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

Short text review sentiment analysis method based on Target-Aspect-Opinion joint extraction

A sentiment analysis, short text technology, applied in the field of artificial intelligence and deep learning, can solve the problem of unable to solve the overlapping of target words, unable to fully consider the inconsistency of simultaneous extraction tasks, etc.

Active Publication Date: 2021-08-06
EAST CHINA NORMAL UNIV
View PDF6 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although the existing research can well capture the interaction information of the emotional sentence context to better obtain the emotion at the attribute level, since Target and Opinion extraction are information extraction tasks, Aspect and sentiment analysis are multi-classification tasks, the existing sentiment analysis The method cannot fully consider the problem of task inconsistency caused by simultaneously extracting the three types of information of Target-Aspect-Opinion
At the same time, the existing research cannot solve the problem of overlapping target words, that is, in different emotional sentences, the same Target and Opinion correspond to different Aspects

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 review sentiment analysis method based on Target-Aspect-Opinion joint extraction
  • Short text review sentiment analysis method based on Target-Aspect-Opinion joint extraction
  • Short text review sentiment analysis method based on Target-Aspect-Opinion joint extraction

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0067] In this embodiment, a short text-based review emotional analysis method based on Target-Aspect-Opinion is divided, including the following steps:

[0068] Step A: Get the user's review of the automotive field "Car Home" user review car data, and build the original data set.

[0069] Step B: Data set pretreatment, remove redundant punctuation, very short comments, and so on.

[0070] Step C: According to the background knowledge in the automotive field, determine the subject of the automotive field to evaluate the subject of the vehicle as the power, price, interior, configuration, security, appearance, control, fuel consumption, space, comfort, etc., determine the emotional tendency to be positive To, neutral, negatively three categories.

[0071] Step D: Task for the dataset completed, labeling, such as evaluation objects, themes, evaluating phrases, emotional clear quadratic group labels.

[0072] Step E: Using the Depth Learning Framework Tensorflow, based on the host de...

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 present invention provides a short text review sentiment analysis method based on Target-Aspect-Opinion joint extraction. This invention first pre-processes the short text data set, screens the effective data, then performs pre-annotation work on the data set, and then builds an emotional analysis model based on Target-Aspect-Opinion joint extraction. The joint extraction model proposed by the present invention solves the problem of incomplete recognition caused by separately extracting Target or Aspect in the existing model, and effectively solves the problem of target word overlap by constructing TargetTaggers and Aspect-OpinionTaggers.

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, specifically to sentiment analysis at the attribute level of review text, and to a short text review sentiment analysis method based on Target-Aspect-Opinion joint extraction. . Background technique [0002] Platforms such as Weibo, forums, and shopping websites provide users with space for information exchange, thus generating a large amount of valuable user comment information. For example, user review data in the automotive field can not only help car manufacturers improve car product design and marketing strategies, but also provide decision-making basis and reference information for users to purchase cars. Therefore, fine-grained sentiment analysis is of great significance to different users. Traditional aspect-level sentiment analysis problems include topic-based senti...

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): G06F16/33G06F16/35G06F16/36G06F40/242G06N3/04G06N3/08
CPCG06N3/049G06N3/08G06F16/3344G06F16/35G06F16/374G06F40/242
Inventor 陈沁蕙赵慧姚婉薇
Owner EAST CHINA NORMAL UNIV
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