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

A Cross-Domain Sentiment Analysis Method

A sentiment analysis and cross-domain technology, applied in the field of data mining, can solve the problem of not paying enough attention to the differences between the source domain and the target domain, and achieve the effect of narrowing the difference

Active Publication Date: 2021-09-17
YUNNAN UNIV
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, deep learning-based methods often need to tune a large number of parameters, and do not pay enough attention to the differences between source and target domains

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
  • A Cross-Domain Sentiment Analysis Method
  • A Cross-Domain Sentiment Analysis Method
  • A Cross-Domain Sentiment Analysis Method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0019] figure 1 The method flowchart provided for the embodiment of the present invention, such as figure 1 As shown, the method may include the following steps:

[0020] Step 101: Quantify source and target domain sentiment text

[0021] Specifically include the following steps:

[0022] First, for the emotional text set in the source domain, extract sn feature words W S ={ w s1 , w s2 ,…, w sn}; For the emotional text set in the target field, extract tm feature words W T ={ w t1 , w t2 ,…, w tm};

[0023] In the implementation of the present invention, feature words are obtained by word segmentation and removal of stop words, and the English text can be restored by part of speech after word segmentation and removal of stop words, and unigram and bigrams words can be extracted as feature words, and the known TF-IDF weight can be used Filter the feature words to reduce the number of feature words.

[0024] Second, for the emotional text set in the sourc...

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 belongs to the technical field of data mining and discloses a cross-domain sentiment analysis method. The present invention regards the source domain and the target domain as the global marginal distribution, and performs modeling based on the Bayesian network; then, constructs the global feature model through the source domain feature model and the target domain feature model; thirdly, establishes the global feature model through the global feature model The association between global features realizes the feature expansion of the source domain and the target domain; finally, the classifier is trained based on the expanded label samples, and the classifier is used to predict the non-label samples of the target domain. The invention can effectively narrow the distance between domains, and provide technical support for analyzing target domains with insufficient label samples.

Description

[0001] The invention belongs to the technical field of data mining, and relates to a cross-domain sentiment analysis method, more specifically, a Bayesian network-based cross-domain sentiment analysis method. Background technique [0002] Emotional texts refer to texts with subjective emotional tendencies. The analysis of the emotional tendency of the text is an important technical basis for applications such as public opinion monitoring, word-of-mouth analysis, and topic monitoring. Cross-domain sentiment analysis studies the technical issue of how to make full use of relevant source domain samples for analysis under the condition that sentiment has topic correlation and domain correlation, and the samples in the target domain are sparse. The key to cross-domain sentiment analysis is to narrow the differences between domains. At present, the main methods of cross-domain sentiment analysis are based on traditional machine learning methods, such as the SFA (Spectral Feature Al...

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 Patents(China)
IPC IPC(8): G06F40/30G06F40/216G06F16/35G06K9/62
CPCG06F40/216G06F40/30G06F18/29
Inventor 李维华刘慧清段云浩王翔
Owner YUNNAN UNIV