Social network text sentiment analysis method

• Main Authors: ZHANG MINGLING, FU WEI
• Format: Patent
• Language: Chinese; English
• Published: 09.04.2024
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR

The invention belongs to the technical field of natural language processing, and particularly relates to a social network text sentiment analysis method which comprises the following steps: acquiring network text information, and preprocessing the text information; performing word vectorization on the preprocessed text data by adopting a pre-trained language model ERNIE to obtain a word vector sequence; a Bi-GRU network layer is adopted to carry out forward and backward information summarization and combination on the word vector sequence, and a final Bi-GRU model implicit vector is obtained; multiplying the implicit vector of each word by the word weight calculated through the self-attention layer to obtain a weighted final vector; inputting the weighted final vector into the trained text sentiment analysis model to obtain a sentiment classification result of the text; according to the method, an innovative method combining ERNIE and Bi-GRU neural networks is adopted, so that the effect of social network tex