamy S,Rastogi R,Shim K.Efficient algorithms for mining outliers from large data sets[C]//Chen W D,Naughton J F,Bern-stein P A.Proc of the ACM SIGMOD Conf.Dallas:ACM Press,2000:427-438.
[6]Breunig M M,Kriegel H P,Ng R,et al.LOF:Identifying density-based local outliers[C]//Chen W D,Naughton J F,Bernstein P A.Proc of the ACM SIGMOD Conf.Dallas:ACM Press,2000:94-104.
[7]Arning A,Agrawal R,Raghavan P.A linear method for deviation detection in large databases[C]//Simoudis E,Han J W,Fayyad U M.Proc of the KDD Conf.Portland:AAAI Press,1996:164-169.
[8]Knorr E M,Ng R T,Tucakov V.Distance-based outliers:Algorithms and applications[J].VLDB Journal:Very Large Databases,2000:237-253.
[9]石剑飞,闫怀志,牛占云.基于凝聚的层次聚类算法的改进[J].北京理工大学学报,2008,28(1):66-69.
[10]王海霞,周晓山.预测人员震害损失的神经网络模型[J].世界地震工程,2007,23(4):194-198.
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心