Working method based on recurrent neural network stock market analysis model

Abstract

The invention discloses a working method based on a recurrent neural network stock market analysis model. The method comprises: collecting stock market data; analyzing and arranging the data; extracting the characteristic values of the stock market data; defining the characteristic values as X, and inputting learning data X = {X1, X2,..., Xtau} according to a sequence; establishing a recurrent neural network for stock market data analysis, wherein the recurrent unit of the recurrent neural network is h<(t)>=f(s<(t-1)>,X<(t)>,theta); giving stock market learning data and classification labels:X={X1, X2,..., Xtau}, y belongs to {1,...,C}; and combining the recurrent neural network with a convolutional neural network. According to the invention, characteristic value extraction is carried outon stock market data, a recurrent neural network for stock market data analysis is established, analysis is carried out through the established recurrent neural network, then stock market data analysis is carried out through the given learning data and classification labels of a stock market, and finally analysis is carried out through combination of the recurrent neural network and a convolutional neural network, so that the analysis efficiency and the analysis accuracy are improved.

Description

technical field [0001] The invention relates to the field of cyclic neural network, in particular to a working method based on a cyclic neural network stock market analysis model. Background technique [0002] The recurrent neural network is a kind of recursive neural network that takes sequence data as input, recurses in the evolution direction of the sequence, and all nodes (recurrent units) are connected in a chain. [0003] Spanish neurobiologists discovered that the anatomy of the cerebral cortex allows stimuli to circulate in neural circuits, leading to the reverberation loop hypothesis. This hypothesis was endorsed in a series of studies at the same time, and is considered to be the reason why living things have short-term memory. Subsequent further studies in neurobiology found that the excitation and inhibition of the echo loop are regulated by the alpha rhythm of the brain and form a circular feedback system in the alpha motor nerve. In the 1970s and 1980s, vario...

AI Technical Summary

Problems solved by technology

In the 1970s and 1980s, various mathematical models established for simulating cyclic feedback systems laid the foundation for the development of cyclic neural networks. When working on cyclic neural network stock market analysis models, the existing working methods are not clear. , resulting in large deviations in the results of the model during analysis, so a working method for the cyclic neural network stock market analysis model is needed