Data classification method and device and readable storage medium

Abstract

The invention provides a data classification method and device and a readable storage medium, and relates to the field of data processing. The method comprises the steps of obtaining to-be-classified data; linear features of the to-be-classified data are extracted; inputting the linear features into an omni-directional self-attention neural network to obtain a feature matrix of the to-be-classified data; performing normalization processing on the feature matrix to obtain a prediction probability that the to-be-classified data belongs to each category label; wherein the to-be-classified data comprises n sub-data, and n is an integer and is greater than or equal to 2; and the omni-directional self-attention neural network obtains a feature matrix of the to-be-classified data by using context features among the sub-data, correlation features among the sub-data and the importance degree of each sub-data to predict a category label of the to-be-classified data. According to the invention, the problem of low data classification prediction accuracy caused by poor universality of a data classification model in the prior art can be solved.

Description

Technical field[0001]Embodiments of the present invention relate to the field of data processing, and more particularly to a data classification method, a device, and a readable storage medium.Background technique[0002]With the successful application and continuous development of deep learning models in other fields, more and more research began to use music's sound spectrum as an input of deep learning models. Common depth learning algorithm networks have problems: such as RNN (Recurrent Neural Network, Cycle Neural Network) can only memorize partial sequences, there is a risk of gradient disappearance or gradient explosion; CNN (Convolutional Neural Networks, Convolution Neural Network The poolization operation will lose a lot of information characteristics; LSTM (Long Short-Term Memory, long short memory network), GRU (GATED Recurrent Unit, Gate Control Circulation Unit) No recognition for long-distance interdependent features; self-attention The mechanism (Self-attention) cannot...

AI Technical Summary

Problems solved by technology

[0004] The embodiment of the present invention provides a data classification method, device and readable storage medium, which solves the problem of low data classification prediction accuracy caused by the poor versatility of the data classification model in the prior art

Images