CN114202755A - Transaction background authenticity auditing method and system based on OCR (optical character recognition) and NLP (non-line segment) technologies - Google Patents
Transaction background authenticity auditing method and system based on OCR (optical character recognition) and NLP (non-line segment) technologies Download PDFInfo
- Publication number
- CN114202755A CN114202755A CN202111503088.5A CN202111503088A CN114202755A CN 114202755 A CN114202755 A CN 114202755A CN 202111503088 A CN202111503088 A CN 202111503088A CN 114202755 A CN114202755 A CN 114202755A
- Authority
- CN
- China
- Prior art keywords
- ocr
- data
- image
- invoice
- area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Abstract
The invention provides a transaction background authenticity auditing method and system based on OCR and NLP technologies, which comprises the following steps: step 1: the method comprises the steps of requesting account receivable information data of a business from a transaction system through an interface, and integrating the data into a data format to be verified after the data are acquired; step 2: acquiring an image data list of the service through an image platform; and step 3: based on OCR and NLP technologies, image recognition is carried out according to the unique number of the image, corresponding information is captured according to preset key element fields, and data to be verified are obtained; and 4, step 4: and establishing a comparison rule according to different service scenes, document types and index types, carrying out authenticity verification on data to be verified, and generating a verification report for assisting verification. The invention introduces intelligent image recognition and verification technology into transaction background authenticity audit, has the characteristics of high recognition accuracy and short time consumption, can effectively prevent artificial audit errors, and effectively saves the cost of human resources.
Description
Technical Field
The invention relates to the technical field of OCR and NLP, in particular to a transaction background authenticity auditing method and system based on OCR and NLP technologies.
Background
Aiming at the current online trade financing application flow and the auditing mode of the trade background image data of each branch, the receivable account registration information in the receivable financing business application is input by a client manager one by one, and then manual comparison and verification are carried out on the trade background image data uploaded by the client or the client manager, such as contracts, invoices and the like one by one, so as to complete the authenticity auditing of the trade background information. With the dramatic increase in the amount of audit material, a great deal of time is wasted in manual entry and comparison of the material and is prone to errors.
Patent document CN113221890A (application number: CN202110574251.0) discloses a cloud mobile phone word content supervision method and system based on OCR, the system includes a text information processing module, a text information comparison module and a sensitive information thesaurus; and performing text line detection, text line identification and sensitive character information comparison on the screen capture data by adopting a text line-based detection and identification algorithm, and performing early warning and warning processing on illegal character contents in the screen capture data of the cloud mobile phone user.
The existing authenticity auxiliary image auditing means based on OCR has the following defects: the image identification accuracy is low, and the requirement of auxiliary examination cannot be met; the adaptability to the diversity of the documents is poor, and only a single specific type of document can be processed; the requirement on the image quality is high, and the low-quality image cannot be processed; for mixed type image materials containing pictures and tables, the problems of inaccurate table positioning, interference of the structure of a chart on the recognition effect and the like exist; and moreover, each information element identified by the OCR is manually compared and verified one by one, and time is consumed.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide a transaction background authenticity auditing method and system based on OCR (optical character recognition) and NLP (non-line segment) technologies.
The transaction background authenticity auditing method based on the OCR and NLP technologies comprises the following steps:
step 1: the method comprises the steps of requesting account receivable information data of a business from a transaction system through an interface, and integrating the data into a data format to be verified after the data are acquired;
step 2: acquiring an image data list of the service through an image platform;
and step 3: based on OCR and NLP technologies, image recognition is carried out according to the unique number of the image, corresponding information is captured according to preset key element fields, and data to be verified are obtained;
and 4, step 4: and establishing a comparison rule according to different service scenes, document types and index types, carrying out authenticity verification on data to be verified, and generating a verification report for assisting verification.
Preferably, the image acquisition device is used for acquiring an image of the material to be verified, and performing OCR area positioning on the acquired image, wherein the area category comprises: a text area, a table area and an invoice area;
training is carried out on the data with the labels, an OCR model based on a deep convolutional neural network is constructed, image data are input into the OCR model, content distribution of each region is output, and a plurality of region sub-images are formed.
Preferably, each regional sub-image is input to an OCR model for content extraction, content characters of each region are output, and business key element extraction is performed according to the output characters;
the key elements of the service in the text area and the table area comprise: contract name, contract number, contract amount, contract signing date, contract effective date and contract signing party;
the business key elements of the invoice area comprise: invoice code, invoice number, tax amount, tax rate, invoicing date and goods category.
Preferably, index types, position information and keywords are introduced simultaneously by combining a named entity recognition model and a rule based on data analysis, and business key elements of a text area and a table area are extracted.
Preferably, the invoice region subgraph is input into a deep convolution neural network for classification, each category corresponds to each business key element in the invoice region, and the invoice region key element extraction is performed by combining AI technologies of sample enhancement, SVM, target detection, table recognition and cross-page table splicing.
The transaction background authenticity auditing system based on the OCR and NLP technologies provided by the invention comprises:
module M1: the method comprises the steps of requesting account receivable information data of a business from a transaction system through an interface, and integrating the data into a data format to be verified after the data are acquired;
module M2: acquiring an image data list of the service through an image platform;
module M3: based on OCR and NLP technologies, image recognition is carried out according to the unique number of the image, corresponding information is captured according to preset key element fields, and data to be verified are obtained;
module M4: and establishing a comparison rule according to different service scenes, document types and index types, carrying out authenticity verification on data to be verified, and generating a verification report for assisting verification.
Preferably, the image acquisition device is used for acquiring an image of the material to be verified, and performing OCR area positioning on the acquired image, wherein the area category comprises: a text area, a table area and an invoice area;
training is carried out on the data with the labels, an OCR model based on a deep convolutional neural network is constructed, image data are input into the OCR model, content distribution of each region is output, and a plurality of region sub-images are formed.
Preferably, each regional sub-image is input to an OCR model for content extraction, content characters of each region are output, and business key element extraction is performed according to the output characters;
the key elements of the service in the text area and the table area comprise: contract name, contract number, contract amount, contract signing date, contract effective date and contract signing party;
the business key elements of the invoice area comprise: invoice code, invoice number, tax amount, tax rate, invoicing date and goods category.
Preferably, index types, position information and keywords are introduced simultaneously by combining a named entity recognition model and a rule based on data analysis, and business key elements of a text area and a table area are extracted.
Preferably, the invoice region subgraph is input into a deep convolution neural network for classification, each category corresponds to each business key element in the invoice region, and the invoice region key element extraction is performed by combining AI technologies of sample enhancement, SVM, target detection, table recognition and cross-page table splicing.
Compared with the prior art, the invention has the following beneficial effects:
(1) the method utilizes artificial intelligence technologies such as an Optical Character Recognition (OCR) technology, a Natural Language Processing (NLP) technology and the like to automatically recognize the image materials submitted in the approval process of the trade financing business and intelligently extract key information elements, provides inspection suggestions including the effectiveness of business contracts, the authenticity verification results of invoices, the consistency of the invoices and the contracts and the like through intelligent logic verification, and provides further judgment and operation for business application personnel and approval personnel, so that the approval efficiency is improved;
(2) the invention can identify the image data which are mixed and arranged according to different systems, different qualities and various content types, and realizes specific content extraction based on NLP technologies such as information extraction and semantic analysis so as to meet the requirements of information comparison and verification;
(3) the invention introduces intelligent image recognition and verification technology into transaction background authenticity audit, has the characteristics of high recognition accuracy and short time consumption, can effectively prevent artificial audit errors, and effectively saves the cost of human resources.
Drawings
Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments with reference to the following drawings:
FIG. 1 is a flow chart of an embodiment of the present invention;
FIG. 2 is a flow chart of the image recognition function of the present invention;
fig. 3 is a flowchart of a business key element extraction function in the present invention.
Detailed Description
The present invention will be described in detail with reference to specific examples. The following examples will assist those skilled in the art in further understanding the invention, but are not intended to limit the invention in any way. It should be noted that it would be obvious to those skilled in the art that various changes and modifications can be made without departing from the spirit of the invention. All falling within the scope of the present invention.
Example (b):
the invention provides a transaction background authenticity auditing method based on OCR and NLP technologies, as shown in figure 1, comprising the following steps:
s100: and extracting service information. And requesting accounts receivable information data of the business from the trading system through the interface. And integrating the information into a data format to be checked after the data is acquired, and temporarily storing the data format in a middle station system for later use. The process can be clicked by an operator or automatically triggered by the system;
s200: and requesting the image data list of the service from the image platform. Selecting image data to be verified by an operator;
s300: and the middle station system submits the unique number of the selected image to an intelligent image recognition system based on OCR and NLP, and clicks to trigger image recognition. Capturing corresponding information according to preset key element fields, forming data to be verified, and returning the data to the middle platform system;
s400: triggering an intelligent verification function. Specific and different comparison rules are constructed according to different service scenes, document types and index types, and verification reports for assisting verification are generated for reference of verification service personnel; and simultaneously triggering an invoice true check function, sending invoice information after intelligent image recognition to a true check interface, and adding a return result to a check report.
As shown in fig. 2, the S300 intelligent image recognition system based on OCR and NLP specifically includes:
s310: collecting image data. Carrying out image acquisition on a material to be verified through an image acquisition device;
s320: the OCR zone is located. The region categories include: a text field, a table field, and a ticket field. Inputting the image data into an OCR model based on a deep convolutional neural network page by page. The OCR model outputs the content distribution of the above regions to form a plurality of regional subgraphs. The OCR model is trained on a large amount of marked data, and has the area positioning capability with high accuracy;
s330: and (5) extracting the OCR content. And inputting the sub-images of the regions into an OCR model for content extraction, and outputting content characters of the regions. The OCR model is trained on a large amount of marked data, and has high-accuracy content recognition capability;
s340: and extracting business key elements. The business key elements of the text area and the table area include but are not limited to: contract name, contract number, contract amount, contract signing date, contract effective date and contract signing party; the business key elements of the invoice area include but are not limited to: invoice code, invoice number, amount (including tax), tax rate, invoicing date, goods category. Can be dynamically configured according to the service requirement.
As shown in fig. 3, the extracting of the business key elements in S340 specifically includes:
s341: and extracting the business key elements in the text area and the table area. In consideration of the diversification of the content format, the extraction of the business key elements combines two technical means of a named entity recognition model and a rule based on data analysis. And adopting a special model aiming at a special industry and a special contract format. For example: for purchasing contracts in the building industry, data marking needs to be carried out on the contract formats, and models special for the contract formats need to be trained. Taking the overall generalization capability of the recognition model into consideration, and simultaneously extracting business key elements by introducing rules such as index types, position information, keywords and the like;
s342: and extracting key elements in the invoice area. And inputting the invoice region subgraph in the step S320 into a deep convolutional neural network for classification, wherein each class corresponds to each business key element in the invoice region in the step S340. And the accuracy of identifying and extracting the key elements of the service is ensured by combining a plurality of AI technical means such as sample enhancement, SVM, target detection, form identification and page-crossing form splicing.
The main technical difficulties of the invention are as follows: 1. the quality of image data uploaded by clients is uneven, multiple types of materials exist in the same image file, and different image types need to be identified and classified through an identification means; 2. the contract files in the image files belong to non-standard files, the format is varied, most text contents are long, key information to be extracted is scattered, and information extraction difficulty such as goods names, unit prices, quantity and the like is high; 3. the invoice file in the image file comprises a plurality of invoice types, and the identification and key element extraction difficulty is increased under the conditions of folding, various layout arrangements, overlap printing and the like; 4. the comparison and verification of the trade background information elements need to be verified and checked according to the information elements extracted from various types of image files, and different elements need to be classified and analyzed through a certain scheme.
The method used in the scheme is as follows:
1. the uploaded image files are segmented page by page, the image files are classified by adopting a convolutional neural network, and meanwhile, the files with context characteristics such as contracts and the like are classified, so that subsequent key business elements can be conveniently identified and extracted;
2. considering the diversification of contract file formats, extracting the contract key elements combines two technical means of a named entity recognition model and a rule based on data analysis, a special model is adopted for special industries and specific contract formats, and the overall generalization capability of the recognition model is improved by combining the extracted rules of key index types, position information, keywords and the like;
the key elements are as follows: such as contract name, contract subject (party A, party B), contract signing date, contract number, total amount of the contract, payment method, goods information, contract validity period, etc.;
the special model is as follows: named entity recognition models trained for specific industries, and the like;
special industry and specific format: the purchasing contract for different industries, such as the construction industry, has a specific contract format.
3. For invoices of different types, a deep convolutional neural network is adopted for classification, and various AI technical means such as sample enhancement, SVM, target detection, table identification and cross-page table splicing are adopted, so that the identification and extraction accuracy of various information elements of different invoice types is improved, and meanwhile, information confirmation and comparison are carried out by combining interface calling of external invoice verification data;
4. the method comprises the steps of comparing and auditing key information elements extracted based on image files with receivable account registration information, adopting technical means of classified file verification and similar information index comparison verification, forming corresponding comparison rules by combining scenes of business application, and providing a result report for auxiliary audit by combining a Natural Language Processing (NLP) technology for reference of audit business personnel.
Classifying files: such as business contracts, agreements, different kinds of invoices, trade lineup notes, etc.;
similar information indexes are as follows: for example, the comparison and verification of the information indexes of the same type, such as the contract amount and the total invoice amount in the business contract, are verified.
The transaction background authenticity auditing system based on the OCR and NLP technologies provided by the invention comprises: module M1: the method comprises the steps of requesting account receivable information data of a business from a transaction system through an interface, and integrating the data into a data format to be verified after the data are acquired; module M2: acquiring an image data list of the service through an image platform; module M3: based on OCR and NLP technologies, image recognition is carried out according to the unique number of the image, corresponding information is captured according to preset key element fields, and data to be verified are obtained; module M4: and establishing a comparison rule according to different service scenes, document types and index types, carrying out authenticity verification on data to be verified, and generating a verification report for assisting verification.
Carry out image acquisition through image acquisition device to the material of waiting to check to carry out OCR regional location to the image of gathering, regional classification contains: a text area, a table area and an invoice area; training is carried out on the data with the labels, an OCR model based on a deep convolutional neural network is constructed, image data are input into the OCR model, content distribution of each region is output, and a plurality of region sub-images are formed. Inputting the sub-images of each region into an OCR model for content extraction, outputting content characters of each region, and extracting business key elements according to the output characters; the key elements of the service in the text area and the table area comprise: contract name, contract number, contract amount, contract signing date, contract effective date and contract signing party; the business key elements of the invoice area comprise: invoice code, invoice number, tax amount, tax rate, invoicing date and goods category. And combining a named entity recognition model and a rule based on data analysis, and introducing an index type, position information and a keyword at the same time to extract the business key elements of the text area and the table area. Inputting the invoice region subgraph into a deep convolution neural network for classification, wherein each category corresponds to each business key element in the invoice region, and extracting the invoice region key elements by combining AI technologies of sample enhancement, SVM, target detection, table recognition and cross-page table splicing.
Those skilled in the art will appreciate that, in addition to implementing the systems, apparatus, and various modules thereof provided by the present invention in purely computer readable program code, the same procedures can be implemented entirely by logically programming method steps such that the systems, apparatus, and various modules thereof are provided in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Therefore, the system, the device and the modules thereof provided by the present invention can be considered as a hardware component, and the modules included in the system, the device and the modules thereof for implementing various programs can also be considered as structures in the hardware component; modules for performing various functions may also be considered to be both software programs for performing the methods and structures within hardware components.
The foregoing description of specific embodiments of the present invention has been presented. It is to be understood that the present invention is not limited to the specific embodiments described above, and that various changes or modifications may be made by one skilled in the art within the scope of the appended claims without departing from the spirit of the invention. The embodiments and features of the embodiments of the present application may be combined with each other arbitrarily without conflict.
Claims (10)
1. A transaction background authenticity auditing method based on OCR and NLP technologies is characterized by comprising the following steps:
step 1: the method comprises the steps of requesting account receivable information data of a business from a transaction system through an interface, and integrating the data into a data format to be verified after the data are acquired;
step 2: acquiring an image data list of the service through an image platform;
and step 3: based on OCR and NLP technologies, image recognition is carried out according to the unique number of the image, corresponding information is captured according to preset key element fields, and data to be verified are obtained;
and 4, step 4: and establishing a comparison rule according to different service scenes, document types and index types, carrying out authenticity verification on data to be verified, and generating a verification report for assisting verification.
2. An auditing method for transaction background authenticity based on OCR and NLP technologies according to claim 1 characterized in that the material to be verified is image-captured by an image capture device and OCR area positioning is performed on the captured image, the area categories include: a text area, a table area and an invoice area;
training is carried out on the data with the labels, an OCR model based on a deep convolutional neural network is constructed, image data are input into the OCR model, content distribution of each region is output, and a plurality of region sub-images are formed.
3. An auditing method for transaction background authenticity based on OCR and NLP technology according to claim 2 characterized in that each region sub-image is input to OCR model for content extraction, content words of each region are output, and business key element extraction is performed according to the output words;
the key elements of the service in the text area and the table area comprise: contract name, contract number, contract amount, contract signing date, contract effective date and contract signing party;
the business key elements of the invoice area comprise: invoice code, invoice number, tax amount, tax rate, invoicing date and goods category.
4. A transaction background authenticity auditing method based on OCR and NLP technologies according to claim 3 characterized in that index type, position information and keywords are introduced simultaneously in combination with named entity recognition model and rules based on data analysis to extract the key elements of text area and table area services.
5. The transaction background authenticity auditing method based on OCR and NLP technologies according to claim 3 characterized in that invoice region sub-graphs are input to a deep convolutional neural network for classification, each class corresponds to each business key element in the invoice region, and invoice region key element extraction is performed in combination with AI technologies of sample enhancement, SVM, target detection, table recognition and cross-page table splicing.
6. A transaction background authenticity auditing system based on OCR and NLP technologies, comprising:
module M1: the method comprises the steps of requesting account receivable information data of a business from a transaction system through an interface, and integrating the data into a data format to be verified after the data are acquired;
module M2: acquiring an image data list of the service through an image platform;
module M3: based on OCR and NLP technologies, image recognition is carried out according to the unique number of the image, corresponding information is captured according to preset key element fields, and data to be verified are obtained;
module M4: and establishing a comparison rule according to different service scenes, document types and index types, carrying out authenticity verification on data to be verified, and generating a verification report for assisting verification.
7. An auditing system for transaction background authenticity based on OCR and NLP technology according to claim 6 where the material to be verified is image captured by an image capture device and OCR area location is performed on the captured image, the area categories include: a text area, a table area and an invoice area;
training is carried out on the data with the labels, an OCR model based on a deep convolutional neural network is constructed, image data are input into the OCR model, content distribution of each region is output, and a plurality of region sub-images are formed.
8. An auditing system for transaction background authenticity based on OCR and NLP technology according to claim 7 characterized in that each region sub-graph is input to OCR model for content extraction, content words of each region are output, and business key element extraction is performed according to the output words;
the key elements of the service in the text area and the table area comprise: contract name, contract number, contract amount, contract signing date, contract effective date and contract signing party;
the business key elements of the invoice area comprise: invoice code, invoice number, tax amount, tax rate, invoicing date and goods category.
9. An auditing system for transaction background authenticity based on OCR and NLP technology according to claim 8 characterized in that index type, position information and keywords are introduced simultaneously to extract the key elements of text area and table area services in combination with named entity recognition model and rules based on data analysis.
10. The transaction background authenticity auditing system based on OCR and NLP technology according to claim 8 characterized in that invoice region sub-graphs are input to a deep convolutional neural network for classification, each class corresponds to each business key element in the invoice region, and invoice region key element extraction is performed in combination with AI technologies of sample enhancement, SVM, target detection, table recognition and cross-page table concatenation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111503088.5A CN114202755A (en) | 2021-12-09 | 2021-12-09 | Transaction background authenticity auditing method and system based on OCR (optical character recognition) and NLP (non-line segment) technologies |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111503088.5A CN114202755A (en) | 2021-12-09 | 2021-12-09 | Transaction background authenticity auditing method and system based on OCR (optical character recognition) and NLP (non-line segment) technologies |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114202755A true CN114202755A (en) | 2022-03-18 |
Family
ID=80651797
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111503088.5A Pending CN114202755A (en) | 2021-12-09 | 2021-12-09 | Transaction background authenticity auditing method and system based on OCR (optical character recognition) and NLP (non-line segment) technologies |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114202755A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114973290A (en) * | 2022-07-26 | 2022-08-30 | 简单汇信息科技(广州)有限公司 | Intelligent order examination method and system based on OCR engine |
| CN115130989A (en) * | 2022-06-24 | 2022-09-30 | 北京百度网讯科技有限公司 | Method, device and equipment for auditing service document and storage medium |
| CN116959018A (en) * | 2023-06-05 | 2023-10-27 | 简单汇信息科技(广州)有限公司 | OCR-based intelligent checking method, system and equipment |
| CN117291740A (en) * | 2023-09-26 | 2023-12-26 | 湖北盈嘉集团有限公司 | Receivables data authenticity intelligent identification auditing system based on big data |
| CN117391068A (en) * | 2023-10-27 | 2024-01-12 | 中国人寿保险股份有限公司山东省分公司 | Method and system for checking life insurance security business information based on RPA |
| WO2024060759A1 (en) * | 2022-09-21 | 2024-03-28 | 深圳前海环融联易信息科技服务有限公司 | Supply chain financial asset auditing method and apparatus, and device and medium |
| CN117291740B (en) * | 2023-09-26 | 2024-04-19 | 湖北盈嘉集团有限公司 | Receivables data authenticity intelligent identification auditing system based on big data |
-
2021
- 2021-12-09 CN CN202111503088.5A patent/CN114202755A/en active Pending
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115130989A (en) * | 2022-06-24 | 2022-09-30 | 北京百度网讯科技有限公司 | Method, device and equipment for auditing service document and storage medium |
| CN114973290A (en) * | 2022-07-26 | 2022-08-30 | 简单汇信息科技(广州)有限公司 | Intelligent order examination method and system based on OCR engine |
| CN114973290B (en) * | 2022-07-26 | 2022-11-04 | 简单汇信息科技(广州)有限公司 | Intelligent order examining method and system based on OCR engine |
| WO2024060759A1 (en) * | 2022-09-21 | 2024-03-28 | 深圳前海环融联易信息科技服务有限公司 | Supply chain financial asset auditing method and apparatus, and device and medium |
| CN116959018A (en) * | 2023-06-05 | 2023-10-27 | 简单汇信息科技(广州)有限公司 | OCR-based intelligent checking method, system and equipment |
| CN116959018B (en) * | 2023-06-05 | 2024-02-23 | 简单汇信息科技(广州)有限公司 | OCR-based intelligent checking method, system and equipment |
| CN117291740A (en) * | 2023-09-26 | 2023-12-26 | 湖北盈嘉集团有限公司 | Receivables data authenticity intelligent identification auditing system based on big data |
| CN117291740B (en) * | 2023-09-26 | 2024-04-19 | 湖北盈嘉集团有限公司 | Receivables data authenticity intelligent identification auditing system based on big data |
| CN117391068A (en) * | 2023-10-27 | 2024-01-12 | 中国人寿保险股份有限公司山东省分公司 | Method and system for checking life insurance security business information based on RPA |
| CN117391068B (en) * | 2023-10-27 | 2024-04-05 | 中国人寿保险股份有限公司山东省分公司 | Method and system for checking life insurance security business information based on RPA |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109887153B (en) | Finance and tax processing method and system | |
| CN114202755A (en) | Transaction background authenticity auditing method and system based on OCR (optical character recognition) and NLP (non-line segment) technologies | |
| CN107608958B (en) | Contract text risk information mining method and system based on unified modeling of clauses | |
| CN110069623B (en) | Abstract text generation method and device, storage medium and computer equipment | |
| Stanisławek et al. | Kleister: key information extraction datasets involving long documents with complex layouts | |
| CN114862540B (en) | Bill auditing system and method thereof | |
| CN111652232B (en) | Bill identification method and device, electronic equipment and computer readable storage medium | |
| US20070237427A1 (en) | Method and system for simplified recordkeeping including transcription and voting based verification | |
| CN110414927B (en) | Method and device for automatically generating voucher during bill processing | |
| WO2019157029A1 (en) | System and method for classifying images of an evidence | |
| CN103678109A (en) | Dump document analysis method, device and system | |
| CN114117171B (en) | Intelligent project file collecting method and system based on energized thinking | |
| CN105243117A (en) | Data processing system and method | |
| US20140207631A1 (en) | Systems and Method for Analyzing and Validating Invoices | |
| CN111932363A (en) | Identification and verification method, device, equipment and system for authorization book | |
| CN111914729A (en) | Voucher association method and device, computer equipment and storage medium | |
| CN111400529B (en) | Data processing method and device | |
| CN111709464A (en) | System and screening method based on data dictionary configuration and seal image recognition technology | |
| CN110717732A (en) | Information authentication method and system | |
| CN105913071A (en) | Information processing device, information processing system and information processing method | |
| CN115907673A (en) | Supply chain system | |
| CN105243584A (en) | Data processing method and system | |
| KR102562186B1 (en) | System for providing rental property management based official letter sending service | |
| KR102416998B1 (en) | Appatus for automatically collecting and classification tax related documents and method thereof | |
| CN113537964A (en) | Application form processing method, device, storage medium and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |