Short-text emotion analysis method and device based on concept based on text emotion

A short text and text technology, applied in the field of information processing, can solve the problem of low accuracy of sentiment analysis and achieve the effect of improving accuracy

Inactive Publication Date: 2018-06-01
EZHOU INST OF IND TECH HUAZHONG UNIV OF SCI & TECH +1
View PDF5 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The embodiment of the present invention provides a short text sentiment analysis method and device based on concepts and text sentiment, which solves the problem that the analysis method in the prior art is affected by the size of the corpus and the feature extraction process, resulting in low accuracy of sentiment analysis. question

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
  • Short-text emotion analysis method and device based on concept based on text emotion
  • Short-text emotion analysis method and device based on concept based on text emotion
  • Short-text emotion analysis method and device based on concept based on text emotion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0061] figure 1 It is a schematic flowchart of a short text sentiment analysis method based on concepts and text sentiment in an embodiment of the present invention. Such as figure 1 As shown, the method includes:

[0062] Step 110: Perform data preprocessing on the short text;

[0063] Further, the data preprocessing of the short text also includes: performing word segmentation processing on the short text; performing part-of-speech tagging on the short text; performing dependency syntax analysis on the short text to obtain binary metaphrase.

[0064] Specifically, preprocessing the short text mainly includes the following aspects:

[0065] Perform word segmentation processing on the text data, divide the short text into text-word sequences, and mark the divided words according to their semantics. Perform dependency syntactic analysis on the text-word sequence, and extract the binary phrases in the sentence according to the syntactic structure of the sentences in the tex...

specific Embodiment approach 1

[0102] The determining the final emotional polarity category according to the text concept feature vector and the text emotion feature vector also includes: performing feature fusion according to the text concept feature vector and the text emotion feature vector to obtain comprehensive text features; According to the integrated text features, the final emotional polarity category is determined.

[0103] Specifically, the two features extracted in the second section: the concept feature vector and the text emotion feature vector are fused to obtain a richer text feature that not only retains the text semantic information but also contains the sentence emotional features, and remembers the final text feature The vector is w(d), then w(d) is expressed as the concatenation of the concept feature vector and the text emotion feature:

[0104] w(d)=concatenate(w c (d),w s (d))

[0105] where w c (d) and w s (d) represent the concept feature vector of the text and the emotional ...

specific Embodiment approach 2

[0106] According to the text concept feature vector and the text emotion feature vector, determining the final emotion polarity category also includes: training an SVM classifier according to the text concept feature vector to obtain a text concept feature classification model; according to the text emotion feature The vector training SVM classifier obtains the text emotion feature classification model; according to the text concept feature classification model and the text emotion feature classification model, the category probability distribution of unknown samples is weighted to determine the final emotion polarity category.

[0107] Specifically, an SVM classifier is trained on the text concept feature vector and the text emotion feature vector to obtain two classification models of feature vectors: a concept feature classification model and a text emotion feature classification model. The final emotional polarity category is obtained by weighting the category probability d...

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 provides a short-text emotion analysis method and device based on concept and text emotion, and relates to the technical field of information processing techniques. The method includes the steps of carrying out data preprocessing on a short-text; according to a dependence grammar relation, obtaining text concept features; according to a constructed emotion dictionary and part-of-speech annotations, carrying out emotion annotation and obtaining text emotion features; according to the text concept features and the text emotion features, obtaining text concept feature vectors and text emotion feature vectors; according to the text concept feature vectors and the text emotion feature vectors, determining a final emotion polarity type. The method and device solve the technical problems that since analysis methods in the prior art are influenced by the volume of a corpus and a feature extraction process, the accuracy of emotion analysis is low. The technical effects of fully utilizing valid information of emotion analysis based on the concept features and the text emotion features is achieved, and the technical effect of improving the accuracy of emotion polarity classification is improved.

Description

technical field [0001] The invention relates to the technical field of information processing, in particular to a short text sentiment analysis method and device based on concepts and text sentiment. Background technique [0002] Sentiment analysis of text, also known as opinion mining, is a process of automatically extracting user intent from text. Traditional sentiment analysis includes dictionary-based analysis methods and machine learning-based methods. [0003] However, in the process of realizing the technical solution of the invention in the embodiment of the present application, the inventor of the present application found that the above-mentioned technology has at least the following technical problems: [0004] The analysis method in the prior art is affected by the size of the corpus and the process of feature extraction, which leads to the technical problem of low accuracy of sentiment analysis. Contents of the invention [0005] The embodiment of the presen...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06F17/30G06F17/27G06K9/62
CPCG06F16/35G06F40/242G06F40/211G06F40/289G06F40/30G06F18/2411
Inventor 莫益军杨帆姚澜
Owner EZHOU INST OF IND TECH HUAZHONG UNIV OF SCI & TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap