Improving the MLP Learning by Using a Method to Calculate the Initial Weights of the Network Based on the Quality of Similarity Measure.
Yaima Filiberto CabreraRafael Bello PérezYailé Caballero MotaGonzalo Ramos-JiménezPublished in: MICAI (2) (2011)
Keyphrases
- similarity measure
- high accuracy
- high quality
- mutual information
- prior knowledge
- learning algorithm
- clustering method
- unsupervised learning
- learning process
- significant improvement
- linear combination
- pairwise
- objective function
- support vector machine
- preprocessing
- detection method
- quality measures
- initial set
- neural network
- neural nets
- segmentation method
- support vector machine svm
- supervised learning
- training set