A form generation method, apparatus, electronic device, and storage medium

By optimizing the form generation process through meta-learning algorithms and RFID technology, the problems of low flexibility and efficiency in form generation in existing technologies are solved, and efficient and accurate form generation that can adapt to different business scenarios is achieved.

CN117540719BActive Publication Date: 2026-11-17WEAVER NETWORK TECH CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202311697638.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-12-11
Publication Date
2026-11-17
Estimated Expiration
2043-12-11

AI Technical Summary

Technical Problem

Existing technologies for generating forms have poor flexibility, low efficiency in manual processing, and are difficult to meet the needs of different business scenarios.

Method used

A meta-learning algorithm model is used to learn from sample forms to generate form templates that adapt to the needs of new scenarios. RFID technology is used to improve data identification and security, and structured feature vectors and analysis algorithms are combined to optimize the form generation process.

Benefits of technology

It improves the flexibility and efficiency of form template generation, enhances the accuracy and security of form data, and reduces human error.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117540719B_ABST
    Figure CN117540719B_ABST
Patent Text Reader

Abstract

This application provides a form generation method, apparatus, electronic device, and storage medium. The method includes: obtaining data to be filled and the data type of the data to be filled; matching a target form template in a preset form template library according to the data type; the form template library includes at least one form template; the form template is obtained by inputting a sample form into a meta-learning algorithm model and learning from the sample form using the meta-learning algorithm model; and writing the data to be filled into the target form template to generate form data. Matching a target form template in the preset form template library according to the data type, and the form templates in the form template library are obtained by learning from the sample form using a meta-learning algorithm model, the meta-learning algorithm model can learn from existing sample forms to create form templates that meet the needs of new scenarios, realizing scenario-based dynamic form generation, improving the flexibility of form templates and the efficiency of form generation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and more specifically, to a form generation method, apparatus, electronic device, and storage medium. Background Technology

[0002] Forms provide an interactive method that allows users to enter and submit data in applications or websites. Currently, numerous businesses across various industries rely heavily on forms. Different business needs require the creation and production of forms. Properly designed forms allow users to clearly and conveniently provide the information they need, improving user experience and interface friendliness. Existing form generation methods typically involve manually designing form content based on business processes, or generating new forms by combining form fields and layout rules, followed by manual filling. Because this relies on fixed rules and templates, form generation lacks flexibility, and manual filling is inefficient. Summary of the Invention

[0003] The purpose of this application is to provide a form generation method, apparatus, electronic device, and storage medium. According to the data type, a target form template is matched in a preset form template library. The form templates in the form template library are obtained by learning from sample forms using a meta-learning algorithm model. The meta-learning algorithm model can learn how to create form templates that meet the needs of new scenarios from existing sample forms, thereby realizing scenario-based dynamic form generation, improving the flexibility of form templates, and increasing the efficiency of form generation.

[0004] In a first aspect, embodiments of this application provide a form generation method, comprising: obtaining data to be filled and the data type of the data to be filled; matching a target form template in a preset form template library according to the data type; the form template library includes at least one form template; the form template is obtained by inputting a sample form into a meta-learning algorithm model and learning the sample form using the meta-learning algorithm model; and writing the data to be filled into the target form template to generate form data.

[0005] In the above implementation process, according to the data type, the target form template is matched in the preset form template library. The form templates in the form template library are obtained by learning from the sample forms using a meta-learning algorithm model. The meta-learning algorithm model can learn from the existing sample forms how to create form templates that meet the needs of new scenarios, thereby realizing scenario-based dynamic form generation, improving the flexibility of form templates and the efficiency of form generation.

[0006] Optionally, in this embodiment of the application, before matching the target form template in the preset form template library according to the data type, the method further includes: obtaining a sample form; performing data preprocessing on the sample form to obtain a training form; parsing and encoding the fields in the training form to obtain a structured feature vector; and using the structured feature vector to train a preset meta-model to obtain a meta-learning algorithm model.

[0007] In the above implementation process, the meta-learning algorithm can learn knowledge and experience from multiple similar tasks, enabling it to adapt and learn more quickly when faced with new tasks. Furthermore, the meta-learning algorithm can be quickly trained and adjusted on a small amount of sample form data, thereby obtaining more form templates that can adapt to multiple application scenarios and improving the efficiency of obtaining form templates.

[0008] Optionally, in this embodiment of the application, training a preset meta-model using structured feature vectors to obtain a meta-learning algorithm model includes: converting the structured feature vectors into a meta-learning task; obtaining model optimization parameters through an optimization algorithm based on the meta-learning task and a predefined loss function; and obtaining the meta-learning algorithm model based on the model optimization parameters.

[0009] In the above implementation process, the metamodel can provide personalized model optimization for different tasks by learning the commonalities and differences of sample forms. By quickly adjusting the model parameters and the expression of form template fields, it can improve applicability and performance and generate more accurate form templates.

[0010] Optionally, in this embodiment, the loss function includes:

[0011] L=∑|G i -S i |

[0012] Where L is the loss function, i is the target field, and G is the target field. i For the representation of the target field in the form template, S i This is how the target field is represented in the sample form.

[0013] In the above implementation process, the meta-learning algorithm model will adjust its internal model parameters to minimize the loss function, that is, reduce the difference between the sample form and the generated form template, so as to improve the accuracy of the meta-learning algorithm model and thus generate a more accurate form template.

[0014] Optionally, in this embodiment of the application, obtaining the data to be filled includes: acquiring the data to be processed and the analysis type; extracting features from the data to be processed and filtering the extracted features to obtain data features; and using the analysis algorithm corresponding to the analysis type to perform data analysis on the data features to obtain the data to be filled.

[0015] In the above implementation process, if the data to be filled cannot be directly obtained from the data to be processed, appropriate data analysis can be performed on the data to be processed to obtain the data to be filled that meets the form filling requirements, so as to generate form data and improve the accuracy of the generated form data.

[0016] Optionally, in this embodiment, the analysis type includes combination analysis, cluster analysis, sales trend analysis, and / or demand analysis; the analysis algorithm corresponding to the analysis type is used to analyze the data features to obtain the data to be filled, including: if the analysis type is combination analysis, then the association rule learning algorithm is used to analyze the data features to obtain data association results; if the analysis type is cluster analysis, then the cluster analysis algorithm is used to analyze the data features to obtain data classification results; if the analysis type is sales trend analysis, then the time series analysis algorithm is used to analyze the data features to obtain product sales trend results; if the analysis type is demand analysis, then the supervised learning algorithm is used to analyze the data features to obtain product demand prediction results; the data to be filled includes data association results, data classification results, product sales trend results, and / or product demand prediction results.

[0017] In the above implementation process, based on the selected data features, the data features are analyzed using the analysis algorithm corresponding to the analysis type. The analysis algorithm can be applied to different types of data and application scenarios to obtain the data to be filled in that meets the form requirements, improve the accuracy of data processing, and increase the speed of generating form data.

[0018] Optionally, in this embodiment of the application, obtaining the data to be processed includes: using an RFID reader to identify RFID tags via radio frequency signals to obtain the data to be processed; the RFID tags are obtained by pre-programming the product data.

[0019] In the above implementation process, wireless communication between electronic tags (RFID tags) and readers (RFID readers) eliminates the need for manual data reading, improving data storage and management efficiency. It also significantly increases identification speed and accuracy, reducing human error. Applying RFID technology to form data generation schemes leverages its security controls to enhance data security and anti-counterfeiting capabilities.

[0020] Secondly, embodiments of this application also provide a form generation apparatus, comprising: obtaining data to be filled and the data type of the data to be filled; matching a target form template in a preset form template library according to the data type; the form template library includes at least one form template; the form template is obtained by inputting a sample form into a meta-learning algorithm model and learning the sample form using the meta-learning algorithm model; and writing the data to be filled into the target form template to generate form data.

[0021] Thirdly, embodiments of this application also provide an electronic device, including: a processor and a memory, the memory storing machine-readable instructions executable by the processor, which, when executed by the processor, perform the method described above.

[0022] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program that, when executed by a processor, performs the methods described above.

[0023] The form generation method, apparatus, electronic device, and storage medium provided in this application match a target form template in a preset form template library according to the data type. The form templates in the form template library are obtained by learning from sample forms using a meta-learning algorithm model. The meta-learning algorithm model can learn how to create form templates that meet the needs of new scenarios from existing sample forms, thereby realizing scenario-based dynamic form generation, improving the flexibility of form templates and the efficiency of form generation. Attached Figure Description

[0024] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 A flowchart illustrating a form generation method provided in an embodiment of this application;

[0026] Figure 2 This is a schematic diagram of the structure of the form generation device provided in the embodiments of this application;

[0027] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0028] The embodiments of the technical solution of this application will now be described in detail with reference to the accompanying drawings. These embodiments are only used to more clearly illustrate the technical solution of this application and are therefore merely examples, and should not be used to limit the scope of protection of this application.

[0029] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this application.

[0030] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.

[0031] Please see Figure 1 The illustrated diagram shows a flowchart of a form generation method provided in an embodiment of this application. The form generation method provided in this application can be applied to electronic devices, which may include a terminal and a server; wherein the terminal may specifically be a smartphone, tablet computer, computer, personal digital assistant (PDA), etc.; the server may specifically be an application server or a web server. The form generation method may include:

[0032] Step S110: Obtain the data to be filled and the data type of the data to be filled.

[0033] Step S120: Match the target form template in the preset form template library according to the data type; the form template library includes at least one form template; the form template is obtained by inputting the sample form into the meta-learning algorithm model and learning the sample form using the meta-learning algorithm model.

[0034] Step S130: Write the data to be filled into the target form template to generate form data.

[0035] In step S110, the data to be filled in refers to the data that needs to be entered into the form, that is, the data recorded by the form in various business applications. The data to be filled in has different data types, and the criteria for classifying data types can be set according to actual needs. For example, the data types of the data to be filled in can include: various types of collected data, combined analysis data, and cluster analysis data, etc. Collected data can be further subdivided into user surveys, market research, user feedback, etc.; combined analysis data can be the results obtained after analyzing the correlation between products; cluster analysis data can be the results obtained after classifying products.

[0036] In step S120, a "form template" refers to a structured framework that includes fields and layouts that need to be filled in, but has not yet been filled with specific data. These templates provide users with a basic form structure, upon which users can further fill in specific information to generate the final form.

[0037] The database is pre-stored the correspondence between the data to be filled in for each data type and at least one form template in the form template library. After obtaining the data type, the target form template that matches it can be determined in the form template library based on the data type.

[0038] For example, if the data to be filled in is a user survey, you can search the form template library for a form template suitable for filling in user survey data and use that form template as the target form template; if the data to be filled in is user survey cluster analysis data, you can search the form template library for a form template suitable for filling in categorical data and use that form template as the target form template.

[0039] The form template library includes at least one form template, meaning it includes a form template corresponding to each data type. These form templates are obtained by learning from sample forms using a meta-learning algorithm model. The meta-learning algorithm model learns from a limited number of provided sample forms and generates form templates that can adapt to new scenarios. The input to the meta-learning algorithm model is the sample forms in the sample form library, and the processing steps involve training the model using the meta-learning algorithm to generate new form templates.

[0040] The following example illustrates the process of generating a form template. A sample form for mobile phone sales is retrieved from the sample form library. First, the meta-learning algorithm model extracts key information from the sample form, such as field names, order, and required fields. This key information is then transformed into a generic sales template.

[0041] When we need to create a new TV sales form, we can adjust the general sales template based on the specific fields in TV sales (such as screen size, resolution, etc.) to generate a TV sales form template, which is then stored in the form template library. If the data to be filled in is TV sales data, we can retrieve the TV sales form template from the form template library as the target form template.

[0042] In step S130, after determining the target form template, the data to be filled in is written into the target form template to generate form data.

[0043] In the above implementation process, according to the data type, the target form template is matched in the preset form template library. The form templates in the form template library are obtained by learning from the sample forms using a meta-learning algorithm model. The meta-learning algorithm model can learn from the existing sample forms how to create form templates that meet the needs of new scenarios, thereby realizing scenario-based dynamic form generation, improving the flexibility of form templates and the efficiency of form generation.

[0044] In an optional embodiment, the generated form template can be validated to verify whether it meets the preset requirements. If it does, the form template is output; if it does not meet the preset requirements, the form template is adjusted using a meta-learning algorithm model and regenerated.

[0045] The following describes the process of validating a form template. Automated scripts can be written to check whether each field in the template conforms to the expected format and type. For example, if a field should be in date format, the validation process will check whether that field only allows date input.

[0046] You can also populate the template with real or simulated data to check if the data can be entered correctly and whether the form layout can present this information reasonably.

[0047] The generated templates can also be manually reviewed to ensure that all key fields are included and that their layout, sorting, and data requirements meet the sales needs of specific product types.

[0048] It can also simulate real-world business scenarios in a simulation environment to test the performance of form templates in actual operation and verify their feasibility and stability.

[0049] Optionally, in this embodiment of the application, before matching the target form template in the preset form template library according to the data type, the method further includes: obtaining a sample form; performing data preprocessing on the sample form to obtain a training form; parsing and encoding the fields in the training form to obtain a structured feature vector; and using the structured feature vector to train a preset meta-model to obtain a meta-learning algorithm model.

[0050] In the specific implementation process: the sample form library stores sample forms. The sample form library can use a database or file system to implement the storage function. The sample forms can be various types of forms collected in advance.

[0051] Preprocess the sample form to obtain the training form. This may include cleaning the fields or data in the sample form, standardizing the data format, removing noise, handling missing values, and other steps to ensure data quality and consistency.

[0052] Obtain the training form. Parse and encode the fields in the training form, performing data analysis based on the characteristics of the field types. For text fields, natural language processing techniques such as word segmentation, stemming, and stop word removal can be used to convert the text into words or phrases. For time fields, they can be broken down into components such as year, month, and day.

[0053] Based on the type and meaning of the fields, an appropriate encoding method is selected to transform the parsed data into a feature vector. Encoding methods can include binary encoding, one-hot encoding, bag-of-words model, and numerical encoding, etc., to obtain a structured feature vector. Each field in the form can generate a corresponding structured feature vector, which will represent various attributes of the field, such as field name, data type, role (e.g., title, description, input), and its relationship with other fields.

[0054] Meta-learning algorithms are used to train data and obtain a meta-learning model. Meta-learning is a class of machine learning algorithms whose goal is to enable the model to learn how to learn. It focuses on how to quickly adapt and learn across different tasks. Using structured feature vectors, a pre-defined meta-model is trained. The meta-learning algorithm automatically trains the optimal model parameters, and then adjusts these parameters to obtain the meta-learning model, enabling it to quickly adapt and learn on various tasks. During training, the dataset can be divided into training and test sets. The model is updated using the training set, and then the performance of the meta-learning algorithm is evaluated on the test set.

[0055] In the above implementation process, the meta-learning algorithm can learn knowledge and experience from multiple similar tasks, enabling it to adapt and learn more quickly when faced with new tasks. Furthermore, the meta-learning algorithm can be quickly trained and adjusted on a small amount of sample form data, thereby obtaining more form templates that can adapt to multiple application scenarios and improving the efficiency of obtaining form templates.

[0056] Optionally, in this embodiment of the application, training a preset meta-model using structured feature vectors to obtain a meta-learning algorithm model includes: converting the structured feature vectors into a meta-learning task; obtaining model optimization parameters through an optimization algorithm based on the meta-learning task and a predefined loss function; and obtaining the meta-learning algorithm model based on the model optimization parameters.

[0057] In the specific implementation process: the meta-learning task is input into the preset meta-model, and the meta-learning algorithm is used for training. The meta-learning algorithm can be a gradient descent-based method (e.g., MAML, Reptile), a Bayesian inference-based method (e.g., Bayesian Meta-Learning), a Hamiltonian Monte Carlo method (e.g., MAML-HMC), etc.

[0058] Specifically, for example, in each training iteration, a meta-learning task is selected from the meta-task set. The meta-features of this meta-learning task are used as input to update the model's parameters through a meta-learning algorithm. In each iteration, a small subset of data from the meta-learning task is used to update the model's gradients, obtaining optimized model parameters that allow the model to quickly adapt to the specific task.

[0059] After obtaining the model optimization parameters, the hyperparameters of the meta-model can be adjusted, the network structure modified, and constraints added, etc., to obtain the meta-learning algorithm model.

[0060] In the above implementation process, the metamodel can provide personalized model optimization for different tasks by learning the commonalities and differences of sample forms. By quickly adjusting the model parameters and the expression of form template fields, it can improve applicability and performance and generate more accurate form templates.

[0061] Optionally, in this embodiment, the loss function includes:

[0062] L=∑|G i -S i |

[0063] Where L is the loss function, i is the target field, and G is the target field. i For the representation of the target field in the form template, S i This is how the target field is represented in the sample form.

[0064] In the specific implementation process: G i The acquisition of this information depends on the inferences made by the meta-learning algorithm during the learning process. Example form S of meta-learning algorithm analysis. i The field layout and content are determined through training to understand which fields are necessary for specific types of products, and based on this, new templates are created. i Space is reserved for these fields. This can be understood as G... i It is a meta-learning algorithm model based on sample form S i The training results depict a "sketched" representation of the target field in the form target.

[0065] The loss function describes the difference between the model's predictions and the true labels in the sample form, thus guiding the tuning and optimization of model parameters. By minimizing the loss function, the optimal parameter values ​​of the model on the given training data can be found. In other words, the loss function represents the total difference between the generated form template and the sample form across all fields.

[0066] In the above implementation process, the meta-learning algorithm model will adjust its internal model parameters to minimize the loss function, that is, reduce the difference between the sample form and the generated form template, so as to improve the accuracy of the meta-learning algorithm model and thus generate a more accurate form template.

[0067] Optionally, in this embodiment of the application, obtaining the data to be filled includes: acquiring the data to be processed and the analysis type; extracting features from the data to be processed and filtering the extracted features to obtain data features; and using the analysis algorithm corresponding to the analysis type to perform data analysis on the data features to obtain the data to be filled.

[0068] In the specific implementation process: the data to be processed can be raw data with fill-in information that needs to be entered into a form. For example, the data to be filled in is cluster analysis data, and the data to be processed is product information for various commodities. It may not be possible to directly obtain the data to be filled in from the data to be processed. Appropriate data analysis can be performed on the data to be processed to obtain the data to be filled in that meets the form's requirements.

[0069] Obtain the data to be processed and the analysis type. The analysis type can be determined based on the data to be entered or business requirements. Perform feature extraction on the data to be processed. Feature extraction refers to transforming the original data into a set of new features that better represent the data characteristics. Methods for feature extraction can include data dimensionality reduction, data transformation, or deriving new variables.

[0070] In an optional embodiment, the process of extracting features from the data to be processed may include time series analysis. For example, if the data to be processed includes the time when goods enter the inventory and the time when they can be released from the inventory, the interval between goods leaving the inventory can be calculated based on the time when goods enter the inventory and the time when they can be released from the inventory. Alternatively, the movement frequency of goods within a certain time period can be calculated.

[0071] After feature extraction of the data to be processed, the extracted features are obtained. Based on feature extraction, the features most helpful for analysis and prediction are further selected. Feature selection can use methods such as information gain, correlation coefficient, and chi-square test to filter out irrelevant or duplicate features and retain the most important variables. For example, if the extracted feature is the inspection speed of goods entering the warehouse, and the analysis type is sales trend analysis, this feature has no obvious correlation with the sales trend analysis of goods, so it can be discarded; however, features such as the manufacturer, raw materials, and historical sales records of the goods have a certain correlation with the sales trend of the goods, so these features are retained.

[0072] Each analysis type can have at least one corresponding analysis algorithm. For example, the algorithm corresponding to sales trend analysis can be a time series analysis algorithm. The correspondence between analysis types and analysis algorithms can be pre-stored in the database. The analysis algorithms are pre-built algorithms that can be called when needed. After obtaining data features through filtering, the analysis algorithm corresponding to the analysis type is used to analyze the data features and obtain the data to be filled in.

[0073] In the above implementation process, if the data to be filled cannot be directly obtained from the data to be processed, appropriate data analysis can be performed on the data to be processed to obtain the data to be filled that meets the form filling requirements, so as to generate form data and improve the accuracy of the generated form data.

[0074] Optionally, in this embodiment, the analysis type includes combination analysis, cluster analysis, sales trend analysis, and / or demand analysis; the analysis algorithm corresponding to the analysis type is used to analyze the data features to obtain the data to be filled, including: if the analysis type is combination analysis, then the association rule learning algorithm is used to analyze the data features to obtain data association results; if the analysis type is cluster analysis, then the cluster analysis algorithm is used to analyze the data features to obtain data classification results; if the analysis type is sales trend analysis, then the time series analysis algorithm is used to analyze the data features to obtain product sales trend results; if the analysis type is demand analysis, then the supervised learning algorithm is used to analyze the data features to obtain product demand prediction results; the data to be filled includes data association results, data classification results, product sales trend results, and / or product demand prediction results.

[0075] In the specific implementation process, the analysis types include combination analysis, cluster analysis, sales trend analysis, and / or demand analysis. Among them, combination analysis can be used to analyze the purchasing patterns between products, such as combinations of products that are frequently shipped together; cluster analysis can be used to analyze and divide products into different categories; sales trend analysis can be used to predict the sales trends of products; and demand analysis can be used to predict the future demand for products to determine when to replenish stock or optimize inventory.

[0076] If the analysis type is a combinatorial analysis, then the association rule learning algorithm is used to analyze the data features. The association rule learning algorithm is such as the Apriori algorithm, which obtains the data association results. The data association results can be specifically the association rules and dependencies between products, such as two products being sold together frequently.

[0077] If the analysis type is cluster analysis, then cluster analysis algorithms are used to analyze the data features. Examples of cluster analysis algorithms include K-Means, which yields data classification results. Based on these classification results, products can be further segmented, allowing for adjustments to product placement and other optimizations.

[0078] If the analysis type is sales trend analysis, then time series analysis algorithms are used to analyze the data characteristics. Time series analysis algorithms, such as the ARIMA model, obtain the sales trend results of the products. Based on the sales trend results of the products, it is possible to assess when to replenish stock or optimize inventory.

[0079] If the analysis type is demand analysis, then supervised learning algorithms are used to analyze the data features. Supervised learning algorithms can include support vector machines (SVM) or random forests, etc., to obtain the product demand prediction results. The product demand prediction results can be the future sales level of the product.

[0080] Data association results, data classification results, product sales trend results, and / or product demand forecast results can all be used as data to be filled in.

[0081] In the above implementation process, based on the selected data features, the data features are analyzed using the analysis algorithm corresponding to the analysis type. The analysis algorithm can be applied to different types of data and application scenarios to obtain the data to be filled in that meets the form requirements, improve the accuracy of data processing, and increase the speed of generating form data.

[0082] Optionally, in this embodiment of the application, obtaining the data to be processed includes: using an RFID reader to identify RFID tags via radio frequency signals to obtain the data to be processed; the RFID tags are obtained by pre-programming the product data.

[0083] In the specific implementation process: RFID (Radio Frequency Identification) readers can identify and read data on RFID tags embedded in goods through wireless radio frequency signals.

[0084] For example, an RFID reader emits a radio frequency (RF) signal, typically an electromagnetic field generated by an antenna. This RF signal is transmitted to the vicinity of the RFID tag. When the RFID tag enters the signal range of the RFID reader, the tag is activated and ready to communicate with the reader. The activated RFID tag receives the RF signal sent by the RFID reader and transmits a response signal using its own antenna. The RFID reader receives the response signal sent by the RFID tag, which contains the data to be processed embedded on the tag. The RFID reader decodes the received response signal to extract the data to be processed stored on the tag.

[0085] RFID tags are obtained by pre-programming product data. This product data corresponds to the data to be processed, which can be understood as encoding the data to be processed in a certain format in advance. For example, RFID programming equipment or software is used to write the data into the tag to obtain the RFID tag.

[0086] In the above implementation process, wireless communication between electronic tags (RFID tags) and readers (RFID readers) eliminates the need for manual data reading, improving data storage and management efficiency. It also significantly increases identification speed and accuracy, reducing human error. Applying RFID technology to form data generation schemes leverages its security controls to enhance data security and anti-counterfeiting capabilities.

[0087] In an optional embodiment, an RFID reader is used to identify RFID tags via radio frequency signals to obtain data to be processed, such as product information for various goods. Features are extracted from the data, and the extracted features are filtered to obtain data features. The data features are then analyzed using an analysis algorithm corresponding to the analysis type to obtain the data to be filled in. For example, a time series analysis algorithm can be used to analyze the data features to obtain computer sales trend results.

[0088] The data to be entered is the computer sales trend result, and the data type is the sales forecast result. A target form template for entering the sales forecast result is matched from the form template library. The target form template can include information such as product name, product information, time, and predicted sales amount. The target form template can be obtained by a meta-learning algorithm model learning from a mobile phone sales sample form. The computer sales trend result is written into the target form template to obtain the form data.

[0089] Please see Figure 2 The diagram shown is a structural schematic of the form generation device provided in an embodiment of this application; this application provides a form generation device 200, including:

[0090] Module 210 is used to obtain the data to be filled and the data type of the data to be filled.

[0091] The matching template module 220 is used to match a target form template in a preset form template library according to the data type. The form template library includes at least one form template. The form template is obtained by inputting a sample form into a meta-learning algorithm model and learning from the sample form using the meta-learning algorithm model.

[0092] The fill module 230 is used to write the data to be filled into the target form template and generate form data.

[0093] Optionally, in this embodiment of the application, the form generation device further includes: a model training module, used to obtain a sample form; perform data preprocessing on the sample form to obtain a training form; parse and encode the fields in the training form to obtain a structured feature vector; and use the structured feature vector to train a preset meta-model to obtain a meta-learning algorithm model.

[0094] Optionally, in the embodiments of this application, the form generation device and the model training module are further configured to convert the structured feature vector into a meta-learning task; obtain model optimization parameters through an optimization algorithm based on the meta-learning task and a predefined loss function; and obtain a meta-learning algorithm model based on the model optimization parameters.

[0095] Optionally, in this embodiment of the application, the form generation apparatus includes a loss function comprising:

[0096] L=Σ|G i -S i |

[0097] Where L is the loss function, i is the target field, and G is the target field. i For the representation of the target field in the form template, S i This is how the target field is represented in the sample form.

[0098] Optionally, in this embodiment of the application, the form generation device and the obtaining module are further configured to obtain the data to be processed and the analysis type; extract features from the data to be processed and filter the extracted features to obtain data features; and perform data analysis on the data features using the analysis algorithm corresponding to the analysis type to obtain the data to be filled in.

[0099] Optionally, in this embodiment of the application, the form generation device performs analysis of various types, including combination analysis, cluster analysis, sales trend analysis, and / or demand analysis. The obtaining module is further configured to: if the analysis type is combination analysis, perform data analysis on the data features using an association rule learning algorithm to obtain data association results; if the analysis type is cluster analysis, perform data analysis on the data features using a cluster analysis algorithm to obtain data classification results; if the analysis type is sales trend analysis, perform data analysis on the data features using a time series analysis algorithm to obtain product sales trend results; and if the analysis type is demand analysis, perform data analysis on the data features using a supervised learning algorithm to obtain product demand prediction results. The data to be filled in includes data association results, data classification results, product sales trend results, and / or product demand prediction results.

[0100] Optionally, in this embodiment of the application, the form generation device and the obtaining module are further configured to use an RFID reader to identify RFID tags via radio frequency signals to obtain data to be processed; the RFID tags are obtained by pre-programming the product data.

[0101] It should be understood that this device corresponds to the form generation method embodiment described above and is capable of performing the various steps involved in the above method embodiment. The specific functions of this device can be found in the description above, and detailed descriptions are omitted here to avoid repetition. The device includes at least one software functional module that can be stored in memory or embedded in the device's operating system (OS) in the form of software or firmware.

[0102] Please see Figure 3 The diagram shows a structural schematic of an electronic device provided in an embodiment of this application. An electronic device 300 provided in this application includes a processor 310 and a memory 320. The memory 320 stores machine-readable instructions executable by the processor 310. When the machine-readable instructions are executed by the processor 310, the method described above is performed.

[0103] This application also provides a storage medium storing a computer program, which is executed by a processor to perform the above-described method.

[0104] The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0105] It should be understood that the disclosed apparatus and methods can also be implemented in other ways, given the several embodiments provided in this application. The apparatus embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, or they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0106] In addition, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0107] The above description is only an optional implementation of the embodiments of this application, but the protection scope of the embodiments of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the embodiments of this application should be covered within the protection scope of the embodiments of this application.

Claims

1. A form generation method, characterized in that, include: Obtain the data to be filled and the data type of the data to be filled; Based on the data type, match the target form template in the preset form template library; The form template library includes at least one form template; The form template is obtained by inputting a sample form into a meta-learning algorithm model and then using the meta-learning algorithm model to learn from the sample form. The data to be filled in is written into the target form template to generate form data; Before matching the target form template in the preset form template library according to the data type, the method further includes: Obtain the sample form; The sample form is preprocessed to obtain the training form; The fields in the training form are parsed and encoded to obtain structured feature vectors; The pre-defined meta-model is trained using the structured feature vectors to obtain the meta-learning algorithm model; The step of training a preset meta-model using the structured feature vectors to obtain the meta-learning algorithm model includes: The structured feature vectors are then converted into a meta-learning task. Based on the meta-learning task and the predefined loss function, the model optimization parameters are obtained through an optimization algorithm. Based on the model optimization parameters, the meta-learning algorithm model is obtained; The loss function includes: in, For loss function, For the target field, The expression of the target field in the form template, the The expression of the target field in the sample form; The method further includes: The generated form template is validated to see if it meets the preset requirements. If it does, the form template is output. If it does not meet the preset requirements, the form template is adjusted and regenerated using a meta-learning algorithm model.

2. The method according to claim 1, characterized in that, The process of obtaining the data to be filled includes: Obtain the data to be processed and the analysis type; Feature extraction is performed on the data to be processed, and the extracted features are filtered to obtain data features; The data features are analyzed using the analysis algorithm corresponding to the analysis type to obtain the data to be filled in.

3. The method according to claim 2, characterized in that, The analysis types include combination analysis, cluster analysis, sales trend analysis, and / or demand analysis; the step of using the analysis algorithm corresponding to the analysis type to analyze the data features and obtain the data to be filled includes: If the analysis type is the combined analysis, then the data features are analyzed using the association rule learning algorithm to obtain data association results; If the analysis type is cluster analysis, then the cluster analysis algorithm is used to perform data analysis on the data features to obtain data classification results; If the analysis type is the sales trend analysis, then the time series analysis algorithm is used to analyze the data features to obtain the product sales trend results; If the analysis type is the demand analysis, then a supervised learning algorithm is used to analyze the data features to obtain the commodity demand prediction results. The data to be filled in includes data association results, data classification results, product sales trend results, and / or product demand forecast results.

4. The method according to claim 2, characterized in that, The acquisition of data to be processed includes: The data to be processed is obtained by using an RFID reader to identify RFID tags via radio frequency signals; the RFID tags are obtained by pre-programming the product data.

5. A form generation device, characterized in that, include: The acquisition module is used to obtain the data to be filled and the data type of the data to be filled; The template matching module is used to match a target form template in a preset form template library based on the data type. The form template library includes at least one form template; The form template is obtained by inputting a sample form into a meta-learning algorithm model and then using the meta-learning algorithm model to learn from the sample form. The data entry module is used to write the data to be entered into the target form template and generate form data. The model training module is used to obtain the sample form; The sample form is preprocessed to obtain the training form; The fields in the training form are parsed and encoded to obtain structured feature vectors; the structured feature vectors are then used to train a preset meta-model to obtain the meta-learning algorithm model. The model training module is also used to convert the structured feature vector into a meta-learning task; and to obtain model optimization parameters through an optimization algorithm based on the meta-learning task and a predefined loss function. Based on the model optimization parameters, the meta-learning algorithm model is obtained; The loss function includes: in, For loss function, For the target field, The expression of the target field in the form template, the The expression of the target field in the sample form; The method also includes: The generated form template is validated to see if it meets the preset requirements. If it does, the form template is output. If it does not meet the preset requirements, the form template is adjusted and regenerated using a meta-learning algorithm model.

6. An electronic device, characterized in that, include: A processor and a memory, the memory storing machine-readable instructions executable by the processor, which, when executed by the processor, perform the method as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the method as described in any one of claims 1 to 4.

Citation Information

Patent Citations

  • Prediction model generation method and device, model prediction method and device and computer equipment

    CN115617798A

  • Table generation method and device, equipment, medium and program product

    CN116306545A