Deep learning problem classification method and system combining multi-level attention mechanism

Abstract

The invention discloses a deep learning problem classification method combining a multi-level attention mechanism, and relates to the field of information processing technology. The deep learning problem classification method comprises the following steps: constructing an interrogative word vector set, wherein the interrogative word vector set includes interrogative word vectors and common word vectors including interrogative word information; extracting window mapping of interrogative sentences by convolution operation according to the interrogative word vector set; extracting sequential characteristics of the interrogative sentences according to the window mapping; and classifying the interrogative sentences according to the sequential characteristics. Through adoption of the method, semantic information of interrogative words in the interrogative sentences is enhanced. Moreover, a convolutional neural network and a long-term and short-term memory model are fused through the attention mechanism in deep learning, thereby effectively increasing the accuracy of problem classification.

Description

technical field [0001] The present invention relates to the technical field of information processing, in particular to a deep learning problem classification method and system combined with a multi-level attention mechanism. Background technique [0002] There are two main types of traditional problem classification methods, one is rule-based methods, and the other is machine learning-based methods. The rule-based method mainly builds a large number of rule bases manually, and adopts the method of rule matching to realize problem classification. The method based on machine learning regards question classification as a supervised learning task, expands the question text by manually formulating features and introducing external corpus, and uses classic support vector machines, naive Bayesian and other machine learning algorithms to classify . [0003] In recent years, due to the maturity of deep learning technology, the problem classification method based on machine learnin...

AI Technical Summary

Problems solved by technology

[0005] Second: The existing problem classification methods cannot well integrate the feature extraction capabilities of the two models

Images