Experiment record data processing method, system and equipment and storage medium

Abstract

According to the experimental record data processing method, system and equipment and the storage medium provided by the invention, the plurality of experimental record templates are established after analysis and classification are carried out according to the layout plates of the different types of experimental record images in advance, and then the structural data identification is carried out on the experimental record images; the method comprises the following steps: identifying experimental record structured data, inputting the identified experimental record structured data into an experimental record template to generate template input data, and determining a template matching degree between the template input data and the experimental record template generating the template input data based on template attribute data of the experimental record template; furthermore, standard experiment record data of the experiment record image is determined in all template input data and is stored, so that historical paper experiment records can be converted into the experiment record image and data input is performed, and automation and standardization of experiment record data input are realized; and the paper experiment record data can be rapidly and accurately input.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to an experiment record data processing method, system, device and storage medium. Background Art

[0002] Scientific experiments are an important means of conducting scientific research. Various types of research and discoveries based on scientific experimental data can promote the rapid development of science and technology and social productivity. In the past, scientific experiments usually recorded the experimental process, experimental results and related information on paper, forming a large amount of paper experimental record data. These massive amounts of paper experimental record data provide a large amount of effective data precipitation for subsequent industry analysis and research.

[0003] In today's information digitization, various paper experiment record data are difficult to associate and inconvenient for relevant statistical analysis, so it is necessary to enter various paper experiment record data into the digital information system to achieve the goals of archiving, retrieval, and statistical analysis of the experiment record data. Due to the complexity of the experimental record data, such as the particularity of the experimental content and typos in the handwriting process, the existing paper experiment record data entry is usually manually entered into the system by relevant personnel. The manual entry of paper experiment record data is difficult to quickly and accurately batch enter paper experiment record data. Summary of the invention

[0004] The present invention provides an experiment record data processing method, system, device and storage medium to solve the problem that the existing paper experiment record data is manually input and it is difficult to input the paper experiment record data in batches quickly and accurately.

[0005] In a first aspect, an embodiment of the present application provides a method for processing experimental record data, comprising:

[0006] Performing structured data recognition on the experimental record image to obtain experimental record structured data;

[0007] Based on at least two preset experimental record templates, data entry is performed on the experimental record structured data respectively to generate at least two template entry data, wherein the experimental record template is a standard template established after analyzing and classifying the layout panels of multiple types of experimental record images in advance;

[0008] Based on the template attribute data of each experimental record template, the template matching calculation is performed on each template input data and the experimental record template that generates the template input data, so as to obtain the template matching degree of each template input data;

[0009] Based on the template matching degree of each template input data, standard experiment record data of the experiment record image is determined from all template input data and stored.

[0010] Optionally, the template attribute data includes user attribute data for characterizing the user's operation behavior characteristics on the experiment record template. Based on the template attribute data of each experiment record template, a template matching calculation is performed on each template input data and the experiment record template that generates the template input data to obtain the template matching degree of each template input data, including:

[0011] Determining a user attribute matching degree of the template input data based on user attribute data of the experiment record template for generating the template input data, wherein the user attribute matching degree includes a matching degree of each user attribute in the user attribute data;

[0012] According to the user attribute matching degree of each template input data, the preset scoring model is used to calculate the template matching of each template input data to obtain the template matching degree of each template input data.

[0013] Optionally, the template attribute data also includes global attribute data. According to the user attribute matching degree of each template input data, a preset scoring model is used to calculate the template matching of each template input data to obtain the template matching degree of each template input data, including:

[0014] Based on the global attribute data of the experiment record template for generating the template input data, determining the global attribute matching degree of the template input data, wherein the global attribute matching degree includes the matching degree of each global attribute in the global attribute data;

[0015] According to the user attribute data and global attribute data of each experimental record template, the preset scoring model is used to calculate the template matching of each template input data to obtain the template matching degree of the template input data.

[0016] Optionally, the preset scoring model includes a scoring value of each user attribute and a scoring value of each global attribute; according to the user attribute matching degree and the global attribute matching degree of each template input data, the preset scoring model is used to perform template matching calculation on each template input data to obtain the template matching degree of the template input data, including:

[0017] Determine the user attribute matching score of the template input data according to the matching degree with each user attribute and the scoring value of each user attribute in the user attribute matching degree;

[0018] Determine the global attribute matching score of the template input data according to the matching degree with each global attribute in the global attribute matching degree and the scoring value of each global attribute;

[0019] According to the user attribute matching score and the global attribute matching score of each template input data, the template matching degree of each template input data is correspondingly determined.

[0020] Optionally, based on at least two preset experiment record templates, data entry is performed on the experiment record structured data respectively to generate at least two template entry data, including:

[0021] According to the structured data of the experimental record, multiple experimental record templates are matched with preset fields, and at least two matching preliminary screening experimental record templates are obtained through preliminary screening;

[0022] The key in the structured data of the experimental record is matched with the fields of each preliminary screening experimental record template and data is entered to generate at least two template entry data.

[0023] Optionally, structured data recognition is performed on the experimental record image to obtain experimental record structured data, including:

[0024] Using optical character recognition technology to locate and identify key-value pairs of the experimental record image, and obtaining key-value pair data of the experimental record image;

[0025] Based on a pre-generated common vocabulary for experiments, field matching and correction are performed on the key-value pair data to obtain the experimental record structured data of the experimental record image.

[0026] Optionally, based on the template matching degree of each template input data, standard experimental record data of the experimental record image is determined and stored in all template input data, including:

[0027] Based on the template matching degree of each template input data, the template input data is sorted and processed to generate recommended input data;

[0028] The recommended entry data is pushed to the user, and the template entry data selected by the user in the recommended entry data is stored as standard experiment record data of the experiment record image.

[0029] Optionally, after storing the template entry data selected by the user from the recommended entry data as standard experiment record data of the experiment record image, the method further includes:

[0030] Update the template attribute data of the corresponding experiment record template according to the template input data selected by the user; and / or,

[0031] According to the template input data selected by the user, the preset scoring model is updated and optimized, so as to facilitate the subsequent template matching calculation of each template input data and the corresponding experimental record template based on the optimized preset scoring model.

[0032] In a second aspect, an embodiment of the present application provides an experiment record data processing system, including an experiment record data processing device, the experiment record data processing device including:

[0033] A recognition module is used to perform structured data recognition on the experimental record image to obtain experimental record structured data;

[0034] An input module, used for inputting data of the experimental record structured data respectively based on at least two preset experimental record templates, so as to generate at least two template input data, wherein the experimental record template is a standard template established after analyzing and classifying the layout of multiple types of experimental record images in advance;

[0035] A calculation module, for calculating the template matching of each template input data and the experimental record template generating the template input data based on the template attribute data of each experimental record template, so as to obtain the template matching degree of each template input data;

[0036] The storage module is used to determine and store the standard experiment record data of the experiment record image in all the template entry data based on the template matching degree of each template entry data.

[0037] In a third aspect, an embodiment of the present application provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned parameter aggregation method based on federated learning when executing the computer program.

[0038] In a fourth aspect, an embodiment of the present application provides a readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned parameter aggregation method based on federated learning are implemented.

[0039] In a solution provided by the above-mentioned experimental record data processing method, system, device and storage medium, multiple experimental record templates are established by pre-analyzing and classifying the layout formats of different types of experimental record images, and then structured data recognition is performed on the experimental record images, and the recognized experimental record structured data is entered into the experimental record template to generate template entry data, and then based on the template attribute data of the experimental record template, the template matching degree between the template entry data and the experimental record template that generates the template entry data is determined, and then the standard experimental record data of the experimental record image is determined and stored in all template entry data. Historical paper experimental records can be converted into experimental record images and data entry can be performed, thereby realizing the automation and standardization of experimental record data entry, and being able to quickly and accurately enter paper-based experimental record data. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0041] Figure 1 is a schematic diagram of an application environment of an experimental record data processing method in one embodiment of the present invention;

[0042] Figure 2 It is a flow chart of a method for processing experimental record data in one embodiment of the present invention;

[0043] Figure 3 is another flow chart of a method for processing experimental record data in one embodiment of the present invention;

[0044] Figure 4 yes Figure 2 A schematic diagram of an implementation process of step S10;

[0045] Figure 5 A layout format diagram of an experimental record image in one embodiment of the present invention;

[0046] Figure 6 yes Figure 2 A schematic diagram of an implementation process of step S20;

[0047] Figure 7 yes Figure 2 A schematic diagram of an implementation process of step S30;

[0048] Figure 8 yes Figure 2 A schematic diagram of an implementation process of step S40;

[0049] Fig. 9 yes Figure 1 A structural schematic diagram of the experimental record data processing device;

[0050] Fig.10 It is a structural diagram of a computer device in one embodiment of the present invention. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or their collections. It should also be understood that the term "and / or" used in the present specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes these combinations.

[0053] In addition, in the description of the present specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0054] References to "one embodiment" or "some embodiments" etc. described in the present specification mean that one or more embodiments of the present invention include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0055] It should be understood that the order of execution of the steps in the following embodiments does not imply a precedence of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0056] In order to illustrate the technical solution of the present invention, specific embodiments are provided below for illustration.

[0057] The experimental record data processing method provided by the embodiment of the present invention can be applied in the following aspects: Figure 1 In the experimental record data processing system shown, the experimental record data processing system includes a user terminal and an experimental record data processing device, wherein the user terminal communicates with the experimental record data processing device through a network.

[0058] When it is necessary to enter data of paper experiment record, the user sends a data entry instruction or an experiment record image to the experiment record data processing device through the user terminal. The experiment record image can be a picture obtained by image acquisition of the paper experiment record file; after receiving the data entry instruction, the experiment record data processing device can acquire the experiment record image by image acquisition of the paper experiment record file through the image acquisition device; after acquiring the experiment record image, the recognition module of the experiment record data processing device performs structured data recognition on the experiment record image to obtain the experiment record structured data. Afterwards, the entry module of the experiment record data processing device respectively enters the experiment record structured data based on at least two preset experiment record templates to generate at least two template entry data, wherein the experiment record template is a data template with a standard format established after analyzing and classifying the layout formats of multiple types of experiment record images in advance. After obtaining the template input data, the calculation module of the experiment record data processing device performs template matching calculation on each template input data and the experiment record template that generates the template input data based on the template attribute data of each experiment record template, and obtains the template matching degree of each template input data; finally, the storage module of the experiment record data processing device determines and stores the standard experiment record data of the experiment record image in all the template input data based on the template matching degree of each template input data.

[0059] In this embodiment, multiple experimental record templates are established by analyzing and classifying the layouts of different types of experimental record images in advance, and then structured data recognition is performed on the experimental record images, and the recognized experimental record structured data is entered into the experimental record template to generate template entry data, and then based on the template attribute data of the experimental record template, the template matching degree between the template entry data and the experimental record template that generates the template entry data is determined, and then the standard experimental record data of the experimental record image is determined and stored in all template entry data. The historical paper experimental records can be converted into experimental record images and data entry can be performed without the need for means entry, thereby realizing the automation of experimental record data entry and improving the entry efficiency; and through the pre-established experimental record template and selecting appropriate template entry data based on the template matching degree, as the standard experimental record data storage of the experimental record image, it can be ensured that the same type of experimental record images are entered into the system with the same layout, and the standardization of experimental record data entry can be achieved. Therefore, the experimental record data processing provided in this embodiment can quickly and accurately perform data entry on paper experimental record data.

[0060] In this embodiment, the experiment record data processing system includes an image acquisition device and an experiment record data processing device. This is only an exemplary description. In other embodiments, the experiment record data processing system also includes other things. For example, it may include an image acquisition device for acquiring images of paper experiment record files to obtain experiment record images; or it may include a data storage container (such as a database) for storing standard experiment record data of experiment record images.

[0061] The user terminal may be any type of terminal device, including but not limited to any personal computer, laptop computer, smart phone, tablet computer and portable wearable device. The experimental record data processing device may be a server, which may be implemented as an independent server or a server cluster composed of multiple servers.

[0062] In one embodiment, if Figure 2 As shown, a method for processing experimental record data is provided, and the method is applied in Figure 1 The experimental record data processing device in is taken as an example to illustrate, which includes the following steps:

[0063] S10: Perform structured data recognition on the experiment record image to obtain experiment record structured data.

[0064] When performing the experiment record data entry task, the experiment record data processing device needs an experiment record image, which can be an image obtained after image acquisition of a paper experiment record file. Among them, the experiment record image can be obtained by image acquisition of a paper experiment record file by an image acquisition device. Among them, the image acquisition device in this embodiment can be a high-definition camera, and by using a high-definition camera to acquire images of paper experiment record files, the clarity and image quality stability of the experiment record image can be improved, thereby improving the subsequent data recognition effect.

[0065] After obtaining the experimental record image, the experimental record data processing device needs to perform structured data recognition on the experimental record image to obtain the experimental record structured data of the experimental record image. It should be understood that the experimental record data is usually structured data arranged in a certain data format. For example, the experimental record data is usually set with standardized keys, such as the experimental name, experimental date, experimental process and other standardized keys, and the data content corresponding to each key. Therefore, it is necessary to use a pre-trained recognition model to perform structured data recognition on the experimental record image to obtain experimental record structured data. The experimental record structured data can be key-value pair data, that is, the experimental record structured data includes multiple key-value pairs, and each key-value pair includes a key and its corresponding value. For example, a key-value pair includes the experimental date (key): XX year XX month XX day (value corresponding to the key).

[0066] Specifically, the pre-trained recognition model can be an optical character recognition (OCR) model, which uses the pre-trained optical character recognition model to perform character recognition on the experimental record image to obtain experimental record recognition data of the experimental record image; then the experimental record recognition data is classified by key and value, each key and its corresponding value are determined, and key-value pair data including multiple key-value pairs are obtained, that is, experimental record structured data is obtained. In order to improve the accuracy of experimental record structured data, it is necessary to perform model training optimization on the OCR model based on the data structure analysis data of multiple experimental record templates by relevant personnel, so that the OCR model can quickly and accurately recognize different experimental record images and obtain experimental record structured data.

[0067] S20: Based on at least two preset experiment record templates, respectively input the experiment record structured data to generate at least two template input data.

[0068] After obtaining the experimental record structured data of the paper experimental record file, the experimental record data processing device needs to obtain at least two pre-established experimental record templates, wherein the experimental record template is a data template with a standard format established after analyzing and classifying the layout formats of multiple types of experimental record images in advance.

[0069] It is important to understand that the content of paper lab record files of different organizations (enterprises, scientific research institutions) will vary due to different types of experiments, but the layout of the same type of paper lab record files will not change easily. Therefore, they can be analyzed and classified according to the layout of different types of lab record images, and then a data template with a standard format for the same type of lab record data can be uniformly established as the lab record template for this type of lab record data. There can be one or more data templates with a standard format for the same type of experiment, that is, the layout of the lab record template for the same type of experiment may be different; the layout of different lab record templates is different, that is, the document layout of different lab record templates is different. In order to ensure the accuracy of lab data entry, all types of experiments and lab record image formats on the market can be investigated and counted, and corresponding lab record templates can be designed and stored for each format. The more formats of lab record templates maintained in the system, the more types of lab record images (paper lab record files) the system supports for entry.

[0070] After obtaining the experimental record template, the experimental record data processing device performs data entry on the experimental record structured data based on at least two preset experimental record templates to obtain at least two template entry data. That is, the experimental record structured data is filled into the experimental record template according to the layout of each experimental record template to generate the template entry data corresponding to each experimental record template. That is, the experimental record structured data in each template entry data is arranged and laid out according to the layout of its experimental record template.

[0071] S30: Based on the template attribute data of each experimental record template, a template matching calculation is performed between each template input data and the experimental record template that generates the template input data to obtain a template matching degree of each template input data.

[0072] After obtaining the template input data, the experimental record data processing device performs template matching calculation on the experimental record structured data in each template input data and the experimental record template that generates the template input data based on the template attribute data of each experimental record template, and obtains the template matching degree of each template input data. Among them, the template attribute data is used to characterize the attribute characteristics of the experimental record template, such as the frequency of use of the experimental record template by the user in the recent period. The template matching degree is used to characterize the matching degree between the experimental record structured data and the experimental record template it enters. The higher the template matching degree, the more accurate the experimental record template generated by the template input data is, and the better the page effect of the generated template input data is.

[0073] S40: Based on the template matching degree of each template input data, determine and store the standard experiment record data of the experiment record image in all the template input data.

[0074] After obtaining the template matching degree of each template input data, the experimental record data processing device determines and stores the standard experimental record data of the experimental record image in all the template input data based on the template matching degree of each template input data. For example, the template input data can be sorted in descending order according to the template matching degree of each template input data, and the template input data ranked first, that is, the template matching degree with the highest template is selected as the standard experimental record data of the experimental record image, and then stored in the database.

[0075] In summary, to complete the experimental record data entry task, you need to complete the following Figure 3 The business process shown:

[0076] 1. The experimental record templates are pre-established and maintained by relevant personnel (such as users / business system engineers), and the business system (i.e., the experimental record data processing device) obtains and stores multiple experimental record templates;

[0077] 2. Relevant personnel (users or OCR system engineers) analyze the experimental record templates, obtain the analysis data of each experimental record template and send it to the business system;

[0078] 3. The business system debugs the recognition algorithm of the recognition model (such as OCR recognition model or program) according to the analysis data of each experimental record template, so as to improve the recognition model to quickly and accurately recognize the data of different experimental record images;

[0079] 4. The user adopts and uploads the paper experiment record image (i.e. the experiment record image after the paper experiment record file is visualized) to the business system;

[0080] 5. The business system obtains the experimental record image and calls the recognition model to recognize the experimental record image;

[0081] 6. The recognition model recognizes the experimental record image in response to the call of the business system, and returns the experimental record structured data of the experimental record image to the business system;

[0082] 7. The business system enters the experimental record structured data into the experimental record template to obtain template entry data, and determines the template matching degree of the template entry data based on the template attribute data of the experimental record template, and then determines the standard experimental record data in all template entry data based on the template matching degree, that is, generates an electronic experimental record, and then stores it.

[0083] In this embodiment, multiple experimental record templates are established by analyzing and classifying the layouts of different types of experimental record images in advance, and then structured data recognition is performed on the experimental record images, and the recognized experimental record structured data is entered into the experimental record template to generate template entry data, and then based on the template attribute data of the experimental record template, the template matching degree between the template entry data and the experimental record template that generates the template entry data is determined, and then the standard experimental record data of the experimental record image is determined and stored in all template entry data. The historical paper experimental records can be converted into experimental record images and data entry can be performed without the need for means entry, thereby realizing the automation of experimental record data entry and improving the entry efficiency; and through the pre-established experimental record template and selecting appropriate template entry data based on the template matching degree, as the standard experimental record data storage of the experimental record image, it can be ensured that the same type of experimental record images are entered into the system with the same layout, and the standardization of experimental record data entry can be achieved. Therefore, the experimental record data processing provided in this embodiment can quickly and accurately perform data entry on paper experimental record data.

[0084] In one embodiment, in order to ensure the accuracy and efficiency of subsequent data entry, the experimental record data processing device can perform batch processing on the experimental record images (of the paper experimental record files). That is, the user sends a data entry instruction for the same batch of experimental record images to the experimental record data processing device. After receiving the batch experimental record data entry instruction, the experimental record data processing device starts and begins to execute steps S10 to S40 for each experimental record image of the same batch, thereby continuously processing the experimental record images of the same batch, further improving the data entry efficiency.

[0085] That is, after receiving the batch experiment record data entry instruction, the experiment record data processing device obtains each experiment record image in the same batch of experiment record images, and performs structured data recognition on the experiment record image to obtain experiment record structured data; based on at least two preset experiment record templates, the experiment record structured data is respectively entered to obtain at least two template entry data; based on the template attribute data of each experiment record template, the template matching calculation is performed on each template entry data and the experiment record template that generates the template entry data to obtain the template matching degree of each template entry data; based on the template matching degree of each template entry data, the standard experiment record data of the experiment record image is determined and stored in all template entry data.

[0086] Among them, the same batch of experimental record images are experimental record images (paper experimental record files) belonging to the same organization (enterprise or institution) or the same experimental type. Through continuous input processing of similar experimental record images, the template attribute data of each experimental record template can record the latest input situation to characterize the adopted template situation of the experimental record template in the recent time period, and then identify the experimental record template with a high frequency of adoption in the recent period, so as to automatically identify the template input data that may be relatively close as the standard experimental record data storage, which improves the efficiency and accuracy of data input.

[0087] In one embodiment, if Figure 4 As shown, in step S10, structured data recognition is performed on the experimental record image to obtain experimental record structured data, which specifically includes the following steps:

[0088] S11: using a pre-trained recognition model to perform character recognition on the experimental record image to obtain experimental record recognition data of the experimental record image;

[0089] S12: Based on the pre-generated common vocabulary for experiments, the keys in the experiment record recognition data are identified and corrected to obtain key-value pair data of the experiment record image, and output as experiment record structured data.

[0090] Among them, the experimental record information in the experimental record image is usually structured data that arranges information according to a certain layout format, that is, the data in the experimental record image includes fixed fields (keys) and data that changes corresponding to the fixed fields (values ​​corresponding to the keys). Since there are more data in the experimental records, and the vocabulary in the experimental records is usually rare, it is easy to identify errors when performing recognition. Therefore, the fixed fields in different types of experimental record images can be statistically analyzed in advance to form a common vocabulary library for experiments including fixed fields of different experimental record images. The fields in the common vocabulary library for experiments can also be expanded according to the fixed fields in the experimental record images, and the fixed fields (keys) of the experimental record images identified later can also be supplemented and imported into the common vocabulary library for experiments, thereby increasing the diversity of the common vocabulary library for experiments, and then increasing the accuracy of subsequent data recognition. Among them, the common vocabulary library for experiments includes vocabulary and symbols related to experimental types, experimental instruments, experimental materials, and basic experimental information. Among them, the basic experimental information includes information such as experimental date, experimental temperature, experimental humidity, and experimenters. The experimenter can be obtained by importing the user lists of each organization.

[0091] like Figure 5 As shown in the figure, an experimental record image of a certain type of experiment is shown. The fixed fields in the experimental record image include: basic experimental information (including experimental date, completion date, experimental temperature, experimental humidity, experimental batch number, experimenter, experimental purpose), instrument used (including instrument number, instrument name, instrument storage location, usage time), physical use (including inventory number, material name, material storage location, usage), and experimental process (including the first step of the experiment, the second step of the experiment... the Nth step of the experiment), etc. Statistical analysis of the fixed fields of different types of experimental record images can be obtained, including Figure 5 The experimental date's associated fields usually include year, month, day, and numbers, the experimental temperature usually includes symbols such as ℃, and the experimenter is usually the experimenter's name. Therefore, the "year", "month", "day" and numbers, as well as ℃ and the names of relevant experimenters are entered into the experimental common word library across fields to facilitate the subsequent correction of the recognition data and provide accuracy of the experimental record structured data.

[0092] When performing the task of entering experimental record data, the experimental record data processing device needs to use a pre-trained recognition model (such as optical character recognition) to locate and identify the key-value pairs of the experimental record image, and obtain the key-value pair data of the experimental record image, that is, use OCR technology to identify and locate each fixed field (key) in the experimental record image and the data corresponding to each fixed field (key value), and obtain key-value pair data including multiple key-value pairs, each key-value pair includes a key (i.e., a fixed field) and a key value (i.e., the data corresponding to the fixed field). Then, the experimental record data processing device performs field matching and correction on the key-value pair data based on a pre-generated common vocabulary for experiments, and obtains the experimental record structured data of the experimental record image.

[0093] For example, Figure 5 As shown, the experimental record image includes fixed fields such as experimental date, completion date, laboratory temperature, experimental humidity, experimental batch number, experimenter, experimental instrument, and materials used. The associated fields in the experimental date are usually texts in fixed fields such as year, month, day, and numbers. When performing OCR recognition on the experimental record image, the "date" in the experimental date may be recognized as the "date", and the "℃" in the experimental temperature may be recognized as the letter "C". Therefore, according to the experimental date and its associated words in the common experimental vocabulary, the "date" of the experimental date in the experimental record recognition data can be corrected to "date", and according to the experimental temperature and its associated words in the common experimental vocabulary, the letter "C" in the experimental temperature can be corrected to "℃", thereby obtaining more accurate experimental record structured data.

[0094] In this embodiment, optical character recognition technology is used to perform character recognition on the experiment record image to obtain experiment record recognition data of the experiment record image, and based on a pre-generated common vocabulary library for experiments, the keys in the experiment record recognition data are identified and corrected to obtain key-value pair data of the experiment record image, and output as experiment record structured data, which can improve the accuracy of the experiment record structured data, thereby improving the accuracy of subsequent data entry.

[0095] In one embodiment, if Figure 6 As shown, in step S20, based on at least two preset experiment record templates, the experiment record structured data is respectively inputted to generate at least two template input data, which specifically includes the following steps:

[0096] S21: According to the experimental record structured data, multiple experimental record templates are matched with preset fields, and at least two matching preliminary screening experimental record templates are preliminarily screened.

[0097] After obtaining the experimental record structured data, the experimental record data processing device needs to perform preset field matching on multiple experimental record templates according to the experimental record structured data, and preliminarily screen and obtain at least two matching preliminary screening experimental record templates.

[0098] Among them, the preset field can be a key in the experimental record structured data (that is, the experimental record file), such as fixed fields such as the experimental type and experimental time, and related experimental keywords. After obtaining the experimental record structured data, the experimental record structured data is subjected to key extraction to obtain at least two preset fields, and then each preset field is field matched with each experimental record template respectively, and the experimental record template that does not match the preset field is eliminated, and the experimental record template that matches the preset field is recorded as a preliminary screening experimental record template, and at least two preliminary screening experimental record templates are obtained.

[0099] In other embodiments, the preset field may also be an auxiliary identification field input for the same batch of experimental record images or for the experimental record images when the user sends the data entry instruction. The auxiliary identification field may include key words such as experimental type, experimental time, and experimental keywords. After obtaining the experimental record structured data, it is determined whether the same batch of experimental record images or the auxiliary identification field input for the experimental record images are received. If the auxiliary identification field is not received, the experimental record structured data is key extracted to obtain at least two preset fields, and then the preset field matching is performed on multiple experimental record templates, and at least two matching preliminary screening experimental record templates are obtained by preliminary screening. The data processing amount of subsequent data entry can be reduced. If the auxiliary identification field is received, the auxiliary identification field is used as the preset field, and the preset field matching is performed on multiple experimental record templates, and at least two matching preliminary screening experimental record templates are obtained by preliminary screening. Preliminary screening of multiple experimental record templates according to the field input by the user can reduce the field extraction process of the experimental record structured data and further reduce the data processing amount.

[0100] S22: Match the key in the structured data of the experimental record with the fields of each preliminary screening experimental record template and enter the data to generate at least two template entry data.

[0101] After obtaining at least two preliminary screening experiment record templates, the experiment record data processing device matches the key in the experiment record structured data with each preliminary screening experiment record template for field matching and data entry, and obtains at least two template entry data. Among them, the key is a fixed field (i.e., key) in the experiment record structured data. For each preliminary screening experiment record template, the key in the experiment record structured data is field matched with the preliminary screening experiment record template, and when the key is matched, the data corresponding to the key (i.e., the value corresponding to the key) is filled into the corresponding position of the matching field in the preliminary screening experiment record template, and all the keys in the experiment record structured data are traversed until the data is filled, and the template entry data is output. Among them, when performing key matching, a similarity matching method can be adopted, that is, the similarity between the key and the fixed field in the preliminary screening experiment record template is calculated, and the field with the highest similarity is used as the matching field.

[0102] In this embodiment, preset field matching is performed on multiple experimental record templates based on the experimental record structured data, and at least two matching preliminary screening experimental record templates are obtained through preliminary screening. Then, the key in the experimental record structured data is matched with the fields and data is entered into each preliminary screening experimental record template to produce at least two template entry data. The process of obtaining the template entry data is clarified, and the subsequent data processing volume is effectively reduced by preliminary screening of many templates, and the accuracy of the template entry data is improved.

[0103] In one embodiment, the template attribute data of the experiment record template includes user attribute data used to characterize the user's operation behavior characteristics on the experiment record template. Figure 7 As shown, in step S30, based on the template attribute data of each experimental record template, the template matching calculation is performed on each template input data and the experimental record template that generates the template input data to obtain the template matching degree of each template input data, which specifically includes the following steps:

[0104] S31: Determine the user attribute matching degree of the template entry data based on the user attribute data of the experiment record template used to generate the template entry data.

[0105] In the present embodiment, the template attribute data of the experiment record template includes user attribute data, and the user attribute data is used to characterize the user's operational behavior characteristics of the experiment record template. After obtaining the template entry data, it is necessary to determine each user attribute in the user attribute data of each experiment record template, and each user attribute is determined according to the user's operational behavior data of the experiment record template. The user can be an experimenter in the experiment record structured data (i.e., paper experiment record file), or a user who currently logs into the system to perform the experiment record entry task. Among them, the user attributes in the user attribute data include the current user's final total number of adoptions of the experiment record template (i.e., the current user selects the experiment record template for data entry and stores it as the total number of standard experiment record data), the number of times the current user creates and updates the experiment record template, and the current user adopts the total number of times the experiment record template is adopted in the most recent preset time period (such as the most recent month).

[0106] After determining each user attribute of each experimental record template, it is necessary to determine the user attribute matching degree of the template entry data based on the user attribute data of the experimental record template that generates the template entry data. Among them, the user attribute matching degree includes the matching degree of each user attribute in the user attribute data. That is, it is necessary to determine the matching degree of each user attribute of the template entry data and the experimental record template that generates the template entry data, and obtain the user attribute matching degree of the template entry data.

[0107] Among them, if the user attribute is non-numeric data, the user attribute of the experimental record template that generates the template input data is matched with the experimental record structured data. If they match, the first preset value (such as 1) is used as the matching degree between the template input data and the user attribute of the experimental record template; if they do not match, the second preset value (such as 0) is used as the matching degree between the template input data and the user attribute of the experimental record template. If the user attribute is numeric data, all experimental record templates are sorted in descending order based on the numerical value of the user attribute, and the matching degree between the template input data and the user attribute of each experimental record template is determined according to the sorting result. The higher the ranking, the greater the matching degree, and the lower the ranking, the smaller the matching degree or even 0. The first preset value is greater than the second preset value.

[0108] For example, based on the numerical value of the user attribute, all experimental record templates are sorted in descending order to determine the top N experimental record templates, and the matching degree between the template input data and the user attribute of the top N experimental record templates is recorded as 1, and the matching degree between the template input data and the user attribute of the remaining experimental record templates is recorded as 0.

[0109] S32: According to the user attribute matching degree of each template input data, a preset scoring model is used to perform template matching calculation on each template input data to obtain the template matching degree of each template input data.

[0110] After obtaining the user attribute matching degree of each template input data, a preset scoring model is used to perform template matching calculations on each template input data according to the user attribute matching degree of each template input data to obtain the template matching degree of each template input data.

[0111] Specifically, the preset scoring model includes a scoring value for each user attribute. According to the matching degree of the user attribute matching degree of the template input data with each user attribute and the scoring value of each user attribute, the matching score of each user attribute of the template input data is determined, and then the matching score of each user attribute is directly added to obtain a comprehensive matching score, that is, the user attribute matching score of the template input data is obtained.

[0112] For example, the user attribute data of the template input data includes user attribute A and user attribute B. The scoring value of user attribute A in the preset scoring model is S1, and the scoring value of user attribute B is S2. Then, the matching degree of user attribute A is multiplied by the scoring value S1 of user attribute A to obtain the matching score of user attribute A, and the matching degree of user attribute B is multiplied by the scoring value S2 of user attribute B to obtain the matching score of user attribute B; the matching score of user attribute A and the matching score of user attribute B are directly added to obtain the user attribute matching score of the template input data.

[0113] In this embodiment, based on the user attribute data of the experimental record template for generating template input data, the user attribute matching degree of the template input data is determined, and according to the user attribute matching degree of each template input data, the template matching degree of each template input data is calculated using a preset scoring model to obtain the template matching degree of each template input data, and the template matching degree acquisition process of each template input data is refined. Through the user attribute data table, the user's behavior data on the experimental record template is pointed out, and then based on the user attribute data of each experimental record template, the user attribute matching degree and template matching degree of the corresponding template input data are determined. Subsequently, the experimental record template with better matching degree with the user behavior can be selected as the data carrier of the paper experimental record file, which improves the accuracy of the input data, thereby improving the experimental record data entry effect.

[0114] In one embodiment, the template attribute data of the experiment record template also includes global attribute data for characterizing the global basic features of the experiment record template. In step S32, the template matching calculation is performed on each template input data using a preset scoring model to obtain the template matching degree of each template input data, which specifically includes the following steps:

[0115] S321: Determine the global attribute matching degree of the template input data based on the global attribute data of the experiment record template used to generate the template input data.

[0116] In this embodiment, the template attribute data of the experimental record template also includes global attribute data, which is used to characterize the global characteristics of the experimental record template. Among them, the global attributes of the local attribute data include the total number of times all users finally adopt the experimental record template, the frequency of adoption of the experimental record template by all users in the near future (such as the past week or month) (including the frequency of adoption and the number of adoptions), the creation time of the experimental record template, the update time of the experimental record template, the application time period of the experimental record template, the experimental type of the experimental record template and other attributes. Among them, the total number of times all users finally adopt the experimental record template refers to the total number of times all users use it to enter data to obtain standard experimental record data after the experimental record template is created. The application time period of the experimental record template refers to the application time period during which the experimental record template should be used; the experimental record template can be updated according to actual needs, and some experimental record templates whose application time does not include the experimental time can be filtered out according to the experimental time in the paper experimental record file.

[0117] After determining the user attribute matching degree of each template entry data, it is also necessary to determine the global attribute matching degree of the template entry data based on the global attribute data of the experiment record template that generates the template entry data. The global attribute matching degree includes the matching degree of each global attribute in the global attribute data, that is, determine the matching degree of each global attribute of the template entry data and the experiment record template that generates the template entry data, and obtain the global attribute matching degree of the template entry data.

[0118] Among them, if the global attribute is non-numeric data, such as the aforementioned creation time, update time, application time period, experiment type, etc., the template input data corresponding to the user attribute of the experiment record template is matched with the experiment record structured data. If it matches, the first preset value (such as 1) is used as the matching degree between the template input data and the global attribute of the experiment record template; if it does not match, the second preset value (such as 0) is used as the matching degree between the template input data and the global attribute of the experiment record template. The first preset value is greater than the second preset value.

[0119] If the global attribute is numerical data, such as the total number of adoptions and the frequency of adoptions, all experimental record templates are sorted in descending order based on the numerical value of the global attribute, and the matching degree between the template input data and the global attribute of each experimental record template is determined according to the sorting result. The higher the ranking, the greater the matching degree, and the lower the ranking, the smaller the matching degree or even 0. For example, all experimental record templates are sorted in descending order based on the numerical value of the global attribute, and the first N experimental record templates with the highest ranking are determined. The matching degree between the template input data and the global attribute of the first N experimental record templates is recorded as 1, and the matching degree between the template input data and the global attribute of the remaining other experimental record templates is recorded as 0.

[0120] S322: According to the user attribute matching degree and global attribute matching degree of each template input data, a preset scoring model is used to perform template matching calculation on each template input data to obtain the template matching degree of the template input data.

[0121] After obtaining the user attribute matching degree and global attribute matching degree of each template input data, the preset scoring model is used to calculate the template matching of each template input data according to the user attribute matching degree and global attribute matching degree of each template input data to obtain the template matching degree of the template input data.

[0122] In this embodiment, based on the global attribute data of the experimental record template for generating template input data, the global attribute matching degree of the template input data is determined, and then according to the user attribute matching degree and global attribute matching degree of each template input data, the template matching degree of each template input data is calculated using a preset scoring model to obtain the template matching degree of the template input data, and the template matching degree acquisition process of each template input data is further refined. According to the user attribute matching degree and global attribute matching degree of each template input data, the template matching degree of each template input data is determined, which further improves the accuracy of the template matching degree, thereby improving the accuracy of the input data, thereby improving the experimental record data input effect.

[0123] In other embodiments, the template attribute data also includes auxiliary attribute data, which are key attributes for template matching input by the user when creating or updating the experiment record template, such as experiment type (such as segmentation type), experiment keywords, and actual creation time, that is, the auxiliary attributes in the auxiliary attribute data include experiment type (such as segmentation type), experiment keywords, and actual creation time, etc.

[0124] When obtaining the user attribute matching degree and global attribute matching degree of each template input data, it is necessary to match the auxiliary attributes of the experimental record structured data and the experimental record template corresponding to the template input data one by one, determine the matching degree of each auxiliary attribute of the template input data and the corresponding experimental record template, and obtain the auxiliary attribute matching degree of the template input data, that is, the auxiliary attribute matching degree of the template input data includes the matching degree of each auxiliary attribute of the template input data and the corresponding experimental record template. Among them, if the experimental record structured data matches the auxiliary attribute of the corresponding experimental record template, the matching degree of the template input data and the auxiliary attribute is the first preset value; if the experimental record structured data does not match the auxiliary attribute of the corresponding experimental record template, the matching degree of the template input data and the auxiliary attribute is the second preset value.

[0125] Finally, according to the auxiliary attribute matching degree, user attribute matching degree and global attribute matching degree of each template input data, the preset scoring model is used to calculate the template matching degree of each template input data, and the template matching degree of the template input data is obtained. This further improves the accuracy of the template matching degree, and then improves the accuracy of the input data, thereby improving the data input effect of the experimental record.

[0126] In one embodiment, the preset scoring model includes the scoring value of each user attribute and the scoring value of each global attribute. In step S322, the template matching calculation of each template input data is performed using the preset scoring model according to the user attribute matching degree and the global attribute matching degree of each template input data to obtain the template matching degree of the template input data, which specifically includes the following steps:

[0127] S3221: Determine the user attribute matching score of the template input data according to the matching degree of each user attribute in the user attribute matching degree of the template input data and the scoring value of each user attribute.

[0128] In this embodiment, the preset scoring model includes a scoring value of each user attribute and a scoring value of each global attribute.

[0129] The user attribute matching score of the template input data is determined according to the matching degree of each user attribute in the user attribute matching degree of the template input data and the scoring value of each user attribute in the preset scoring model. That is, the matching degree of each user attribute in the user attribute matching degree can be directly multiplied by the scoring value of the corresponding user attribute to obtain the matching score of the user attribute in the template input data, and the matching scores of each user attribute in the template input data are added to obtain the user attribute matching score of the template input data.

[0130] Among them, the scoring values ​​of each user attribute can be the same or different, and the scoring values ​​of each user attribute can be adjusted in real time according to actual needs. Similarly, the scoring values ​​of each global attribute (and each auxiliary attribute) can be the same or different, and the scoring values ​​of each user attribute can be adjusted in real time according to actual needs.

[0131] For example, when the preset scoring model is built, the scoring values ​​of each user attribute can be the same. Subsequently, when the template input data is selected as the standard experimental record data, the scoring values ​​of each user attribute are adjusted in real time, thereby optimizing the preset scoring model and improving the accuracy of subsequent template matching.

[0132] In addition, different types of attributes have different scoring values. The scoring values ​​of each attribute of the preset scoring model include scoring levels and scoring values. The scoring level of each attribute is designed according to a preset step size. For example, the scoring values ​​of different types of attributes can be graded according to the preset step size to obtain the first level and the second level. If the preset step size can be 5, the scoring value in the first level can be 1-5 points, and the scoring value in the second level can be 6-10 points.

[0133] The scoring values ​​of all attributes are graded according to the importance of different attributes. The scoring levels of attributes of the same type can be different. The higher the scoring level, the greater the score value. When adjusting the scoring values ​​of each attribute later, keep the scoring level unchanged and adjust the score within the preset step size. For example, the scoring value of the experiment type in the global attributes can be 10 points in the second level, and the scoring value of the template creation time in the global attributes can be 5 points in the first level; when the scoring of the experiment type in the global attributes needs to be adjusted, the lowest score can be adjusted to 6 points in the second level.

[0134] S3222: Determine the global attribute matching score of the template input data according to the matching degree of each global attribute in the global attribute matching degree of the template input data and the scoring value of each global attribute.

[0135] The global attribute matching score of the template input data is determined according to the matching degree of each global attribute in the global attribute matching degree of the template input data and the scoring value of each global attribute in the preset scoring model. That is, the matching degree of each global attribute in the global attribute matching degree can be directly multiplied by the scoring value of the corresponding global attribute to obtain the matching score of the global attribute in the template input data, and the matching scores of each global attribute in the template input data are added to obtain the global attribute matching score of the template input data.

[0136] S3223: Determine the template matching degree of each template input data according to the user attribute matching score and the global attribute matching score of each template input data.

[0137] After obtaining the user attribute matching score and the global attribute matching score, the template matching degree of each template input data is determined according to the user attribute matching score and the global attribute matching score of each template input data. For example, the user attribute matching score and the global attribute matching score of the template input data can be directly added to obtain the template matching degree of the template input data, which is simple and convenient.

[0138] Among them, the preset scoring model can also include the scoring value of each auxiliary attribute in the auxiliary attribute data; after obtaining the auxiliary attribute matching degree of each template input data, the auxiliary attribute matching score of the template input data is determined according to the matching degree of each auxiliary attribute in the auxiliary attribute matching degree and the scoring value of each auxiliary attribute, and the calculation process participates in the calculation process of the user attribute matching score; then, according to the user attribute matching score, global attribute matching score and auxiliary attribute matching score of each template input data, the template matching degree of each template input data is determined accordingly, which is more accurate.

[0139] In other embodiments, the preset scoring model also includes the score rate of each attribute, that is, the score rate of the user attribute, the score rate of the global attribute (and the score rate of the auxiliary attribute). When determining the template matching degree of each template input data, it is necessary to use the score rate of each attribute as a weight, and perform weighted summation on the user attribute matching score and the global attribute matching score (and the auxiliary attribute matching) of the template input data to obtain the template matching degree of the template input data.

[0140] Among them, the importance of each type of attribute is different, and the score rate of each type of attribute is different. For example, if the user attribute is more in line with the current user's behavior data and preferences, the user attribute has the highest score rate. Among them, the score rate of the user attribute is greater than the score rate of the auxiliary attribute and the score rate of the global attribute, and the score rate of the auxiliary attribute is greater than the score rate of the global attribute. For example, the score rate of the user attribute can be 1.5, the score rate of the global attribute can be 0.8, and the score rate of the auxiliary attribute can be 1. In this embodiment, the weighted summation of the attribute matching scores is performed according to the score rate of each type of attribute, which can increase the emphasis on a certain type of attribute and improve the accuracy of template matching.

[0141] In this embodiment, the user attribute matching score of the template input data is determined according to the matching degree of each user attribute in the user attribute matching degree of the template input data and the scoring value of each user attribute, and the global attribute matching score of the template input data is determined according to the matching degree of each global attribute in the global attribute matching degree of the template input data and the scoring value of each global attribute. After accommodating, the template matching degree of each template input data is correspondingly determined according to the user attribute matching score and the global attribute matching score of each template input data, and the template matching degree acquisition process of the template input data is clarified. By setting scoring values ​​for different attributes by presetting the scoring model, it is convenient to quickly calculate the template matching degree according to the matching score of each attribute, thereby improving the accuracy of the template matching degree.

[0142] In one embodiment, if Figure 8 As shown, in step S40, based on the template matching degree of each template input data, the standard experimental record data of the experimental record image is determined and stored in all the template input data, which specifically includes the following steps:

[0143] S41: sorting the template input data based on the template matching degree of each template input data to generate recommended input data;

[0144] S42: Push the recommended entry data to the user, and store the template entry data selected by the user in the recommended entry data as standard experiment record data of the experiment record image.

[0145] After obtaining the template matching degree based on each template input data, each template input data can be sorted (such as descending order) based on the template matching degree of each template input data to generate recommended input data. Each template input data in the recommended input data is sorted from high to low according to the template matching degree. Then, the recommended input data is pushed to the user, and the template input data selected by the user in the recommended input data is stored as the standard experimental record data of the experimental record image.

[0146] In addition, each template input data can also be screened according to the preset matching degree, and the template input data with a template matching degree less than the preset matching degree can be eliminated. Then, the remaining template input data cases are sorted from high to low in terms of template matching degree to obtain recommended input data and improve the recommendation effect.

[0147] In this embodiment, based on the template matching degree of each template input data, each template input data is sorted to generate recommended input data, and then the recommended input data is pushed to the user, and the template input data selected by the user in the recommended input data is stored as the standard experimental record data of the experimental record image, and the template input data selected by the user is stored as the standard experimental record data of the experimental record image, which has higher accuracy.

[0148] In one embodiment, after step S42, that is, after storing the template entry data selected by the user in the recommended entry data as the standard experiment record data of the experiment record image, the following steps are specifically included:

[0149] S43: updating the template attribute data of the corresponding experiment record template according to the template input data selected by the user; and / or,

[0150] S44: updating and optimizing the preset scoring model according to the template input data selected by the user, so as to facilitate the subsequent calculation of template matching between each template input data and the corresponding experimental record template based on the optimized preset scoring model.

[0151] When storing the template entry data selected by the user in the recommended entry data as the standard experimental record data of the experimental record image, it is necessary to update the template attribute data of the experimental record template corresponding to the template entry data according to the template entry data selected by the user. Timely updating of template attribute data can improve the accuracy of template matching scores during subsequent data entry, thereby improving the accuracy and effect of data entry. For example, determine the experimental record template corresponding to the template entry data selected by the user, and then update the attributes related to user behavior in the template attribute data of the experimental record template, such as updating the total number of final adoptions of the experimental record template by the current user in the user attributes, the total number of adoptions of the experimental record template by the current user in the most recent preset time period, and the total number of final adoptions of the experimental record template by all users in the global attributes, and the recent (adoption frequency of the experimental record template by all users).

[0152] Alternatively, the preset scoring model is updated and optimized according to the template input data selected by the user, so as to facilitate the subsequent calculation of template matching between each template input data and the corresponding experimental record template based on the optimized preset scoring model. Among them, the template attribute matching score is compared based on the template input data selected by the user and the template input data not selected by the user, so as to update and optimize the score value of the corresponding attribute in the preset scoring model, thereby dynamically adjusting the score value of each attribute in the preset scoring model according to the user's selection, improving the accuracy of the score value of each attribute, and thus improving the accuracy of the template matching degree calculated by the preset scoring model in the future.

[0153] Specifically, when the template entry data selected by the user multiple times are not the template entry data with the highest template attribute matching score, it is necessary to compare the matching scores of each attribute (at least user attributes and global attributes) in the template entry data selected by the user, and the matching scores of each attribute in the template entry data with the highest template attribute matching score, determine the global attributes and user attributes that are different, and then adjust (increase or decrease) the score values ​​of the global attributes and user attributes that are different. The adjustment range cannot exceed the scoring level of the attribute. After the adjustment is completed, whether the template entry data selected by the user has the highest template attribute matching score among all template entry data.

[0154] For example, by comparing the template entry data A selected by the user with the template entry data B with the highest template attribute matching score, it is found that the matching score of the global attribute experiment type in data B is greater than the matching score of the global attribute experiment type in data A. At this time, it is necessary to adjust the preset scoring model to adjust the scoring value of the global attribute experiment type, and recalculate the template attribute matching scores in all template entry data until the template entry data selected by the user has the highest template attribute matching score among all template entry data.

[0155] When adjusting the score of each attribute, the score level is kept unchanged and the score is adjusted within the preset step size. For example, the score of the current experiment type in the global attribute is 10 points in the second level. When the score of the experiment type in the global attribute needs to be adjusted, the lowest score can be adjusted to 6 points in the second level.

[0156] In one embodiment, after the template entry data selected by the user in the recommended entry data is stored as the standard experimental record data of the experimental record image, the template attribute data of the corresponding experimental record template can also be updated according to the template entry data selected by the user. At the same time, the preset scoring model needs to be updated and optimized according to the template entry data selected by the user, so as to facilitate the subsequent calculation of template matching between each template entry data and the corresponding experimental record template based on the optimized preset scoring model. The template attribute data and the preset scoring model are optimized at the same time, which improves the accuracy of subsequent data entry.

[0157] In this embodiment, the template attribute data of the corresponding experimental record template is updated according to the template input data selected by the user, and the template attribute data is updated in time, which can improve the accuracy of the template matching score during subsequent data entry. According to the template input data selected by the user, the preset scoring model is updated and optimized, so as to facilitate the subsequent calculation of the template matching between each template input data and the corresponding experimental record template based on the optimized preset scoring model; the score value of each attribute in the preset scoring model is dynamically adjusted according to the user's selection, so as to improve the accuracy of the score value of each attribute, thereby improving the accuracy of the template matching degree calculated by the preset scoring model in the future.

[0158] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0159] In one embodiment, a device for processing experimental record data is provided, and the device for processing experimental record data corresponds to the method for processing experimental record data in the above embodiment. Fig. 9 As shown, the experimental record data processing device includes an identification module 901, an input module 902, a calculation module 903 and a storage module 904. The detailed description of each functional module is as follows:

[0160] The recognition module 901 is used to perform structured data recognition on the experimental record image to obtain the experimental record structured data;

[0161] An input module 902 is used to input the experimental record structured data based on at least two preset experimental record templates, respectively, to generate at least two template input data, wherein the experimental record template is a standard template established after analyzing and classifying the layout of multiple types of experimental record images in advance;

[0162] The calculation module 903 is used to calculate the template matching of each template input data and the experiment record template that generates the template input data based on the template attribute data of each experiment record template, so as to obtain the template matching degree of each template input data;

[0163] The storage module 904 is used to determine and store the standard experiment record data of the experiment record image in all the template entry data based on the template matching degree of each template entry data.

[0164] Optionally, the template attribute data includes user attribute data for characterizing the user's operation behavior characteristics on the experiment record template, and the calculation module 903 is specifically used to:

[0165] Determining a user attribute matching degree of the template entry data based on the user attribute data of the experiment record template for generating the template entry data;

[0166] According to the user attribute matching degree of each template input data, the preset scoring model is used to calculate the template matching of each template input data to obtain the template matching degree of each template input data.

[0167] Optionally, the template attribute data also includes global attribute data, and the calculation module 903 is further used to:

[0168] Determining a global attribute matching degree of the template entry data based on the global attribute data of the experiment record template for generating the template entry data;

[0169] According to the user attribute data and global attribute data of each experimental record template, the preset scoring model is used to calculate the template matching of each template input data to obtain the template matching degree of the template input data.

[0170] Optionally, the input module 902 is specifically used for:

[0171] According to the structured data of the experimental record, multiple experimental record templates are matched with preset fields, and at least two matching preliminary screening experimental record templates are obtained through preliminary screening;

[0172] The key in the structured data of the experimental record is matched with the fields of each preliminary screening experimental record template and the data is entered to obtain at least two template entry data.

[0173] Optionally, the identification module 901 is specifically used for:

[0174] Using optical character recognition technology to locate and identify key-value pairs of the experimental record image, and obtaining key-value pair data of the experimental record image;

[0175] Based on a pre-generated common vocabulary for experiments, field matching and correction are performed on the key-value pair data to obtain the experimental record structured data of the experimental record image.

[0176] Optionally, the storage module 904 is specifically used for:

[0177] Based on the template matching degree of each template input data, the template input data is sorted and processed to generate recommended input data;

[0178] The recommended entry data is pushed to the user, and the template entry data selected by the user in the recommended entry data is stored as standard experiment record data of the experiment record image.

[0179] Optionally, the experimental record data processing device further includes an updating module 905, which stores the template input data selected by the user in the recommended input data as the standard experimental record data of the experimental record image, and then the updating module 905 is used to:

[0180] Update the template attribute data of the corresponding experiment record template according to the template input data selected by the user; and / or,

[0181] According to the template input data selected by the user, the preset scoring model is updated and optimized, so as to facilitate the subsequent template matching calculation of each template input data and the corresponding experimental record template based on the optimized preset scoring model.

[0182] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / units are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0183] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0184] The present application also provides a computer device, such as Fig.10 As shown, the computer device includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor implements the steps in any of the above-mentioned method embodiments when executing the computer program, or implements the functions of the modules / units in the above-mentioned device embodiments when executing the computer program.

[0185] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present application. The one or more modules / units may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program in the computer device.

[0186] Those skilled in the art will understand that Fig.10 These are merely examples of the computer device and do not constitute a limitation of the computer device. The computer device may include more or fewer components than those shown in the figure, or a combination of certain components, or different components. For example, the computer device may also include input and output devices, network access devices, buses, etc.

[0187] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0188] The memory may be an internal storage unit of the computer device, such as a hard disk or memory of the computer device. The memory may also be an external storage device of the computer device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the computer device. Furthermore, the memory may also include both an internal storage unit of the computer device and an external storage device.

[0189] An embodiment of the present application further provides a readable storage medium, wherein the readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments can be implemented.

[0190] An embodiment of the present application provides a computer program product. When the computer program product runs on a mobile terminal, the mobile terminal can implement the steps in the above-mentioned method embodiments when executing the computer program product.

[0191] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium can at least include: any entity or device that can carry the computer program code to the camera / terminal device, recording medium, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, RandomAccess Memory), electric carrier signal, telecommunication signal and software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electric carrier signals and telecommunication signals.

[0192] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0193] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0194] In the embodiments provided in the present application, it should be understood that the disclosed devices / equipment and methods can be implemented in other ways. For example, the device / equipment embodiments described above are merely schematic, for example, the division of the modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0195] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0196] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for processing experimental record data, characterized in that: include: Performing structured data recognition on the experimental record image to obtain experimental record structured data; Based on at least two preset experiment record templates, data entry is performed on the experiment record structured data respectively to generate at least two template entry data, wherein the experiment record template is a standard template established after analyzing and classifying the layout panels of multiple types of the experiment record images in advance; Based on the template attribute data of each of the experiment record templates, a template matching calculation is performed on each of the template input data and the experiment record template that generates the template input data to obtain a template matching degree of each of the template input data; Based on the template matching degree of each of the template entry data, the standard experiment record data of the experiment record image is determined and stored in all the template entry data.

2. The experimental record data processing method according to claim 1, characterized in that: The template attribute data includes user attribute data used to characterize the user's operation behavior characteristics on the experiment record template. The template attribute data based on each of the experiment record templates is used to calculate the template matching of each of the template input data and the experiment record template that generates the template input data, and obtain the template matching degree of each of the template input data, including: Determining a user attribute matching degree of the template entry data based on the user attribute data of the experiment record template that generates the template entry data; According to the user attribute matching degree of each template input data, a preset scoring model is used to perform template matching calculation on each template input data to obtain the template matching degree of each template input data.

3. The experimental record data processing method according to claim 2, characterized in that: The template attribute data also includes global attribute data. According to the user attribute matching degree of each template input data, a preset scoring model is used to perform template matching calculation on each template input data to obtain the template matching degree of each template input data, including: Determining a global attribute matching degree of the template entry data based on the global attribute data of the experiment record template for generating the template entry data; According to the user attribute matching degree and the global attribute matching degree of each template input data, the preset scoring model is used to perform template matching calculation on each template input data to obtain the template matching degree of the template input data.

4. The experimental record data processing method according to claim 1, characterized in that: The step of inputting data into the experimental record structured data based on at least two preset experimental record templates to generate at least two template input data includes: According to the experimental record structured data, a plurality of the experimental record templates are matched with preset fields, and at least two matching preliminary screening experimental record templates are preliminarily screened and obtained; The key in the structured data of the experimental record is matched with each of the preliminary screening experimental record templates for field matching and data entry, and at least two template entry data are generated.

5. The experimental record data processing method according to claim 1, characterized in that: The step of performing structured data recognition on the experimental record image to obtain the experimental record structured data includes: Using optical character recognition technology to locate and identify key-value pairs of the experimental record image to obtain key-value pair data of the experimental record image; Based on a pre-generated common word library for experiments, field matching and correction are performed on the key-value pair data to obtain the experiment record structured data of the experiment record image.

6. The experimental record data processing method according to any one of claims 1 to 5, characterized in that: The step of determining and storing the standard experimental record data of the experimental record image in all the template entry data based on the template matching degree of each template entry data comprises: Based on the template matching degree of each template input data, sorting the template input data to generate recommended input data; The recommended entry data is pushed to the user, and the template entry data selected by the user in the recommended entry data is stored as standard experiment record data of the experiment record image.

7. The experimental record data processing method according to claim 6, characterized in that: After storing the template entry data selected by the user in the recommended entry data as standard experiment record data of the experiment record image, the method further includes: According to the template input data selected by the user, updating the template attribute data corresponding to the experiment record template; and / or, The preset scoring model is updated and optimized according to the template input data selected by the user, so as to facilitate the subsequent template matching calculation between each template input data and the corresponding experiment record template based on the optimized preset scoring model.

8. An experimental record data processing system, characterized in that: The device comprises an experimental record data processing device, wherein the experimental record data processing device comprises: A recognition module is used to perform structured data recognition on the experimental record image to obtain experimental record structured data; An input module, used for inputting data of the experimental record structured data respectively based on at least two preset experimental record templates to generate at least two template input data, wherein the experimental record template is a standard template established after analyzing and classifying the layout panels of multiple types of experimental record images in advance; A calculation module, for performing template matching calculation on each of the template input data and the experiment record template for generating the template input data based on the template attribute data of each of the experiment record templates, to obtain the template matching degree of each of the template input data; The storage module is used to determine and store the standard experiment record data of the experiment record image in all the template entry data based on the template matching degree of each template entry data.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the experimental record data processing method as described in any one of claims 1 to 7 are implemented.

10. A readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the experimental record data processing method as described in any one of claims 1 to 7 are implemented.

Images