A method, system and medium for realizing data products based on digital public products

Through the method based on digital public products, the product demand characteristics are extracted and digital vouchers with anti-counterfeiting watermarks are generated, which solves the security and normative problems in data product transactions and improves the management convenience and security of data products.

CN118735686BActive Publication Date: 2024-11-26GUIZHOU UNIVERSITY OF FINANCE AND ECONOMICS +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411217144.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-02
Publication Date
2024-11-26
Estimated Expiration
2044-09-02

AI Technical Summary

Technical Problem

The existing data product management mechanism is imperfect, resulting in the risk of leakage and information tampering in data product transactions, and the lack of unified production standards, which reduces the security and standardization of data products.

Method used

Using a method based on digital public products, the product demand characteristics are extracted through neural network models, digital public product templates that meet the needs are generated, and digital vouchers that are anti-counterfeiting watermarks are generated through data encoding and watermark embedding, for the transaction and management of data products.

Benefits of technology

It improves the production quality and standardization of data products, enhances the management convenience and security of data products, and avoids the risks of information leakage and tampering.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118735686B_ABST
    Figure CN118735686B_ABST
Patent Text Reader

Abstract

The present application discloses a method, system and medium for realizing data products based on digital public products, wherein the method includes: extracting a number of demand features from the acquired product demand; performing weighted matching and feature splicing on the demand features and preset standard product template regulations according to the priority of the demand features to generate a digital public product template; performing data association between the preset product data set and the label of the digital public product template to generate a data product; inputting the data product into the voucher generation model for data encoding and watermark embedding to obtain a digital voucher containing an anti-counterfeiting watermark for each data product; uploading the digital voucher to the data product trading platform. The present application uses digital public products as the unified standard for producing data products, improves the production quality and standardization of data products, and uses digital vouchers for the circulation of data products, which improves the convenience of data product management and data security.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application belongs to the field of data product management, and specifically relates to a method, system and medium for realizing data products based on digital public products. Background Art

[0002] Data products are digital products that are uniquely identified based on blockchain technology. Compared with physical products, data products are popular among the public due to their advantages such as easy portability, dissemination and permanent preservation. However, due to the imperfect management mechanism of data products, it is easy to leak and tamper with product information during the transaction of data products, which reduces the security of data products.

[0003] At present, the production method of data products is based on different production versions and models. Although data products are accurately produced, it is difficult to unify the production standards of data products, resulting in a decline in the standardization of data products. Blockchain storage technology is still commonly used to support data product transactions, but due to the lack of corresponding identity identification of data products, it is difficult to verify data products during data transactions and circulation, and directly using the data products themselves for verification is prone to information leakage. Summary of the invention

[0004] This application proposes a data product implementation method, system and medium based on digital public products, which uses digital public products as the unified standard for producing data products, improves the production quality and standardization of data products, and uses digital certificates for the circulation of data products, thereby improving the convenience of data product management and data security.

[0005] The first aspect of the present application provides a method for realizing a data product based on a digital public product, the method comprising:

[0006] Extracting several demand features from the acquired product demand through a preset neural network model;

[0007] According to the priority of the demand feature, weighted matching and feature splicing are performed on the demand feature and the preset standard product template regulations to generate a digital public product template that meets the product demand;

[0008] Associating a preset product data set with the label of the digital public product template to generate a plurality of data products;

[0009] Input the data products into the trained credential generation model, perform data encoding and watermark embedding on each data product, and obtain a digital credential containing an anti-counterfeiting watermark for each data product;

[0010] Upload the digital certificate to the data product trading platform to enable the data product to enter the market for trading.

[0011] The above scheme first extracts demand features from the acquired product requirements, and then matches and splices the demand features and standard product template regulations according to the priority of the demand features, ensuring that demand features with higher priority account for a larger proportion in the selection of digital public product templates that meet product requirements, thereby improving the accuracy of the generated digital public product templates. Then, the product data set and the digital public product template are data-associated to uniformly generate a large number of data products, thereby improving the production quality and standardization of data products. In addition, through data encoding and watermark production, a digital certificate containing an anti-counterfeiting watermark is generated for each data product. By uploading the digital certificate to the data product trading platform, the transaction of data products is indirectly realized through the digital certificate, thereby avoiding the leakage of data product information during the transaction process, and improving the convenience of data product management and data security.

[0012] In a possible implementation method of the first aspect, a plurality of demand features are extracted from the acquired product demand through a preset neural network model, specifically:

[0013] Inputting the product demand into a preset neural network model, performing semantic recognition and keyword labeling on the product demand, and obtaining a number of product keywords;

[0014] The product keywords and the preset standard feature data set are multi-dimensionally mapped to obtain a number of demand features.

[0015] The above scheme inputs the product demand into a preset neural network model to propose product keywords, and then performs multi-dimensional mapping of the product keywords and the preset standard feature data set to obtain a data format that conforms to the generation of digital public products, thereby improving the efficiency of subsequent data matching.

[0016] In a possible implementation method of the first aspect, according to the priority of the demand feature, weighted matching and feature splicing are performed on the demand feature and the preset standard product template rules to generate a digital public product template that meets the product demand, specifically:

[0017] According to the priority of the demand feature, a calculation weight of the demand feature is assigned; wherein the higher the priority of the demand feature, the higher the corresponding calculation weight;

[0018] According to the calculated weights of the demand characteristics, the demand characteristics and the preset standard product template regulations are weighted matched to obtain the optimal template regulations for each demand characteristic;

[0019] Based on the preset standard product template, the optimal template regulations are feature scanned and spliced ​​to obtain a digital public product template that meets the product requirements.

[0020] The above scheme assigns the calculation weight of the demand feature according to the priority of the demand feature, so that the demand features with higher priority have a higher proportion in the data matching, ensuring that the high-priority demand features are met first, and by weighted matching the demand features with the preset standard product template regulations, a digital public product template that better meets the product requirements is obtained. In addition, the optimal template regulations are scanned and spliced ​​according to the format of the standard product template to obtain a digital public product template with a compliant format and in line with the product requirements, thereby improving the standardization of data products.

[0021] In a possible implementation method of the first aspect, according to the calculated weight of the demand feature, the demand feature and the preset standard product template regulations are weighted matched to obtain the optimal template regulations for each demand feature, specifically:

[0022] Calculating the cosine similarity between each requirement feature and the standard product template regulations, and determining the maximum cosine similarity of each requirement feature;

[0023] Perform weighted calculation based on the calculated weights of all demand features and the corresponding maximum cosine similarity to obtain the similarity evaluation result;

[0024] If the similarity evaluation result is greater than the first threshold, the standard product template rule corresponding to the maximum cosine similarity is used as the optimal template rule, so as to obtain the optimal template rule for each demand feature;

[0025] If the similarity evaluation result is less than or equal to the first threshold, the demand feature with the maximum cosine similarity less than the second threshold and the corresponding standard product template regulations are feature fused to obtain the optimal template regulations for each demand feature.

[0026] The above scheme first calculates the cosine similarity between each demand feature and the standard product template regulations, selects the maximum cosine similarity and the calculation weight of the demand feature for calculation, and obtains the comprehensive similarity of all demand features, that is, the similarity evaluation result. When the similarity evaluation result is greater than the first threshold, it means that all the current optimal template regulations can generally meet the product requirements, so they can be used to generate digital public products; and when the similarity evaluation result is less than or equal to the first threshold, it means that all the optimal template regulations do not meet the product requirements. It is necessary to integrate the demand features that are significantly different from the standard product template regulations to improve the similarity between the standard product template regulations and the corresponding demand features, so as to obtain the optimal template regulations that are closest to each demand feature and improve the accuracy of the subsequently generated digital public products.

[0027] In a possible implementation method of the first aspect, the cosine similarity and the similarity evaluation result are specifically:

[0028] The specific calculation formula of the cosine similarity is:

[0029] ;

[0030] In the formula, s ij is the i-th demand feature and the jth standard product template regulations The cosine similarity between ;

[0031] The specific calculation formula of the similarity evaluation result is:

[0032] ;

[0033] Where P is the similarity evaluation result, N is the total number of demand features, and A i is the calculation weight of the i-th demand feature, S m,i is the maximum cosine similarity of the i-th demand feature.

[0034] In a possible implementation method of the first aspect, a preset product data set is associated with a label of the digital public product template to generate a plurality of data products, specifically:

[0035] Labeling the product dataset according to the label of the digital public product template;

[0036] Based on the same label, data association is performed on the product data set and the digital public product template to obtain a mapping relationship between the product data set and the digital public product template;

[0037] Inputting the product data set into a pre-trained data processing model for data segmentation and logical integration to obtain a number of product description information;

[0038] According to the mapping relationship, the product description information is input into the digital public product template to generate a plurality of data products.

[0039] The above scheme first labels the product data set, and then associates the product data set with the digital public product template based on the same label, and the obtained mapping relationship provides support for subsequent data filling; then, the product data set is segmented to obtain a number of independent product description information, each of which can be used to generate a data product; according to the mapping relationship, the product description information is input into the digital public product template for data filling, and a number of standardized data products are obtained.

[0040] In a possible implementation method of the first aspect, the data product is input into a trained credential generation model, and data encoding and watermark embedding are performed on each data product to obtain a digital credential containing an anti-counterfeiting watermark for each data product, specifically:

[0041] Extracting product characterization features from the data product; wherein the product characterization features include product number, product type and product basic information;

[0042] Input the product characterization features into the trained voucher generation model, encode the product type and basic product information to obtain a product business card containing a QR code, so that scanning the QR code can display the basic product information of the data product;

[0043] An anti-counterfeiting watermark is generated according to the product number and embedded into the corresponding product business card to obtain a digital certificate containing the anti-counterfeiting watermark for each data product, so that the data product and its source can be verified through the anti-counterfeiting watermark.

[0044] The above scheme extracts product characterization features from the data product, and data encodes the product characterization features to obtain a product business card containing a QR code. Through the product business card, the information of the data product can be displayed during product transactions without direct contact with the data product, thereby ensuring the convenience of data product management and data security; then an anti-counterfeiting watermark is generated according to the product number, and the anti-counterfeiting watermark is embedded in the corresponding product business card to obtain a digital certificate containing the anti-counterfeiting watermark, so that the digital certificate can be used to prove the authenticity of the data product and the authority of the data source.

[0045] In a possible implementation method of the first aspect, uploading the digital certificate to a data product trading platform so that the data product enters the market for trading is specifically as follows:

[0046] The digital certificate is stored in the blockchain, and the storage node address of the digital certificate is hashed and embedded in the digital certificate;

[0047] Upload the digital certificate to the data product trading platform to enable the data product to enter the market for trading.

[0048] The above solution directly stores the digital certificate in the blockchain instead of storing the data product in the blockchain, saving data storage capacity; and hashing the storage node address of the digital certificate and embedding it into the digital certificate, so that the digital certificate can be modified and moved through the blockchain, and the circulation of data products can be completed using the digital certificate, which improves the convenience of data product management.

[0049] The second aspect of the present application provides a data product realization system based on digital public products, the system comprising: a demand feature extraction module, a product template generation module, a data product generation module, a digital certificate generation module and a digital certificate upload module;

[0050] Wherein, the demand feature extraction module is used to extract a number of demand features from the acquired product demand through a preset neural network model;

[0051] The product template generation module is used to perform weighted matching and feature splicing on the demand features and preset standard product template regulations according to the priority of the demand features, so as to generate a digital public product template that meets the product requirements;

[0052] The data product generation module is used to associate the preset product data set with the label of the digital public product template to generate a plurality of data products;

[0053] The digital certificate generation module is used to input the data product into the trained certificate generation model, perform data encoding and watermark embedding on each data product, and obtain a digital certificate containing an anti-counterfeiting watermark for each data product;

[0054] The digital certificate uploading module is used to upload the digital certificate to the data product trading platform so that the data product can enter the market for trading.

[0055] A third aspect of the present application provides a storage medium, which stores computer-readable program code. When the computer-readable program code is executed, the steps of a method for realizing a data product based on a digital public product as described in any one of the embodiments of the present application are implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the technical solution of the present application, the drawings required for use in the implementation manner will be briefly introduced below. Obviously, the drawings described below are only some implementation manners of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0057] Figure 1 It is a specific flow chart of a method for realizing a data product based on a digital public product provided in a certain embodiment of the present application;

[0058] Figure 2 It is a specific structural diagram of a data product implementation system based on a digital public product provided in a certain embodiment of the present application. DETAILED DESCRIPTION

[0059] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0060] It should be understood that the step numbers used in this article are only for the convenience of description and are not intended to limit the order in which the steps are executed.

[0061] like Figure 1 As shown, Figure 1 A specific flow chart of a method for realizing a data product based on a digital public product is provided for a certain embodiment of the present application. The method for realizing a data product based on a digital public product of this embodiment includes steps S1 to S5, which are described in detail as follows:

[0062] Step S1, extracting a number of demand features from the acquired product demand through a preset neural network model;

[0063] In this step, product requirements are first collected and input into the trained neural network model. In some embodiments, the neural network model is built based on YOLOv3.

[0064] In the neural network model, the product requirements are first semantically identified to obtain a number of semantically complete words, and then these words are keyword-tagged, and the words related to product features are marked to obtain a number of product keywords. Among them, product keywords can be separate words or short sentences to enrich the content of product keywords. The product requirements can be content descriptions or functional descriptions of data products that users or operators want to generate, such as product usage, product name, and product-related fields.

[0065] The product keywords and the preset standard feature data set are multi-dimensionally mapped to match more similar and standardized standard data in the standard feature data set for the product keywords, to obtain a mapping relationship between each product keyword and the standard feature data, and to determine the demand feature corresponding to the product keyword based on the mapping relationship and the feature vector corresponding to the standard feature data.

[0066] For example, if you want to generate a batch of membership redemption cards for market transactions, first obtain product keywords based on the input product requirements for membership redemption cards, such as "usage platform", "customizable interface background" and "redemption period". After multi-dimensional mapping, it can be determined that the standard feature data corresponding to "customizable interface background" is "membership privileges", so as to achieve standardization and classification of product keywords, and then obtain the demand characteristics corresponding to "customizable interface background" based on the mapping relationship between "customizable interface background" and "membership privileges" and the feature vector corresponding to "membership privileges". It is worth mentioning that the feature vector corresponding to "membership privileges" is not consistent with the demand characteristics corresponding to "customizable interface background", because the demand characteristics also include the mapping relationship between "customizable interface background" and "membership privileges".

[0067] Step S2, according to the priority of the demand feature, weighted matching and feature splicing are performed on the demand feature and the preset standard product template rules to generate a digital public product template that meets the product demand;

[0068] In this step, the demand features have specific priorities. The demand features with high priorities can obtain more accurate digital public product templates that meet product requirements. Therefore, it is necessary to assign calculation weights to the demand features according to their priorities to reflect the importance of high-priority demand features. Among them, the demand features with higher priorities include product category and product safety level.

[0069] In order to highlight the importance of high-priority demand features, it is necessary to assign the calculation weights of the demand features according to the priorities of the demand features; wherein the assignment range of the calculation weights is [1,10], the highest priority is assigned to 10, and the lowest priority is assigned to 1.

[0070] Then, according to the calculated weights of the demand features, the demand features are weighted matched with preset standard product template regulations to obtain the optimal template regulations for each demand feature.

[0071] Furthermore, the cosine similarity between each demand feature and the standard product template rules is calculated. The standard product template rules are information stored in the database and used to generate digital public products. The information is multiple rules, each of which corresponds to the component information required for a digital public product. It can quickly determine the data format required to integrate the demand features into the digital public product template, thereby improving the accuracy of the digital public product template.

[0072] Exemplarily, some standard product template regulations may be information descriptions about the standard feature data "membership privileges", such as "administrators can configure content independently according to their permissions", "support uploading multiple files without limit" and "exclusive online customer service", etc. For the demand characteristics corresponding to "customizable interface background", the corresponding optimal template regulation can be determined by calculating the cosine similarity of the standard product template regulations with the above three types of "membership privileges" respectively.

[0073] The specific calculation formula of the cosine similarity is:

[0074] ;

[0075] In the formula, s ij is the i-th demand feature and the jth standard product template regulations The cosine similarity between . ij The value range is [-1,1].

[0076] When the cosine similarity is -1, it means that the corresponding demand characteristics and the standard product template regulations are completely inconsistent; when the cosine similarity is 1, it means that the corresponding demand characteristics and the standard product template regulations are completely consistent.

[0077] According to the above formula, after obtaining the cosine similarity between each demand feature and all standard product template regulations, the one with the largest value is selected from the cosine similarities as the maximum cosine similarity of each demand feature, and then the standard product template regulation corresponding to the maximum cosine similarity is determined. The standard product template regulation is the existing standard product template regulation that is most similar to the demand feature.

[0078] Then, the calculation weight of each demand feature and the corresponding maximum cosine similarity are weighted and calculated, and the weighted calculation results of all demand features are summed to obtain the similarity evaluation result. The larger the calculation weight, the greater the impact of the weighted calculation result on the similarity evaluation result.

[0079] The specific calculation formula of the similarity evaluation result is:

[0080] ;

[0081] Where P is the similarity evaluation result, N is the total number of demand features, and A i is the calculation weight of the i-th demand feature, S m,i is the maximum cosine similarity of the i-th demand feature.

[0082] The similarity evaluation result is the overall similarity between all demand features and the corresponding standard product template regulations. Only when the similarity evaluation result is greater than the first threshold, the standard product template regulations corresponding to the maximum cosine similarity are used as the optimal template regulations, indicating that all demand features as a whole can obtain information that meets product requirements and is used to generate digital public products.

[0083] However, when the similarity evaluation result is less than or equal to the first threshold, it means that it is impossible to obtain information for generating digital public products that meets the product requirements as a whole, so it is necessary to fuse the demand features with the maximum cosine similarity less than the second threshold and the corresponding standard product template regulations, and further increase the similarity between the demand features with the maximum cosine similarity less than the second threshold and the corresponding standard product template regulations, so as to obtain information for generating digital public products that meets the product requirements. The standard product template regulations after feature fusion are the optimal template regulations for the corresponding demand features.

[0084] Among them, the calculation formula of the feature fusion is:

[0085] ;

[0086] In the formula, is the i-th feature point of the demand feature x, is the i-th feature point of the standard product template y, is the feature fusion result, is the fusion weight factor corresponding to the i-th feature point, for and Activation map of .

[0087] Finally, the optimal template regulations for each demand feature are scanned according to the preset standard product template to obtain a scanning result that conforms to the format of the standard product template, and then the scanning results are spliced ​​in the order provided by the standard product template to obtain a complete digital public product template that meets the product requirements; wherein the standard product template is a set data product template.

[0088] Step S3, data association is performed between the preset product data set and the label of the digital public product template to generate a plurality of data products;

[0089] In this step, the product data set is data that meets the product requirements and is used to fill the digital public product template. Before data filling, in order to ensure that the product data set can be accurately used to generate data products, the product data set and the digital public product template need to be accurately associated.

[0090] First, according to the label of the digital public product template, the product data set is labeled accordingly, and information of the same type as the label is marked in the product data set. Then, based on the same labels in the product data set and the digital public product template, the product data set and the digital public product template are associated with each other to generate a mapping relationship between the product data set and the digital public product template.

[0091] The product data set is then input into a pre-trained data processing model for data segmentation, the product data set is segmented into a number of semantically coherent phrase data according to a preset semantic logic, and the phrase data are logically integrated and spliced ​​to obtain a number of semantically complete product description information.

[0092] Finally, according to the mapping relationship, the product description information is input into the digital public product template for accurate data matching. Each product description information can generate a data product, thereby obtaining a plurality of data products.

[0093] Furthermore, in the process of generating data products, each data product will obtain a corresponding product number according to the order of generation, and the product number is a unique identifier for each data product.

[0094] Step S4, inputting the data product into the trained voucher generation model, performing data encoding and watermark embedding on each data product, and obtaining a digital voucher containing an anti-counterfeiting watermark for each data product;

[0095] In this step, product characterization features are extracted from the data product, including product number, product type and basic product information, etc. The basic product information is a brief description of the data product.

[0096] The product characterization features are input into the trained voucher generation model for data processing, and the product type and basic product information are data encoded to obtain a product business card containing a QR code; wherein, the product business card is used to display brief information of the corresponding data product, such as the product type of the data product, the digital public product template used, the product number and the product price, etc.; by scanning the QR code, more detailed basic product information can be viewed, such as the functional use of the data product and a detailed appearance display.

[0097] Finally, an anti-counterfeiting watermark is generated for each product business card according to the product number, and the anti-counterfeiting watermark is embedded in a designated position of the product business card to obtain a digital certificate containing an anti-counterfeiting watermark for each data product. According to the anti-counterfeiting watermark, the source authority and uniqueness of the corresponding data product can be confirmed.

[0098] Step S5, uploading the digital certificate to the data product trading platform so that the data product can enter the market for trading.

[0099] In this step, the digital certificate is selected to be stored in the blockchain, and the storage node address of the digital certificate is hashed and hidden and embedded in the digital certificate. After purchasing the data product, a key can be obtained, and the corresponding digital certificate can be decoded using the key to obtain the storage node address of the digital certificate, and the digital certificate can be modified to belong to the purchaser.

[0100] Furthermore, the storage node of the digital certificate also stores the storage address of the corresponding data product, and binds the storage address to the storage node address, so that after obtaining the storage node address, the data product can be obtained according to the storage address, thereby improving data security.

[0101] The generated digital certificate is uploaded to the data product trading platform, and the circulation and transaction of the data product are completed through the digital certificate. The basic information of the data product can be understood without direct access to the data product.

[0102] Furthermore, in order to execute the data product realization system based on digital public products corresponding to the above method embodiment to achieve the corresponding functions and technical effects, Figure 2 A structural diagram of a data product implementation system based on digital public products is provided. For ease of description, only the parts related to this embodiment are shown. The data product implementation system based on digital public products provided in the embodiment of the present application includes:

[0103] The demand feature extraction module 201 is used to extract a number of demand features from the acquired product demand through a preset neural network model;

[0104] The product template generation module 202 is used to perform weighted matching and feature splicing on the demand features and preset standard product template rules according to the priority of the demand features, so as to generate a digital public product template that meets the product requirements;

[0105] The data product generation module 203 is used to associate the preset product data set with the label of the digital public product template to generate a plurality of data products;

[0106] A digital certificate generation module 204 is used to input the data product into the trained certificate generation model, perform data encoding and watermark embedding on each data product, and obtain a digital certificate containing an anti-counterfeiting watermark for each data product;

[0107] The digital certificate uploading module 205 is used to upload the digital certificate to the data product trading platform so that the data product can enter the market for trading.

[0108] In some embodiments, the demand feature extraction module 201 further includes:

[0109] The demand feature extraction unit is used to input the product demand into a preset neural network model, perform semantic recognition and keyword labeling on the product demand, and obtain a number of product keywords; perform multi-dimensional mapping on the product keywords and a preset standard feature data set to obtain a number of demand features.

[0110] In some embodiments, the product template generation module 202 further includes:

[0111] The digital public product template generation unit is used to assign a calculation weight to the demand feature according to the priority of the demand feature; wherein, the higher the priority of the demand feature, the higher the corresponding calculation weight; according to the calculation weight of the demand feature, the demand feature and the preset standard product template regulations are weighted matched to obtain the optimal template regulations for each demand feature; based on the preset standard product template, the optimal template regulations are feature scanned and spliced ​​to obtain a digital public product template that meets the product requirements.

[0112] The optimal template regulation determination unit is used to calculate the cosine similarity between each demand feature and the standard product template regulation, and determine the maximum cosine similarity of each demand feature; perform weighted calculation according to the calculation weights of all demand features and the corresponding maximum cosine similarities to obtain a similarity evaluation result; if the similarity evaluation result is greater than a first threshold, the standard product template regulation corresponding to the maximum cosine similarity is used as the optimal template regulation to obtain the optimal template regulation for each demand feature; if the similarity evaluation result is less than or equal to the first threshold, the demand feature whose maximum cosine similarity is less than the second threshold and the corresponding standard product template regulation are feature fused to obtain the optimal template regulation for each demand feature.

[0113] In some embodiments, the data product generation module 203 further includes:

[0114] The data product unit is used to label the product data set according to the label of the digital public product template; based on the same label, the product data set and the digital public product template are data associated to obtain a mapping relationship between the product data set and the digital public product template; the product data set is input into a pre-trained data processing model for data segmentation and logical integration to obtain a number of product description information; according to the mapping relationship, the product description information is input into the digital public product template to generate a number of data products.

[0115] In some embodiments, the digital credential generation module 204 further includes:

[0116] A digital certificate generating unit is used to extract product characterization features from the data product; wherein the product characterization features include product number, product type and basic product information; the product characterization features are input into a trained certificate generating model, the product type and basic product information are data-encoded to obtain a product business card containing a QR code, so that the basic product information of the data product can be displayed by scanning the QR code; an anti-counterfeiting watermark is generated according to the product number, and the anti-counterfeiting watermark is embedded in the corresponding product business card to obtain a digital certificate containing the anti-counterfeiting watermark for each data product, so that the data product and the source of the data product can be verified through the anti-counterfeiting watermark.

[0117] In some embodiments, the digital certificate uploading module 205 further includes:

[0118] The data product circulation unit is used to store the digital certificate in the blockchain, hash the storage node address of the digital certificate and embed it into the digital certificate; upload the digital certificate to the data product trading platform to enable the data product to enter the market for trading.

[0119] An embodiment of the present application provides a storage medium, which stores computer-readable program codes. When the computer-readable program codes are executed, the steps of the above-mentioned method for realizing a data product based on a digital public product are implemented.

[0120] The embodiment of the present application proposes a data product realization method, system and medium based on digital public products: extracting several demand features from the acquired product requirements through a preset neural network model; performing weighted matching and feature splicing on the demand features and the preset standard product template regulations according to the priority of the demand features, and generating a digital public product template that meets the product requirements; performing data association between the preset product data set and the label of the digital public product template, and generating several data products; inputting the data products into the trained voucher generation model, performing data encoding and watermark embedding on each data product, and obtaining a digital voucher containing an anti-counterfeiting watermark for each data product; uploading the digital voucher to the data product trading platform, so that the data product can enter the market for trading. Its beneficial effects are: using digital public products as the unified standard for producing data products improves the production quality and standardization of data products, and using digital vouchers for the circulation of data products improves the convenience of data product management and data security.

[0121] The specific embodiments described above further describe the purpose, technical solutions and beneficial effects of the present application in detail. It should be understood that the above description is only a specific embodiment of the present application and is not intended to limit the scope of protection of the present application. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of protection of the present application.

Claims

1. A method for realizing a data product based on a digital public product, characterized in that: include: Extracting several demand features from the acquired product demand through a preset neural network model; According to the priority of the demand feature, the demand feature and the preset standard product template regulations are weighted matched and feature spliced ​​to generate a digital public product template that meets the product requirements, specifically: according to the priority of the demand feature, the calculation weight of the demand feature is assigned; wherein, the higher the priority of the demand feature, the higher the corresponding calculation weight; according to the calculation weight of the demand feature, the demand feature and the preset standard product template regulations are weighted matched to obtain the optimal template regulations for each demand feature; based on the preset standard product template, the optimal template regulations are feature scanned and spliced ​​to obtain a digital public product template that meets the product requirements; The optimal template regulations are specifically as follows: calculating the cosine similarity between each demand feature and the standard product template regulations, and determining the maximum cosine similarity of each demand feature; performing weighted calculation according to the calculation weights of all demand features and the corresponding maximum cosine similarities, and obtaining a similarity evaluation result; if the similarity evaluation result is greater than a first threshold, taking the standard product template regulations corresponding to the maximum cosine similarity as the optimal template regulations, thereby obtaining the optimal template regulations for each demand feature; if the similarity evaluation result is less than or equal to the first threshold, performing feature fusion on the demand feature whose maximum cosine similarity is less than the second threshold and the corresponding standard product template regulations, thereby obtaining the optimal template regulations for each demand feature; Associating a preset product data set with the label of the digital public product template to generate a plurality of data products; Input the data products into the trained credential generation model, perform data encoding and watermark embedding on each data product, and obtain a digital credential containing an anti-counterfeiting watermark for each data product; Upload the digital certificate to the data product trading platform to enable the data product to enter the market for trading.

2. The method for realizing data products based on digital public products according to claim 1, characterized in that: The preset neural network model is used to extract several demand features from the obtained product demand, specifically: Inputting the product demand into a preset neural network model, performing semantic recognition and keyword labeling on the product demand, and obtaining a number of product keywords; The product keywords and the preset standard feature data set are multi-dimensionally mapped to obtain a number of demand features.

3. The method for realizing data products based on digital public products according to claim 1, characterized in that: The cosine similarity and similarity evaluation results are specifically: The specific calculation formula of the cosine similarity is: ; In the formula, s ij is the i-th demand feature and the jth standard product template regulations The cosine similarity between ; The specific calculation formula of the similarity evaluation result is: ; Where P is the similarity evaluation result, N is the total number of demand features, and A i is the calculation weight of the i-th demand feature, S m,i is the maximum cosine similarity of the i-th demand feature.

4. The method for realizing data products based on digital public products according to claim 1, characterized in that: The preset product data set is associated with the label of the digital public product template to generate a number of data products, specifically: Labeling the product dataset according to the label of the digital public product template; Based on the same label, data association is performed on the product data set and the digital public product template to obtain a mapping relationship between the product data set and the digital public product template; Inputting the product data set into a pre-trained data processing model for data segmentation and logical integration to obtain a number of product description information; According to the mapping relationship, the product description information is input into the digital public product template to generate a plurality of data products.

5. The method for realizing data products based on digital public products according to claim 1, characterized in that: The data product is input into the trained credential generation model, and data encoding and watermark embedding are performed on each data product to obtain a digital credential containing an anti-counterfeiting watermark for each data product, specifically: Extracting product characterization features from the data product; wherein the product characterization features include product number, product type and product basic information; Input the product characterization features into the trained voucher generation model, encode the product type and basic product information to obtain a product business card containing a QR code, so that scanning the QR code can display the basic product information of the data product; An anti-counterfeiting watermark is generated according to the product number and embedded into the corresponding product business card to obtain a digital certificate containing the anti-counterfeiting watermark for each data product, so that the data product and its source can be verified through the anti-counterfeiting watermark.

6. The method for realizing data products based on digital public products according to claim 1, characterized in that: The uploading of the digital certificate to the data product trading platform so that the data product can enter the market for trading is specifically: The digital certificate is stored in the blockchain, and the storage node address of the digital certificate is hashed and embedded in the digital certificate; Upload the digital certificate to the data product trading platform to enable the data product to enter the market for trading.

7. A data product realization system based on digital public products, characterized in that: include: Demand feature extraction module, product template generation module, data product generation module, digital certificate generation module and digital certificate upload module; Wherein, the demand feature extraction module is used to extract a number of demand features from the acquired product demand through a preset neural network model; The product template generation module is used to perform weighted matching and feature splicing on the demand feature and the preset standard product template regulations according to the priority of the demand feature, so as to generate a digital public product template that meets the product demand, specifically: assigning a calculation weight of the demand feature according to the priority of the demand feature; wherein the higher the priority of the demand feature, the higher the corresponding calculation weight; according to the calculation weight of the demand feature, weighted matching is performed on the demand feature and the preset standard product template regulations to obtain the optimal template regulations for each demand feature; based on the preset standard product template, feature scanning and splicing are performed on the optimal template regulations to obtain a digital public product template that meets the product demand; wherein , the optimal template regulations are specifically as follows: calculating the cosine similarity between each demand feature and the standard product template regulations, and determining the maximum cosine similarity of each demand feature; performing weighted calculation according to the calculation weights of all demand features and the corresponding maximum cosine similarities, and obtaining a similarity evaluation result; if the similarity evaluation result is greater than a first threshold, taking the standard product template regulations corresponding to the maximum cosine similarity as the optimal template regulations, thereby obtaining the optimal template regulations for each demand feature; if the similarity evaluation result is less than or equal to the first threshold, performing feature fusion on the demand feature whose maximum cosine similarity is less than the second threshold and the corresponding standard product template regulations, thereby obtaining the optimal template regulations for each demand feature; The data product generation module is used to associate the preset product data set with the label of the digital public product template to generate a plurality of data products; The digital certificate generation module is used to input the data product into the trained certificate generation model, perform data encoding and watermark embedding on each data product, and obtain a digital certificate containing an anti-counterfeiting watermark for each data product; The digital certificate uploading module is used to upload the digital certificate to the data product trading platform so that the data product can enter the market for trading.

8. A storage medium, characterized in that: The storage medium stores computer-readable program codes, which, when executed, implement the steps of a method for realizing a data product based on a digital public product according to any one of claims 1 to 6.