Detecting method and device for personal abnormal emotion based on personal microblog

Abstract

The invention provides a detecting method and device for a personal abnormal emotion based on a personal microblog. The method comprises that steps that first preset several microblog text data of a designated user in a preset time quantum is acquired; a support vector machine is utilized to perform emotion recognition and annotation on the first preset several microblog text data, and second preset several emotions are obtained; according to a preset time unit, emotion distribution statistics is carried out on the microblog text data of the second preset several emotions, and a multi-dimension data set of the designated user is obtained; multivariate normal distribution is adopted to carry out fitting on the multi-dimension data set in the preset time quantum, and a normal distribution curve of corresponding emotions is obtained; a joint probability density value of the multi-dimension data set is counted; based on the normal distribution curve and / or the joint probability density value, the abnormal emotions of the designated user are determined. Therefore, qualitative and quantitative analysis is carried out on the abnormal emotions of the designated user by acquiring the normaldistribution curve and joint probability density value, and the detecting accuracy and reliability can be improved.

Description

technical field [0001] The invention relates to the technical field of data processing, in particular to a method and device for detecting personal abnormal emotions based on personal microblogs. Background technique [0002] At present, the personal abnormal emotion detection schemes based on personal Weibo mainly include: [0003] Option 1, abnormal text dictionary. In this scheme, words such as sadness, anger, and depression are collected into a dictionary, and then emotional word matching is performed to detect abnormal words in individual Weibo text data, and then the abnormal text and users with abnormal emotions are judged. It can be seen that the detection method of the abnormal text dictionary is relatively simple, and the accuracy of the detection results is high. In this way, the detection result can only detect abnormal emotions, but cannot detect the abnormal emotions implied by the user. And some users choose to use Weibo as a platform to vent or their perso...

AI Technical Summary

Problems solved by technology

[0005] In view of the defects in the prior art, the present invention provides a personal abnormal emotion detection method and device based on personal Weibo, which is used to solve the problem that the abnormal dictionary method or negative text abnormal judgment method in the related art cannot accurately detect the abnormal emotion of the user question

Images