Modular Neural Network Rule Extraction Technique in Application to Country Stock Cooperate Governance Structure

摘要

Neural networks learn knowledge from data. For a monolithic structure, this knowledge can be easily used but not isolated. The many degrees of freedom while learning make ruler extraction a computationally intensive process as the representation is nor unique. Based on the technology of modular neural network data mining, this paper applied modular neural network ruler extraction to the data mining of country stock cooperate governance structure. Meanwhile, it investigated the relationship among gerentocratic constitutes of country stock cooperate, farmers' educational level, labor force and corporation performance of country stock cooperate.