Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Comment text attribute-level sentiment analysis method based on deep learning

A technology of deep learning and sentiment analysis, applied in the field of comment text attribute-level sentiment analysis, can solve problems such as failure to consider the fusion of comment text information and attribute information, improve accuracy, shorten model training iteration time, and promote information interaction Effect

Pending Publication Date: 2020-09-18
EAST CHINA NORMAL UNIVERSITY
View PDF0 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although the existing methods can pay more attention to the part of the sentence for the specific attribute to a certain extent, so as to perform attribut

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
  • Comment text attribute-level sentiment analysis method based on deep learning
  • Comment text attribute-level sentiment analysis method based on deep learning
  • Comment text attribute-level sentiment analysis method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0049] Example 1

[0050] The deep learning-based sentiment analysis method for comment text attribute level in this embodiment includes the following steps:

[0051] Step A: Obtain the text data of online user reviews in the automotive field of a forum to form the original data set.

[0052] Step B: According to the existing knowledge in the field, manually determine twenty attribute categories such as version, body color, and power system. Tag the original data set one by one with two-tuple labels like (attribute category, emotional orientation).

[0053] Step C: Text preprocessing and word segmentation. Text preprocessing mainly includes the removal of special symbols such as "@" and "&", and the removal of stop words such as "的" and "啊". The word segmentation adopts the separation method based on "character" as the unit.

[0054] Step D: Use the deep learning framework tensorflow, based on the host device memory 32G, GPU: NVIDIA GTX1080Ti, use the new self-attention fusion networ...

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 comment text attribute-level sentiment analysis method based on deep learning, which comprises the following steps of: A, obtaining public comment text data, such as comment text data related to a certain forum in a certain field and the like to form an original data set; b, marking a two-tuple label with the shape such as (attribute category and emotional tendency) on theoriginal data set; c, carrying out text preprocessing and word segmentation; and step D, performing training to obtain a final model by utilizing a deep learning framework and adopting a novel self-attention fusion network, so as to carry out attribute-level sentiment analysis. According to the method, the comment text and the corresponding attribute information are effectively fused through thenovel self-attention fusion network, information interaction between the comment text and the corresponding attribute information is better promoted, and the accuracy of predicting the attribute-levelemotional tendency of the comment text is effectively improved; and compared with a general RNN-based deep learning solution, the method has shorter model training iteration time.

Description

technical field [0001] The present invention relates to the technical fields of artificial intelligence and deep learning, in particular to research and analysis related to natural language processing, and in particular to attribute-level sentiment analysis of comment texts. Background technique [0002] With the development of Internet technology, the way of interaction between customers and retailers / producers has changed dramatically. Nowadays, more and more companies tend to collect customer feedback, such as reviews, online forum discussions, etc., to improve customer experience, product design, and more. However, a major challenge we face is how to extract useful information from the massive overload of information. Sentiment analysis is a key means to solve the above problems. Sentiment analysis can provide a lot of valuable information for companies and customers. Going a step further, some companies are not only interested in the overall emotional orientation of ...

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): G06F16/35G06F16/33G06K9/62G06N3/04G06N3/08
CPCG06F16/35G06F16/3335G06F16/3344G06N3/08G06N3/047G06N3/045G06F18/2415G06F18/241Y02D10/00
Inventor 黄泽林赵慧陈沁蕙
Owner EAST CHINA NORMAL UNIVERSITY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com