A Text Information Extraction System and Method for a Joint Reading Course Learning Mechanism
A learning mechanism and text information technology, applied in the field of information processing, can solve the problems of reducing practicability, error propagation, and complex structure of information extraction model, so as to improve the ability and promote the effect of reasoning speed
Active Publication Date: 2021-06-25
杭州识度科技有限公司
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] (1) There is an error propagation problem in the traditional pipeline method, and the results of the previous round of model extraction will affect the performance of the next round of models
[0005] (2) The joint method cannot handle the many-to-many situation well for open-domain text, and the model trained by it cannot capture the contextual representation information of entities and relationships well
[0006] (3) The current popular information extraction model has a complex structure and a large amount of reasoning and calculation, which further reduces the practicability of application in the industry
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 approach
[0044] As an embodiment of the present invention, the specific steps of outputting the prediction result set in the fine-grained extraction module are:
[0045] Generation of description questions: Aiming at the acquired entity set and relationship set, construct a description question set based on the logical relationship template;
[0046] For each relationship of the relationship set pt in the preceding example, sort out the associated subject type and object type, and then build a description for the entity set et based on the logical relationship template , the specific process is as follows figure 2 shown. For the relationship set pt in the above example, the subject type and object type associated with the relationship 'chairman' are 'company' and 'person' respectively, and the subject type and object type associated with the relationship 'date of establishment' are 'company' and 'Date', and then construct the description question based on the logical relationship tem...
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 present invention relates to a text information extraction system of a joint reading course learning mechanism, comprising: a preprocessing module: used to perform error correction operations on input texts to reduce noise; a coarse-grained extraction module: used to perform text processing on the preprocessing module Entity recognition and relationship extraction, to obtain the corresponding entity set and relationship set; fine-grained extraction module: used to verify the entity relationship set, and output the prediction result set; post-processing module: used to rule the prediction result set to obtain Enter the structured information corresponding to the text. The invention can improve the ability to capture entity and relationship context representation information; at the same time, due to the simple structure of the adopted model, it can reduce the influence of pipeline method error propagation to a certain extent, and further improve the reasoning speed.
Description
technical field [0001] The invention belongs to the field of information processing, and in particular relates to a system and method for extracting text information of a joint reading course learning mechanism. Background technique [0002] In the "New Generation Artificial Intelligence Development Plan" issued by the State Council, it is clearly pointed out that "associated understanding and knowledge mining, knowledge map construction and learning, knowledge evolution and reasoning, intelligent description and generation and other technologies" are the key common technology systems of the new generation of artificial intelligence. key breakthrough areas. The two key modules involved in knowledge graph construction and learning, knowledge evolution and reasoning are all supported by information extraction technology. Information extraction techniques can be divided into two types: pipeline methods and joint methods. Among them, the pipeline method is divided into two ste...
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): G06F16/31G06F16/35G06F40/295
CPCG06F16/313G06F16/353G06F40/295
Inventor 刘广峰
Owner 杭州识度科技有限公司