Feature selection method using neural network
作者： V. Onnia , M. Tico , J. Saarinen
关键词: Feature extraction 、 Artificial neural network 、 k-nearest neighbors algorithm 、 Feedforward neural network 、 Feature vector 、 Feature (computer vision) 、 Feature selection 、 Time delay neural network 、 Artificial intelligence 、 Data mining 、 Feature (machine learning) 、 Dimensionality reduction 、 Deep learning 、 Probabilistic neural network 、 Computer science 、 Linear classifier 、 Pattern recognition 、 Feature learning
摘要: Feature selection is an important part of most learning algorithms. used to select the relevant features from data. By selecting only data, higher predictive accuracy can be achieved and computational load classification system reduced. A simple method for feature using feedforward neural networks presented. The starts by one input neuron adds at time until wanted has been or all attributes have chosen. algorithm also with other methods. Test results are given they promising. Our reduces size space significantly improves accuracy. Tests were performed on commonly databases. Average accuracy, when selected features, was between 79% 100% depending dataset.
sci-hub.se 参考文章(9)
William F. Punch, Richard J. Enbody, Paul D. Hovland, Min Pei, Erik D. Goodman, Lai Chia-Shun, Further Research on Feature Selection and Classification Using Genetic Algorithms international conference on genetic algorithms. pp. 557- 564 ,(1993)
A.H.S. Solberg, A.K. Jain, A study of the invariance properties of textural features in SAR images international geoscience and remote sensing symposium. ,vol. 1, pp. 670- 672 ,(1995) , 10.1109/IGARSS.1995.520488
G. Castellano, A.M. Fanelli, Feature selection: a neural approach international joint conference on neural network. ,vol. 5, pp. 3156- 3160 ,(1999) , 10.1109/IJCNN.1999.836157
Kyusik Chung, Jongmin Yoon, Performance comparison of several feature selection methods based on node pruning in handwritten character recognition international conference on document analysis and recognition. ,vol. 1, pp. 11- 15 ,(1997) , 10.1109/ICDAR.1997.619805
Anil K. Jain, Aditya Vailaya, IMAGE RETRIEVAL USING COLOR AND SHAPE Pattern Recognition. ,vol. 29, pp. 1233- 1244 ,(1996) , 10.1016/0031-3203(95)00160-3
M.F. Redondo, C.H. Espinosa, A comparison among feature selection methods based on trained networks Neural Networks for Signal Processing IX: Proceedings of the 1999 IEEE Signal Processing Society Workshop (Cat. No.98TH8468). pp. 205- 214 ,(1999) , 10.1109/NNSP.1999.788139
Rudy Setiono, Huan Liu, Neural-network feature selector IEEE Transactions on Neural Networks. ,vol. 8, pp. 654- 662 ,(1997) , 10.1109/72.572104
R. Battiti, Using mutual information for selecting features in supervised neural net learning IEEE Transactions on Neural Networks. ,vol. 5, pp. 537- 550 ,(1994) , 10.1109/72.298224
M.F. Redondo, C.H. Espinosa, Input selection by multilayer feedforward trained networks international joint conference on neural network. ,vol. 3, pp. 1834- 1839 ,(1999) , 10.1109/IJCNN.1999.832658