Disclosure of Invention
Aiming at the defects, the technical problem to be solved by the invention is how to realize automatic execution and intelligent decision by means of sensing and identifying various information of a user by means of an artificial intelligence technology and a natural language processing technology and modeling a subsequent decision flow.
In view of the above-mentioned drawbacks, an object of the present invention is to provide a method, a system, an electronic device, a computer storage medium, and a program product for labeling and identifying information entities of digital files.
The method is applied to a server side, full text information in a digital file is extracted or data which is ready for word segmentation is collected, the data is input into a label function, word segmentation training is carried out on the information based on regular matching and labels are generated, according to the input parameters of a model, after the label data and original data are integrated, the label data and the original data are input into the model to carry out entity recognition model training, and a result set and a corresponding scoring result are generated.
Preferably, the tasks are created according to requirements, the training models are associated, and the data sets to be marked are uploaded after the calculation space is distributed.
Preferably, the plain text is processed into data quadruplet data, keywords, entity types, positions and text subscripts in the table through the word list.
Preferably, the data is used as an input of the tag function, and an entity is generated for each row of data after the training of the Snorkel model.
Preferably, the method specifically comprises the following steps:
s1, extracting full text information from the text material input by the user;
s2, performing word segmentation processing on the text information;
s3, processing the plain text into quaternary data in a table through a word list;
s4, associating the Snorkel training model and distributing calculation space;
s5, uploading a data set needing marking;
s6, generating a label function and training a model;
s7, generating a corresponding entity for the input data through Snorkel training;
and S8, fusing the marked data and the original data to generate the Bert training data.
Preferably, the digital file is a coronary angiogram report sheet and/or a coronary angiogram case report.
Preferably, the method comprises the following steps:
s201, extracting full text information including a coronary angiography case number from a coronary angiography case report list and/or a coronary angiography case report through OCR;
s202, processing the plain text into data quadruple data in a table through a word list, wherein the labels are 1, 2 and … …, and the word list is a left front descending branch, a right circumflex branch and … …;
s203, taking the data obtained in the step S202 as input of a label function, training by Snorkel, and then generating an entity aiming at the data of each row, wherein the coronary angiography report sheet corresponds to a right coronary artery label, and the coronary angiography case number corresponds to a conventional posture angiography label;
and S204, fusing the marked data and the original data to generate the Bert training data.
The invention provides a digital file information entity labeling and identifying method, which is applied to an internet medical platform, acquires user authorization permission based on terminal equipment, acquires a digital file uploaded by a user and sends the digital file to a data center processing system of a background server, the data center processing system carries out OCR (optical character recognition) on the digital file to acquire full-text information or collects data ready for word segmentation, inputs the data into a label function, carries out word segmentation training on the information based on regular matching and generates a label, integrates the label data and original data according to the input parameters of the model, inputs the integrated label data and the original data into the model to carry out entity identification model training to generate a result set and a corresponding scoring result, and outputs a specific solution to the user by combining with an application product of the internet medical platform.
Preferably, the internet medical platform simulates insurance claim settlement examination by using a rule engine through a diagnosis decision model by combining the result set and the corresponding scoring result, performs decision tree logic execution, and obtains and outputs an intelligent examination conclusion to a user.
Preferably, the internet medical platform calculates a recommendation decision factor by combining the result set and the corresponding scoring result, calculates a decision by the factor, and obtains an insurance recommendation combination by using a recommendation algorithm.
Preferably, the internet medical platform combines the result set and the corresponding scoring result based on the rehabilitation data model, and gives related rehabilitation instruments and a proposal through model calculation.
Preferably, the internet medical platform combines the result set and the corresponding grading result with a national hospital big database to give a recommended medical treatment party or docking medical specialist.
Preferably, the internet medical platform creates tasks according to requirements, associates training models, and uploads data sets to be marked after calculation space is allocated.
The invention provides a system for marking and identifying digital file information entities, which comprises at least one terminal device, at least one internet medical platform and at least one server, the terminal equipment collects digital files uploaded by a user, the internet medical platform acquires user authorization permission based on the terminal equipment, collects the digital files uploaded by the user and sends the digital files to a data center processing system of a background server, the data center processing system carries out OCR (optical character recognition) on the digital files to obtain full text information or collects data ready for word segmentation, the data is input into a label function, word segmentation training is carried out on the information based on regular matching and labels are generated, according to the parameters of the model, after the label data and the original data are integrated, the integrated data are input into the model for entity recognition model training, a result set and a corresponding grading result are generated, and a specific solution is output to a user by combining with an application product of an internet medical platform.
Preferably, the tasks are created according to requirements, the training models are associated, and the data sets to be marked are uploaded after the calculation space is distributed.
Preferably, the plain text is processed into data quadruplet data, keywords, entity types, positions and text subscripts in the table through the word list.
The present invention provides a computer readable storage medium having stored thereon a computer program/instructions which, when executed by a processor, implement the steps of the above-described method.
The present invention provides a computer program product comprising computer programs/instructions which, when executed by a processor, implement the steps of the above-described method.
The present invention provides an electronic device, including:
a processor; and
a memory arranged to store computer-executable instructions that, when executed, cause the processor to:
the method comprises the steps that user authorization permission is obtained based on terminal equipment, the terminal equipment collects digital files uploaded by a user, the internet medical platform obtains the user authorization permission based on the terminal equipment, the digital files uploaded by the user are collected and sent to a data center processing system of a background server, the data center processing system conducts OCR (optical character recognition) on the digital files to obtain full text information or collect data ready for word segmentation, the data are input into a label function, word segmentation training is conducted on the information based on regular matching and labels are generated, the label data and original data are integrated according to parameters of a model and then input into the model to conduct entity recognition model training to generate a result set and a corresponding scoring result, and a specific solution is output to the user in combination with an application product of the internet medical platform.
The invention solves the problems of timeliness and cost of information entity marking of massive digital files through model training, enables non-algorithm personnel to quickly realize operation through a program realization mode, belongs to great innovation of tools, can be widely applied to data marking in the field of medical health, provides convenience for medical insurance, health, medical resource butt joint and the like, and greatly saves time and capital cost.
Detailed Description
Features and exemplary embodiments of various aspects of the present invention will be described in detail below, and in order to make objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not to be construed as limiting the invention. It will be apparent to one skilled in the art that the present invention may be practiced without some of these specific details. The following description of the embodiments is merely intended to provide a better understanding of the present invention by illustrating examples of the present invention.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
One embodiment of the present specification provides a method for labeling and identifying an information entity of a digital file, which includes extracting textual information in the digital file or collecting data ready for word segmentation, inputting the data into a label function, performing word segmentation training on the information based on regular matching and generating a label, integrating the label data and original data according to parameters of a model, inputting the integrated data into the model, and performing entity identification model training to generate a result set and a corresponding scoring result.
In some embodiments, tasks are created on demand, training models are associated, and data sets to be marked are uploaded after computing space is allocated.
In some embodiments, plain text is processed through a vocabulary into data tuple data in a table, keywords, entity types, locations, and text indices.
In some embodiments, the data is used as an input to the tag function, and an entity is generated for each row of data after training of the Snorkel model.
In the mental disease category, mental diseases and sleep disorders are caused by cardiovascular reasons, and in the case of cardiovascular coronary arteries, the digital file is a coronary angiography report sheet and/or a coronary angiography case report.
As shown in fig. 1, an embodiment of the present specification provides a method for labeling and identifying an information entity of a digital file, including:
s101, extracting full text information in a digital file or collecting data ready for word segmentation;
s102, inputting data into a tag function;
s103, performing word segmentation training on the information based on the regular matching and generating a label;
s104, integrating the label data and the original data according to the parameters of the model;
and S105, inputting the result into the model to train the entity recognition model, and generating a result set and a corresponding scoring result.
As shown in fig. 2, the present specification provides an embodiment of a method for labeling and identifying information entities of a digital file, including:
s201, extracting full text information including a coronary angiography case number from a coronary angiography case report list and/or a coronary angiography case report through OCR;
s202, processing the plain text into data quadruple data in a table through a word list, wherein the labels are 1, 2 and … …, and the word list is a left front descending branch, a right circumflex branch and … …;
s203, taking the data obtained in the step S202 as input of a label function, training by Snorkel, and then generating an entity aiming at the data of each row, wherein the coronary angiography report sheet corresponds to a right coronary artery label, and the coronary angiography case number corresponds to a conventional posture angiography label;
and S204, fusing the marked data and the original data to generate the Bert training data.
As shown in fig. 3, the present specification provides an embodiment of a method for labeling and identifying an information entity of a digital file, which specifically includes:
s1, extracting full text information from the text material input by the user;
s2, performing word segmentation processing on the text information;
s3, processing the plain text into quaternary data in a table through a word list;
s4, associating the Snorkel training model and distributing calculation space;
s5, uploading a data set needing marking;
s6, generating a label function and training a model;
s7, generating a corresponding entity for the input data through Snorkel training;
and S8, fusing the marked data and the original data to generate the Bert training data.
The specification provides an embodiment of a digital file information entity labeling and identifying method, which is applied to an internet medical platform, acquires user authorization permission based on terminal equipment, acquires a digital file uploaded by a user and sends the digital file to a data center processing system of a background server, the data center processing system performs OCR (optical character recognition) on the digital file to acquire full-text information or collect data ready for word segmentation, inputs the data into a label function, performs word segmentation training on the information based on regular matching and generates labels, integrates the label data and original data according to parameters of the model, inputs the information into the model to perform entity identification model training to generate a result set and corresponding scoring results, and outputs a specific solution to the user in combination with an application product of the internet medical platform.
In some embodiments, the internet medical platform simulates insurance claim settlement review by using a rule engine through a diagnosis decision model in combination with the result set and corresponding scoring results, performs decision tree logic execution, and obtains and outputs an intelligent review conclusion to a user.
In some embodiments, the internet medical platform calculates a recommendation decision factor by combining the result set and the corresponding scoring result, calculates a decision by the factor, and obtains an insurance recommendation combination by using a recommendation algorithm.
In some embodiments, the internet medical platform combines the result set and the corresponding scoring result based on a rehabilitation data model, and gives related rehabilitation instruments and a proposal through model calculation.
In some embodiments, the internet medical platform combines the result set and corresponding scoring results with a national hospital big database to give recommended medical parties or docking medical professionals.
In some embodiments, the internet medical platform creates tasks according to requirements, associates training models, and uploads data sets to be marked after computing space is allocated.
As shown in fig. 4, the present specification provides an embodiment of a digital file information entity labeling and recognition system, which includes at least one terminal device, at least one internet medical platform, and at least one server, where the terminal device collects digital files uploaded by users, the internet medical platform obtains user authorization based on the terminal device, collects digital files uploaded by users and sends the digital files to a data center processing system of a background server, the data center processing system performs OCR recognition on the digital files to obtain full text information or collect data ready for word segmentation, inputs the data into a tag function, performs word segmentation training on the information based on regular matching and generates tags, integrates the tag data and original data according to parameters of a model, inputs the integrated data into the model to perform entity recognition model training to generate a result set and a corresponding scoring result, and outputting a specific solution to a user by combining an application product of the Internet medical platform.
In some embodiments, tasks are created on demand, training models are associated, and data sets to be marked are uploaded after computing space is allocated.
In some embodiments, plain text is processed through a vocabulary into data tuple data in a table, keywords, entity types, locations, and text indices.
In one specific example, the insurance includes health insurance, human insurance, and the various insurance includes health condition notification.
In a specific example, the corresponding rehabilitation apparatus or suggestion is given through the recommendation algorithm model and imported into the corresponding third-party electronic commerce platform, such as the Ali health, the Jingdong health and the like.
In one specific example, a recommended treatment or docking medical specialist is given in conjunction with the result set and corresponding scoring results and a national hospital big database, and the specialist includes a hospital or doctor on an online medical platform such as Ali health, clove garden, spring rain doctor, etc.
In some specific examples, such as nerve decline, sleep disorder, cardiovascular disease, etc., products or health products that help relieve stress and improve sleep quality, such as electrocardiographs, health pillows, melatonin, etc., can be given.
In some specific examples, the digital files uploaded by the system to the user include, but are not limited to, diagnostic proofs, case reports, prescriptions, and hospitalizations.
One embodiment of the present specification provides a computer readable storage medium having stored thereon a computer program/instructions which, when executed by a processor, implement the method of: the method comprises the steps of obtaining user authorization permission based on terminal equipment, collecting digital files uploaded by a user and sending the digital files to a data center processing system of a background server, carrying out OCR (optical character recognition) on the digital files to obtain full-text information or collecting data ready for word segmentation, inputting the data into a label function, carrying out word segmentation training on the information based on regular matching and generating labels, integrating the label data and original data according to parameters of a model, inputting the integrated label data and original data into the model to carry out entity recognition model training to generate a result set and a corresponding scoring result, and outputting a specific solution to the user by combining with an application product of an Internet medical platform.
One embodiment of the present specification provides a computer program product comprising computer programs/instructions which, when executed by a processor, implement the method of: the method comprises the steps of obtaining user authorization permission based on terminal equipment, collecting digital files uploaded by a user and sending the digital files to a data center processing system of a background server, carrying out OCR (optical character recognition) on the digital files to obtain full-text information or collecting data ready for word segmentation, inputting the data into a label function, carrying out word segmentation training on the information based on regular matching and generating labels, integrating the label data and original data according to parameters of a model, inputting the integrated label data and original data into the model to carry out entity recognition model training to generate a result set and a corresponding scoring result, and outputting a specific solution to the user by combining with an application product of an Internet medical platform.
One embodiment of the present specification provides an electronic apparatus including:
a processor; and
a memory arranged to store computer-executable instructions that, when executed, cause the processor to:
the method comprises the steps of obtaining user authorization permission based on terminal equipment, obtaining user authorization permission based on the terminal equipment, collecting digital files uploaded by a user and sending the digital files to a data center processing system of a background server, carrying out OCR (optical character recognition) on the digital files to obtain full text information or collecting data ready for word segmentation, inputting the data into a label function, carrying out word segmentation training on the information based on regular matching and generating labels, integrating the label data and original data according to parameters of a model, inputting the label data into the model to carry out entity recognition model training to generate a result set and a corresponding scoring result, and outputting a specific solution to the user by combining with application products of an Internet medical platform.
The invention solves the problems of timeliness and cost of information entity marking of massive digital files through model training, enables non-algorithm personnel to quickly realize operation through a program realization mode, belongs to great innovation of tools, can be widely applied to data marking in the field of medical health, provides convenience for medical insurance, health, medical resource butt joint and the like, and greatly saves time and capital cost.
For convenience of description, the above devices are described as being divided into various units by function, and are described separately. Of course, the functionality of the units may be implemented in one or more software and/or hardware when implementing the present application.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The application may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The application may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.