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

Attention mechanism-based copywriting input recognition and classification method for judicial scenes

A technology for recognition classification and attention, applied in neural learning methods, character and pattern recognition, semantic analysis, etc., can solve problems such as ambiguity, increase the difficulty of classification, and non-standard multi-intent, so as to improve accuracy and improve text quality. The effect of sorting efficiency

Pending Publication Date: 2020-04-03
BEIMING SOFTWARE
View PDF4 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] Search engine technology is a new technology that has emerged in the development of the Internet in recent years. The purpose of its application is to help Internet users quickly find and display the information they need in the vast amount of information. Quickly find and display the information they need, there are many types of judicial texts, and to put it bluntly, search engines are a means of classification. Sorting technologies such as the inverted index of the company return the results that users want. In the process, the user input text has the characteristics of ambiguity, multi-intent, and irregularity, so there are various input methods such as irregular input text and irregular user input. Irregular feature words and unregistered words that cannot be recognized by word segmentation dictionaries appear in the text. Traditional text classification cannot understand the primary and secondary of user query intentions, which increases the difficulty of classification and to a certain extent leads to the inability to accurately obtain short texts. feature representation, the efficiency of classification is not high, therefore, we propose a judicial scene based on attention mechanism to recognize and classify copywriting input

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

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0021] An attention mechanism-based method for identifying and classifying copywriting input in judicial scenarios, the specific steps of which are as follows:

[0022] A: Obtain the text data set, perform data preprocessing on the collected judicial text data, vectorize and represent each word and form a mapping matrix, and then use the mapping matrix and the kernel matrix with a weight of 1 to perform one-dimensional volume Product operation to detect potential multi-scale candidate semantic units in short texts;

[0023] B: Construct a semantic extension matrix as the extended information of the short text, and at the same time obtain the semantic features of the text sequence in the short text, assign the semantic feature vector of the short text to the classifier, and predict the category of the short text;

[0024] C: Select the features of the preprocessed text to form the feature space of the text data set, sort out the data in the feature space to build a vocabulary, ...

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 attention mechanism-based copywriting input recognition and classification method for judicial scenes in the technical field of copywriting input recognition and classification. The attention mechanism-based copywriting input recognition and classification method comprises the following specific steps: first, obtaining a text dataset, and constructing a semantic extension matrix; performing feature selection on the preprocessed text, and forming a feature space of the text data set; importing data in the embedded matrix into an attention mechanism weight model of anattention mechanism module, a double-layer LSTM neural network layer and a CNN module; splicing and fusing the data of different dimensions; importing the processed data into a Softmax classifier to carry out normalization processing; using global long-term dependence to pay attention to local semantic features, so as to remove redundant or irrelevant features, and improve the text classificationefficiency; and fusing the extracted features, so that extract deep semantic features in the text corpus are extracted in a rich and careful mode, and then the judicial short text feature recognitioncapacity of the model is improved, and the judicial copywriting text recognition and classification efficiency is improved.

Description

technical field [0001] The invention relates to the technical field of text input recognition and classification, in particular to an attention mechanism-based method for text input recognition and classification in judicial scenarios. Background technique [0002] Search engine technology is a new technology that has emerged in the development of the Internet in recent years. The purpose of its application is to help Internet users quickly find and display the information they need in the vast amount of information. Quickly find and display the information they need, there are many types of judicial texts, and to put it bluntly, search engines are a means of classification. Sorting technologies such as the inverted index of the company return the results that users want. In the process, the user input text has the characteristics of ambiguity, multi-intent, and irregularity, so there are various input methods such as irregular input text and irregular user input. Irregular...

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): G06F40/30G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/044G06N3/045G06F18/24
Inventor 袁馨谢文锐
Owner BEIMING SOFTWARE
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