Air Quality Modelling by Kohonen's Self-organizing F eature Maps and LVQ Neural Networks

摘要: The paper presents a design of parameters for air quality modelling and the classification districts into classes according to their pollution. Further, it model design, data pre-processing, designs various structures Kohonen's Self-organizing Feature Maps (unsupervised methods), clustering by K-means algorithm Learning Vector Quantization neural networks (supervised methods). Therefore, generates well-separated clusters has good generalization ability as well.