Evaluation object extraction method for case-related microblog comments based on multi-head attention mechanism

A technology for commenting and evaluating objects on Weibo, applied in neural learning methods, instruments, network data retrieval, etc., can solve problems such as inconsistent and irrelevant expressions, and achieve good extraction results.

Active Publication Date: 2020-08-25
KUNMING UNIV OF SCI & TECH
View PDF8 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The present invention provides a method for extracting evaluation objects of microblog comments involved in the case based on a multi-head attention mechanism, which is used to solve the problem that the evaluation objects of public comments in the microblog comments involved in the case are not related to the case involved and the public is concerned about the microblog comments involved in the case. Inconsistencies in the expression of the evaluation object when commenting

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
  • Evaluation object extraction method for case-related microblog comments based on multi-head attention mechanism
  • Evaluation object extraction method for case-related microblog comments based on multi-head attention mechanism
  • Evaluation object extraction method for case-related microblog comments based on multi-head attention mechanism

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0044] Embodiment 1: as Figure 1-2 As shown, the evaluation object extraction method of microblog comments involved in the case based on the multi-head attention mechanism, the specific steps of the method are as follows:

[0045] Step1. Use Scrapy as a crawling tool to imitate user operations, log in to Weibo, and formulate a template based on the XPath path of the page data to obtain Weibo text and Weibo comments related to the case, such as the "Mercedes-Benz female car owner's rights protection case" crawling Weibo comments "More than 90,000 entries" and nearly 25,000 entries of the "Chongqing River Falling Case" were experimentally studied, and Weibo comments were deduplicated and screened;

[0046] Step2. According to the case involved in the text of the microblog, set the case elements of the case for the screened microblog comments, mark the evaluation objects of the microblog comments involved in the case, and divide the marked microblog comment corpus into training ...

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 relates to an evaluation object extraction method for case-related microblog comments based on a multi-head attention mechanism. The evaluation object extraction method comprises the following steps: firstly, collecting case-related microblog texts and microblog comments; marking evaluation objects of the case-related microblog comments; obtaining character vectors of the case-related microblog comments, and obtaining word vectors, part-of-speech vectors related to part-of-speech of nouns, context vectors and case element vectors; splicing the obtained character vector and word vector, obtaining the correlation between the part-of-speech vector and the context vector by adopting a multi-head attention mechanism, and calculating the similarity probability weight of the maximumcase element and the part-of-speech vector; and performing point multiplication on the obtained splicing vector, correlation weight and maximum similarity probability weight, taking an obtained result as input of a BiLSTM model, and then obtaining a final sequence label, namely an extraction result, through a conditional random field CRF. According to the method, a relatively good extraction effect is achieved, and the two typical case-related microblog comment data sets are respectively improved by 1% and 1.9% in comparison with a baseline model.

Description

technical field [0001] The invention relates to an evaluation object extraction method of microblog comments involved in a case based on a multi-head attention mechanism, and belongs to the technical field of data mining. Background technique [0002] With the advent of the self-media era, the vigorous development of my country's new media and the deepening of Internet public opinion, the focus cases tried by some courts are rapidly fermented on the Internet, causing widespread controversy in society, and even affecting judicial trials, seriously affecting social stability. In this context, the country is also promoting the construction of smart courts. Judicial departments can effectively grasp and correctly guide online public opinion in real time to help ensure that courts independently exercise judicial power in accordance with the law and maintain social fairness and justice. Weibo, as one of the social media for free comments by the general public, has a large number o...

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): G06F16/951G06F40/117G06F40/284G06N3/04G06N3/08
CPCG06F16/951G06F40/284G06F40/117G06N3/049G06N3/08
Inventor 余正涛赵培莲毛存礼相艳郭军军黄于欣
Owner KUNMING UNIV OF SCI & TECH
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