Textual emotion marking method, device and system

An emotion tagging and text technology, applied in instruments, computing, character and pattern recognition, etc., can solve the problems of large manpower and material resources, consume a lot of time and money, and it is difficult for annotators to express and classify, and achieve high accuracy and economical efficiency. The effect of time and money

Inactive Publication Date: 2016-06-15
HARBIN INST OF TECH SHENZHEN GRADUATE SCHOOL
View PDF3 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The method of manual labeling mainly has the following defects: 1. On the one hand, in order to ensure the labeling accuracy of the emotional corpus, it is necessary to conduct a large amount of training on domain-related knowledge and emotional labeling specifications for each staff member participating in the labeling, which will consume a lot of time and money
On the other hand, due to the lack of systematic labeling specifications for emotional corpus, it is difficult to ensure that labelers can accurately and efficiently label emotional corpus after the training is over.
2. Different annotators often have different emotional cognition experiences, resulting in different or even opposite results when different annotators annotate the same corpus
When this happens, it is usually necessary for the annotators to discuss and decide the final annotation results together. This process often consumes a lot of time and energy of the annotators, and will eventually seriously slow down the annotation process.
[0004] To sum up, due to the complexity of human emotion cognition mechanism, it is often difficult for annotators to accurately express and classify their own true emotions, resulting in serious inaccurate annotation of corpus, and the annotating process is very cumbersome and consumes huge manpower and material resources

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
  • Textual emotion marking method, device and system
  • Textual emotion marking method, device and system
  • Textual emotion marking method, device and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0020] Please refer to figure 1 , the embodiment of the present application provides a text emotion labeling method, comprising the following steps:

[0021] 101. Acquire the electroencephalogram signal of the text reader output by the sensor.

[0022] Please also refer to image 3 , before step 101, obtaining the EEG signal of the text reader output by the sensor, it also includes:

[0023] 101A. Convert the text to be tagged into the form of multiple phrases through the chunk analysis technology, and present it to the text readers.

[0024] The process of converting the text to be marked into multiple phrases through the block analysis technique is as follows:

[0025] Original Text: "Chinese athletes will win glory for the country in the 2008 Olympic Games."

[0026] Phrase forms converted by chunk analysis: "Chinese athletes", "will", "in the 2008 Olympic Games", "glory for the country".

[0027] Another example: original corpus: "Xiao Ming handed in homework on time....

Embodiment 2

[0048] Please refer to Figure 4 , this example provides a text emotion labeling device, including:

[0049] The acquiring unit 30 is configured to acquire the electroencephalogram signal of the text reader output by the sensor.

[0050] The calculation unit 31 is used to calculate the power mean value of the denoised EEG signal in four frequency bands respectively, as the feature vector of emotion analysis. The four frequency bands are delta wave, theta wave, alpha wave and beta wave.

[0051] The predicting unit 32 is configured to input the feature vector of the sentiment analysis into the classification model, and predict and obtain the labeling result of the sentiment of the text.

[0052] Wherein, the classification model includes: a corresponding model of spectrum power mean values ​​on four types of frequency bands and emotion labels, and the corresponding model is obtained by pre-training training samples.

[0053] Such as Figure 5 As shown, in one embodiment, it ...

Embodiment 3

[0062] Please refer to Image 6 , this example provides a text sentiment tagging system, including:

[0063] A sensor 50 and a processor 51.

[0064] The sensor 50 is used to collect the EEG signal of the text reader and output it to the processor 51 .

[0065] The processor 51 is used to acquire the EEG signal of the text reader output by the sensor 50, and respectively calculate the power mean value of the denoised EEG signal on the four types of frequency bands as the feature vector of the sentiment analysis, wherein the four The class frequency bands are delta waves, theta waves, alpha waves and beta waves; and, input the feature vectors of the sentiment analysis into the classification model, and predict and obtain the labeling results of the sentiments of the text.

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 application provides a textual emotion marking method, device and system. According to the present invention, a tester reads the to-be-classified texts, a brain electrical signal of a text reader is acquired, and then the emotion mark is carried out on the text according to the brain electrical signal, so that the emotion of a marker can be reflected really from a cognitive neuroscience angle, and a very high accuracy is guaranteed. Moreover, by the application, the marking personnel do not need the long-time training, when a text emotion analysis system is developed, a lot of time and funds can be saved.

Description

technical field [0001] The present invention relates to the technical field of natural language processing, in particular to a text emotion labeling method, device and system. Background technique [0002] With the vigorous development of Internet technology, the Internet gradually adopts the open architecture concept of user participation, so a large amount of information about user participation is generated on the Internet, such as a large number of comment information on people, events, products, etc. These comments express people's various emotional colors and emotional tendencies, such as joy, anger, sorrow, joy, criticism, praise, etc. Obviously, the rest of the users can learn about public opinion on a certain event or product by browsing these comments with subjective emotions. [0003] In the process of building a traditional emotional corpus, the generation method of emotional tags is usually marked manually. These annotated corpora consist of two parts: raw tex...

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): G06K9/62G06K9/00
CPCG06F2218/16G06F2218/08G06F18/2415
Inventor 徐睿峰杜嘉晨桂林黄锦辉
Owner HARBIN INST OF TECH SHENZHEN GRADUATE SCHOOL
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