CAJ | 학술논문

该文利用竞争型神经网络来对空气质量进行分类。竞争型神经网络模拟生物神经网络中的神经元之间的兴奋、抑制与竞争的机制,进行网络的学习与训练。竞争型神经网络的输出层神经元能够通过竞争以确定胜者,胜者将指出哪一种原型模式最能代表输入模式。将不同空气质量等级下的各空气指标作为原型模式,通过输入样本模式,利用竞争网络的竞争特点得到胜者,以此得出空气质量等级。
해문이용경쟁형신경망락래대공기질량진행분류。경쟁형신경망락모의생물신경망락중적신경원지간적흥강、억제여경쟁적궤제,진행망락적학습여훈련。경쟁형신경망락적수출층신경원능구통과경쟁이학정성자,성자장지출나일충원형모식최능대표수입모식。장불동공기질량등급하적각공기지표작위원형모식,통과수입양본모식,이용경쟁망락적경쟁특점득도성자,이차득출공기질량등급。
Air quality plays an important role in human＇s life.We wish to know the air quality grade of living area and measure the effect of the environment on our body with a simple and effective method to classify air quality.This paper uses competitive neural network to classify air quality,to simulate the excitation,inhibition and competitive mechanism existing in biological neural networks for undertaking network learning and training.The neuron in the output layer through competition to determine the winner,who will demonstrate which prototype model can represent the input mode.Air targets can be used in the different level of air quality as prototype model and input sample mode.Then,we the competitive characteristics of the competitive neural network can be taken advantage of so as to produce the winner and get the level of air quality.