Table information extraction method, system and device, medium and program product
A technology of information extraction and tables, applied to neural learning methods, instruments, biological neural network models, etc., can solve problems such as difficult maintenance, cumbersome methods, and many rules, and achieve the effect of high accuracy and simple training rules
Active Publication Date: 2022-05-13
ANHUI DIGITAL INTELLIGENT CONSTR RES INST CO LTD
View PDF6 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, the current extraction methods based on deep learning are mostly rule-based methods that are cumbersome, require more rules to be defined, are difficult to maintain, cannot cope with complex and diverse table structures, and cannot meet the conditions of irregular tables. Requirements for Accurately Extracting Specified Content Information
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
[0053] Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numerals in different drawings refer to the same or similar elements unless otherwise indicated. The implementations described in the following exemplary examples do not represent all implementations consistent with the present invention. Rather, they are merely examples of apparatuses and methods consistent with aspects of the invention as recited in the appended claims.
[0054] figure 1 is a flow chart of a method for extracting table information according to an exemplary embodiment, such as figure 1 shown, including:
[0055] In step S101, a table for information extraction is obtained.
[0056] Specifically, the user uploads the form that needs to be extracted according to his own needs, so as to obtain the form to be extracted, so as to perform subsequent ...
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 a table information extraction method, system and device, a medium and a program product. The table information extraction method comprises the steps of obtaining a table to be subjected to information extraction; the table is input into a pre-trained graph neural network, the content in a target cell extracted from the table by the graph neural network is obtained, and the graph neural network is pre-trained based on the following mode: obtaining a table sample, and performing annotation classification according to a header cell and a content cell of the table sample to obtain a table sample; labeling and classifying header cells and content cells corresponding to the concerned content; and constructing a graph structure according to the labeled table structure of the table sample, and training the graph neural network based on the graph structure and a training task. According to the method, the table to be subjected to information extraction is extracted through the pre-trained graph neural network, and specified content information can be accurately extracted under regular and irregular table conditions.
Description
technical field [0001] The present disclosure relates to the technical field of form information extraction, and in particular to a form information extraction method, system, device, medium and program product. Background technique [0002] Tables are a very important and common semi-structured data, widely used in documents and web pages. Tabular information is clear and easy for humans to understand. However, manually extracting specified information from a large number of forms is usually very tedious and time-consuming, so there is a method of automatically extracting information from the form by machine. [0003] In the existing related technologies, research on tables includes table recognition, table structure extraction, table understanding, etc., wherein table understanding is further divided into table-based information retrieval, table question answering, and table content extraction. The extraction method of table content is faced with complex and diverse tabl...
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 Applications(China)
IPC IPC(8): G06V30/413G06V30/414G06V10/82G06N3/04G06N3/08
CPCG06N3/08G06N3/045
Inventor 宋恒刘道学仇明清李亚楠耿天宝程维国孙朝福张志强
Owner ANHUI DIGITAL INTELLIGENT CONSTR RES INST CO LTD