Multi-modal emotion analysis method for emoji package of social platform

Abstract

The invention provides a multi-modal emotion analysis method for an emoji package of a social platform, and the method comprises the following steps: S1, crawling an emoji package image from the social platform through a crawler tool, carrying out the emotion labeling of the emoji package image, and then carrying out the preprocessing; S2, obtaining semantic information of the emoji package images crawled in the step S1, and obtaining text information feature vector representation corresponding to each emoji package image; S3, obtaining visual features of the emoji package imageS crawled in the step S1, and obtaining visual feature vector representation corresponding to each emoji package image; S4, performing multi-modal fusion on the text information feature vector representation and the visual feature vector representation to obtain multi-modal fusion feature vector representation; and S5, enabling the multi-modal fusion feature vector to represent that an emotion recognition result is obtained through a classifier, selecting the emotion recognition result with the highest confidence coefficient as a predicted emotion, and introducing text semantic features of the image, so that implicit semantic information in the expression package can be better captured.

Description

technical field [0001] The present invention relates to the field of artificial intelligence, and more specifically, relates to a multimodal sentiment analysis method for emoticons on social platforms. Background technique [0002] Sentiment analysis is an important task in the field of artificial intelligence and one of the research hotspots in the field of natural language processing. Sentiment analysis mines people's real opinions and emotions by processing information data. For example, judging whether the information reflects positive emotions or negative emotions, or dividing emotions into several different rating levels from very satisfied to very dissatisfied according to the scale. The task of sentiment analysis was originally widely studied in the field of natural language processing, which refers to the sentiment recognition of text content, that is, sentiment classification. Early sentiment analysis only focused on plain text information, using traditional mach...

AI Technical Summary

Problems solved by technology

[0009] The present invention provides a multi-modal emotion analysis method for emoticons on social platforms, which solves the lack of an emotional analysis method for emoticons on social platforms in the existing research on emotional analysis. Solved the problem of the text content on the emoticon pack picture

Images