Neural network modelling method

Abstract

Based on principle of minimization risk of configuration, combined with cooperative collaboration evolution algorithm, and learning network structure of neural network and connection weight value, the invention obtains optimal compromise between network structure and connection weight value finally. The method includes three basic steps: data processing, network learning and network estimated forecast. Configuring network and learning connection weight value are carried out at same time in the invention so as to better solve practical problems existed in traditional neural network learning: correlation between result and initial value, slow convergence rate, easy to run to local minimum value as well as derivable error function needed and over learning. The invention raises learning capability and generalization capability of network, applicable to intelligent diagnosing heart disease, fault diagnosis in industries, stock and goods price forecasting etc.

Description

technical field

[0001] The invention relates to a modeling method, in particular to a neural network modeling method. It belongs to the technical field of intelligent information processing. Background technique

[0002] Traditional mathematical modeling methods, including mechanism modeling, multivariate statistical methods, Kalman filter methods, model-based regression methods, etc., have achieved certain results in application. However, with the increasing complexity of the problems to be solved, it is difficult to accurately describe them by traditional mathematical methods, and the selection of evaluation indicators is not appropriate, and the actual results are generally not very satisfactory. Therefore, a neural network-based modeling method was proposed, which improved the performance of the model to a considerable extent. However, since the connection weight learning usually uses an algorithm that is essentially a gradient descent, and the network structure depends...

Images