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

An Attribute-Level Sentiment Analysis Method with Full Attention Mechanism

A technology of sentiment analysis and attention, applied in the field of deep learning, can solve the problems of time overhead and low accuracy, and achieve the effect of improving accuracy and reducing costs

Active Publication Date: 2021-06-22
GUILIN UNIV OF ELECTRONIC TECH
View PDF10 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The present invention provides an attribute-level sentiment analysis method with a full attention mechanism, aiming at the problem of low time overhead and low accuracy when using deep learning to conduct sentiment analysis in user comments

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
  • An Attribute-Level Sentiment Analysis Method with Full Attention Mechanism
  • An Attribute-Level Sentiment Analysis Method with Full Attention Mechanism
  • An Attribute-Level Sentiment Analysis Method with Full Attention Mechanism

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0038] In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in combination with specific examples and with reference to the accompanying drawings.

[0039] see figure 1, an attribute-level sentiment analysis method with a full attention mechanism, including the following steps:

[0040] 1) Use the sample data to train the network model based on the attention mechanism:

[0041] Step 1. Sample data preprocessing

[0042] The given user comment sentence with real sentiment label classification is taken as sample data, and the sample data is preprocessed.

[0043] The content of the emotional polarity mark is the emotional polarity of the user comment sentence under the corresponding feature, specifically including three emotions: positive emotion, negative emotion, and neutral emotion.

[0044] The purpose of data preprocessing is to standardize data and construct a tra...

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 discloses an attribute-level sentiment analysis method of a full attention mechanism, which combines a self-attention mechanism network SAM‑NN and a specific aspect attention mechanism network AAM‑NN to generate semantic features at the vocabulary level and sentence level respectively, and finally The emotional polarity of the review sentence content is calculated through a fully connected neural network FC‑NN output layer. The method proposed by the present invention is implemented in a parallel structure, and in each network computing module, the present invention integrates the characteristics of specific aspects of information, ensuring that the method further analyzes user comment information based on specific aspects of information as much as possible Sentiment polarity in relation to specific attributes of the target object. Compared with the prior art, the method of the present invention not only effectively improves the accuracy of sentiment analysis tasks in specific aspects, but also effectively reduces the time spent on model training.

Description

technical field [0001] The invention relates to the technical field of deep learning, in particular to an attribute-level sentiment analysis method with a full attention mechanism. Background technique [0002] The rapid development and popularization of Internet technology has led to the generation of a large amount of online comment information, such as various e-commerce trading platforms, social network platforms, etc. How to effectively mine useful information from a large number of comments has become a problem in the field of natural language processing in recent years. Research hotspots, including opinion mining tasks in user review information. The emotions expressed in online review information are usually diverse, and a user review may contain different emotions in terms of multiple attributes of the review target. For example, a user review from a restaurant: "Noodles are delicious, soup is terrible". This comment contains characteristic information on two attr...

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): G06F16/35G06F40/211G06F40/289G06F40/30G06N3/04G06N3/08
CPCG06F16/35G06N3/08G06N3/045
Inventor 林煜明傅裕李优周娅张敬伟
Owner GUILIN UNIV OF ELECTRONIC TECH