Unlock instant, AI-driven research and patent intelligence for your innovation.

An aspect attitude mining method integrating an emoji database and a subject model

A technology of emoticons and topic models, applied in the field of artificial intelligence, can solve problems such as ignoring emoticons, misjudgment of praise and criticism attitudes, etc.

Pending Publication Date: 2019-01-29
NANTONG UNIVERSITY +1
View PDF0 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Just relying on the text content, but ignoring the emotion of the emoji, is likely to produce a misjudgment of the comment's positive or negative attitude

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • An aspect attitude mining method integrating an emoji database and a subject model
  • An aspect attitude mining method integrating an emoji database and a subject model
  • An aspect attitude mining method integrating an emoji database and a subject model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0047] The technical solution of the present application will be further described below in conjunction with the accompanying drawings.

[0048] Such as figure 1 In this embodiment, the method for mining praise and derogation attitudes of aspects that integrate emoticon databases and topic models includes the following steps:

[0049] Step 1) Preprocess the original microblog content and comments, calculate the similarity matrix between the aspects in the original microblog and comments, and obtain the explicit aspect set related to the original microblog aspects and the original A set of implicit facets that are not related to Weibo facets. Aspect is a professional term in opinion mining technology, which refers to the most fine-grained evaluation object in reviews. This embodiment introduces the concept of aspect views into Sina Weibo comments, extracts nouns and noun phrases in comment sentences as aspects of the comments, and extracts adjectives and adjective phrases, ve...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses an aspect attitude mining method integrating an emoji database and a subject model, which comprises the following steps: firstly, calculating similarity matrix between nouns inoriginal micro-blog and comments, and obtaining explicit aspect set and implicit aspect set related to the aspect mentioned in original micro-blog by using spectral clustering algorithm; secondly, constructing an aspect attitude mining model which combines emoticons and theme model, and estimate the parameters. Finally, combined with the viewpoint mining model and explicit and implicit aspects ofthe micro-blog comments for the attitude analysis, get the attitude of each user comments on the original micro-blog content. The invention applies the fusion of subject model and emoji database to mining viewpoints and analyzing attitude of praise and disapproval of user comments under original content of micro-blog. At the same time, combining the explicit aspects related to the original micro-blog and the implicit aspects existing in the comments, in order to better obtain the comment users' attitude towards the original micro-blog, and to improve the judgment of the overall set of comments' attitude tendency.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to a method for mining praise and derogation attitudes of aspects, viewpoints, and themes that integrate emoticon databases and theme models. Background technique [0002] Weibo 2.0 is one of the most popular applications. It gives users more freedom and faster ways to communicate information, express opinions, and record emotions, making the daily update of Sina Weibo a huge amount of information data, increasing the content of Weibo Difficulty of aspect and aspect-based viewpoint mining. In recent years, topic models based on LDA can effectively avoid the shortcomings of traditional unsupervised learning methods relying on sentiment dictionaries, and achieve better mining results. [0003] In Sina Weibo, there are many user comments under each original Weibo. Some users expressed their tendency to praise or criticize the original Weibo content, and some users wro...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06F16/35G06F16/2458G06F17/27
CPCG06F40/284
Inventor 张士兵张茜张晓格
Owner NANTONG UNIVERSITY