Method for extracting emotional triggers

Abstract

The invention discloses a method for extracting emotional triggers. The method includes steps of a corpus processing procedure: dividing each section into a plurality of clauses according to punctuation marks and labeling each clause according to relations among the clause and the emotional triggers; a feature extracting and training procedure: extracting features of the clauses of the training corpus and acquiring a classifier for all the clauses in the training corpus by means of training; a emotional trigger extracting procedure: dividing the sections to be extracted into a plurality of clauses according to punctuation marks, extracting features of each clause, classifying each clause by the trained classifier, judging whether each clause is an emotional trigger or not and outputting the clauses which are the emotion triggers. The method for extracting the emotional triggers has the advantages that the method is beneficial to automatically processing and analyzing events relevant to emotions, further can be applied to other tasks such as monitoring sudden events and abstracting emotion, and is high in practical value.

Description

technical field [0001] The present invention relates to the technical field of natural language processing and pattern recognition, in particular to a method for extracting emotion-triggered events. Background technique [0002] Text sentiment analysis is an important research direction in natural language processing (NLP) research. This direction aims to study how to automatically analyze the emotions expressed in text and information related to emotions. At present, mainstream sentiment analysis research mainly focuses on emotion classification, which aims to automatically classify the emotions expressed in text (such as: happy, sad, surprised, etc.). However, the emotional information that this task focuses on is only the category of emotions, and it is still relatively shallow emotional information. In order to better understand the emotional information expressed in the text, it is urgent to conduct further exploration and research on the deeper information about the ...

AI Technical Summary

Problems solved by technology

[0004] Although emotion analysis research has been carried out for many years, most of the research focuses on emotion recognition and classification, and research on the analysis of emotional trigger events has not yet been carried out.

Images