Sentiment classification method and device

Abstract

The invention provides a sentiment classification method and device. The method comprises the following steps: inputting various sentiment data of a target object into corresponding feature extraction modules in a classification model, and outputting first feature vectors of the various sentiment data; inputting the first feature vectors of all the sentiment data into a feature fusion module in the classification model, and outputting second feature vectors after the first feature vectors of all the sentiment data are fused; forming the first feature vectors and the second feature vectors of the various sentiment data into input vectors, inputting the input vectors into corresponding attention mechanism modules in the classification model, and outputting third feature vectors of the various sentiment data; and splicing the third feature vectors of all the sentiment data, inputting the spliced third feature vectors into a classification module of the classification model, and outputting a sentiment category of the target object. According to the invention, sentiments of the target object are classified by fusing the various sentiment data of the target object, and the accuracy of a classification result is effectively improved.

Description

technical field [0001] The invention relates to the technical field of data processing, in particular to an emotion classification method and device. Background technique [0002] Sentiment classification is widely used in medicine, policing, and recruitment. Accurate emotion classification can effectively assist doctors to formulate corresponding treatment plans for patients. [0003] In the prior art, a single data information is usually used to classify emotions, such as using a convolutional neural network to perform feature extraction on the spectrogram of the speech data, and then classifying the speech data for emotion; or using Bert (Bidirectional Encoder Representation From Transformer, two-way encoder) model and Transformer model etc. perform feature extraction on text data, and perform sentiment classification on text data. [0004] However, since there are many factors that affect emotion evaluation, it is difficult to accurately obtain emotion-related features...

AI Technical Summary

Problems solved by technology

[0005] The present invention provides an emotion classification method and device, which are used to solve the defect in the prior art that it is difficult to accurately obtain emotion-related features only based on text data or voice data, resulting in inaccurate emotion classification results, and realize accurate classification of emotions

Images