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

A goal-based sentiment classification method incorporating descriptive knowledge

A technology for sentiment classification and knowledge description, applied in the field of deep learning and knowledge base, can solve problems such as inability to handle sentiment analysis well

Active Publication Date: 2020-10-09
BEIJING INSTITUTE OF TECHNOLOGYGY
View PDF9 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

(3) Existing methods cannot handle target-sensitive sentiment analysis problems well, such as two sentences: "The waitress is so hot!", "The meal is so hot!", a sentence is extremely emotional for 'waitress' Sex is positive, and the second sentence has a negative emotional polarity for 'meal'. If you don't know what 'waitress' and 'meal' are, you will not be able to correctly classify the sentiment

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 goal-based sentiment classification method incorporating descriptive knowledge
  • A goal-based sentiment classification method incorporating descriptive knowledge
  • A goal-based sentiment classification method incorporating descriptive knowledge

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0064] A goal-based sentiment classification method incorporating descriptive knowledge, such as figure 1 shown, including the following steps:

[0065] Step 1. Implement verification experiments on the benchmark dataset SemEval 2014Task4 dataset. The data set includes a restaurant review data set (restaurant) and a laptop computer review data set (laptop), each of which is divided into a training set and a test set, and the samples contain one of three emotional polarities (positive, negative, and neural). The training set is used to optimize the model parameters, and the test set is used to test the model effect.

[0066] Step 2. From the online knowledge base, extract the description information (target description sentence) in the text form of the target word or phrase.

[0067] Further, the method of extracting the description information of the target in step 2 is:

[0068] Step 2.1. Use the SPARQL statement to query the description of the target in DBpedia. If it exi...

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 target-based emotion classification method integrated with description knowledge, and belongs to the technical field of deep learning and knowledge bases. In the emotional analysis task based on the target, the description knowledge of the target is utilized and the description information is fused into the target representation, and the description of the target comes from a knowledge base which contains mass knowledge about the entity and is easy to access. Aiming at target-based sentiment analysis, the invention provides a new model-DEGN, the model uses a gating mechanism to closely and effectively combine description knowledge of a given target into an end-to-end neural network, and uses a supervision attention mechanism to extract important information from the target, a target context and a description sentence of the target to generate more effective representations thereof. The invention surpasses all existing models in both F1-score and strict accuracyto reach the highest level.

Description

technical field [0001] The invention belongs to the technical field of deep learning and knowledge base, and relates to a target-based emotion classification method incorporating description knowledge. Background technique [0002] With the rapid development of information technology and social media and the advent of the era of big data, the Internet has penetrated into all aspects of people's daily lives, and more and more people are posting their opinions and comments on services, commodities, politics, news, etc. online , the Internet has accumulated massive texts containing opinions. When an individual is trying to make a decision and translate it into behavior, he or she usually considers and considers other people's opinions, suggestions, views and views. In other words, the opinions and opinions of others will have a significant impact on a person's decisions and behaviors. If an enterprise or institution can effectively analyze the public's views on products or se...

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): G06K9/62G06N3/04G06N3/08
Inventor 宋丹丹邢博文廖乐健
Owner BEIJING INSTITUTE OF TECHNOLOGYGY