Event recognition and classification method and device based on multilingual attention mechanism
A classification method and attention technology, applied in semantic analysis, neural learning methods, biological neural network models, etc., can solve the problem that the recognition effect of monolingual features cannot meet the needs, so as to improve the accuracy rate and recall rate, and improve the recognition effect Effect
Active Publication Date: 2020-07-28
INST OF AUTOMATION CHINESE ACAD OF SCI
View PDF6 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0008] In order to solve the above problems in the prior art, that is, in order to solve the problem that the recognition effect of monolingual features cannot meet the requirements, the present invention proposes an event recognition and classification method based on a multilingual attention mechanism, including:
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0061] Preferred embodiments of the present invention are described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the present invention, and are not intended to limit the protection scope of the present invention.
[0062] A kind of event identification and classification method based on multilingual attention mechanism of the present invention, such as figure 2 shown, including:
[0063] Step 1. Map the data that only labels monolingual event information into multilingual parallel data;
[0064] Step 2, performing lexical-level alignment on the multilingual parallel data, and obtaining consistent representations of the same event in multiple different languages through the multilingual alignment relationship, that is, multilingual alignment data;
[0065] Step 3, use step 2 to obtain the consistent representation of the same event in multiple differ...
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
Login to View More Abstract
The invention relates to the technical field of natural language processing, in particular to a multilingual attention mechanism-based event identification and classification method and apparatus, andaims to solve the problem that the monolingual feature identification effect cannot meet the demands. The method comprises the steps of mapping data only marked with monolingual event information into multilingual parallel data; performing vocabulary-level alignment on the multilingual parallel data, and obtaining a consistency representation of a same event in multiple different languages through a multilingual alignment relationship; obtaining multilingual consistency information based on a monolingual attention model; obtaining multilingual complementarity information based on a multilingual attention model; and performing cooperative reasoning based on the multilingual consistency information and the multilingual complementarity information, and outputting a final identification result through a nonlinear neural network judgment model. The event identification effect can be improved.
Description
technical field [0001] The invention relates to the technical field of natural language processing, in particular to an event recognition and classification method and device based on a multilingual attention mechanism. Background technique [0002] With the large-scale development of the Internet, it is becoming easier to obtain information, and people are exposed to massive amounts of information in various fields almost all the time, such as news in sports, entertainment, military and other fields. However, these information are generally disordered, messy, unstructured, and there is a certain degree of information redundancy. How to discover interesting events from massive amounts of information with the aid of computers is an urgent problem to be solved. Event Identification (Event Identification) technology is a powerful means to solve this problem. Event recognition is an independent subtask of Event Extraction (Event Extraction), which mainly studies how to identif...
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
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06F40/30G06N3/08
CPCG06F40/30G06N3/08
Inventor 陈玉博刘康赵军刘健
Owner INST OF AUTOMATION CHINESE ACAD OF SCI