Deep neural network method based on principal component analysis and clustering analysis

Abstract

The invention relates to a deep neural network method based on principal component analysis and clustering analysis, and the method specifically comprises the following steps: S1, dividing label datainto training data and test data, and the training data being used for the training learning of a model; S2, performing feature dimension reduction on all initial data features of the training data byutilizing PCA, and extracting a new principal component; S3, performing K-Means clustering analysis on all the training samples according to the principal components extracted by PCA; and S4, forminga single-layer neural network for training by taking an upper-layer data result as an input and combining the tag obtained by clustering, thereby obtaining a network weight parameter. According to the deep neural network method based on principal component analysis and clustering analysis provided by the invention, a statistical feature learning method and the neural network are combined and applied, a training mode of a traditional multilayer neural network is optimized in a training process, and a better test effect is obtained for learning of a common deep neural network.

Description

technical field [0001] The invention relates to the technical fields of machine learning and artificial intelligence, in particular to a deep neural network method based on principal component analysis and cluster analysis. Background technique [0002] In recent years, artificial intelligence technology has attracted widespread attention in both industry and academia, and machine learning methods play a central role in the field of artificial intelligence and have been applied in many fields, such as biometric sequences, natural language processing, computer Vision, image recognition, financial market analysis and many other fields have developed rapidly. Among them, the performance of deep learning in many fields is remarkable. As a classic deep learning method, Deep Belief Network (DBN) has high research significance in feature extraction and classification learning. [0003] The main idea of ​​Deep Belief Network (DBN) is divided into two stages of learning. The first ...

AI Technical Summary

Problems solved by technology

[0004] In order to overcome the above-mentioned shortcomings of the prior art, the present invention proposes a deep neural network method based on principal component analysis and cluster analysis to solve the problem of poor learning effect of the existing common deep neural network

Images