基于概念格的BP神经网络算法及仿真

A BP Neural Network Algorithm Based on Concept Lattice and Its Matlab Simulation

针对在研究人工智技术能领域中,利用概念格和BP神经网络的各自优势,提出了一种基于概念格的BP神经网络算法。算法首先利用概念格的理论对样本数据进行属性约简,提取其中关键要素作为BP神经网络的训练样本,用简化的训练样本对BP神经网络进行训练,建立优化的基于概念格的BP神经网络算法进行仿真实验。仿真结果表明,基于概念格的BP神经网络算法能简化BP神经网络的训练样本,优化BP神经网络,提高了系统的学习效率和精度。证明方法是有效可行的,具有理论意义和实用价值。

Based on the advantages of concept lattices and BP neural networks,this paper presents a concept lattice -based BP neural network algorithm.In terms of Concept Lattice theory,attribute reduction of concept lattice is carried on and then the key elements are extracted,which can be used as input of BP neural network.Furthermore, the concept lattice - based BP neural network algorithm can be set up after the sample training of BP neural network. Finally,the results of simulative experiment show that the algorithm can simplify the training samples of BP neural network,optimize the structure of BP neural network and also enhance the study efficiency and precision of the system. So,this novel method is effective and feasible,what＇s more,the theoretical significance and practical value are also outstanding.