Blockchain-based method and system for electronic single certificate storage and verification of cross-border trade

Abstract

The present application relates to a blockchain-based cross-border trade electronic document storage and verification method and system, relating to the field of data storage based on blockchain, the system comprises: an account management module for establishing a plurality of unified authentication trusted accounts through a plurality of blockchain nodes and managing permissions; a document checking module for receiving an upload request of the unified authentication trusted account, determining the category of the electronic document according to the uploaded electronic document, checking the electronic document, and determining the secret level of the electronic document; a document management module for storing the electronic document to the blockchain through a smart contract according to the category and secret level of the electronic document; and a document retrieval module for receiving a document retrieval request initiated by the unified authentication trusted account, querying the target electronic document, and feeding back, which has the advantages of realizing electronic filing and management of the document, reducing the use of paper documents, improving management efficiency, and reducing the risk of document loss.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data storage based on blockchain, in particular to a method and system for electronic single certificate storage and verification in cross-border trade based on blockchain. BACKGROUND

[0002] The export tax rebate filing certificate is a series of single certificate documents that need to be filed and saved according to regulations by an export enterprise in the process of handling export tax rebate. These single certificate documents are important basis for tax authorities to audit export tax rebate, and play a key role in ensuring the authenticity and legality of export tax rebate. The export tax rebate filing certificate includes but is not limited to the purchase and sale contract of the export enterprise (such as export contract, foreign trade comprehensive service contract, foreign trade enterprise purchase contract, production enterprise purchase contract for non-self-produced goods, etc.), the transportation documents of the exported goods (such as sea transport bill, air transport bill, railway transport bill, cargo transport documents, and postal receipt), and the single certificate of the export enterprise entrusting other units to declare customs (such as customs declaration agreement, and the agent customs declaration service fee invoice issued by the entrusted customs declaration unit).

[0003] The traditional export tax rebate filing mainly relies on paper single certificate, and the management process is complicated. In addition, the single certificate is easily lost or damaged due to human factors, which affects the normal handling of export tax rebate business. The paper single certificate needs to be checked manually, which is low in efficiency and easy to make mistakes. The paper single certificate is easy to be tampered with or forged, and the data security is low, and it is difficult to trace the authenticity and source of the single certificate.

[0004] Therefore, it is necessary to provide a method and system for electronic single certificate storage and verification in cross-border trade based on blockchain, which realizes the electronic filing and management of single certificate, reduces the use of paper single certificate, improves the management efficiency, and reduces the risk of single certificate loss. SUMMARY

[0005] The present application provides a system for electronic single certificate storage and verification in cross-border trade based on blockchain, which includes: an account management module for establishing a plurality of unified authentication trusted accounts through a plurality of blockchain nodes, and managing the permissions of the plurality of unified authentication trusted accounts; a single certificate checking module for receiving an upload request of a unified authentication trusted account, determining the category of the electronic single certificate according to the uploaded electronic single certificate, and checking the electronic single certificate to determine the secret level of the electronic single certificate; a single certificate management module for storing the electronic single certificate to the blockchain through a smart contract according to the category and secret level of the electronic single certificate; and a single certificate retrieval module for receiving a single certificate retrieval request initiated by a unified authentication trusted account, querying the target electronic single certificate according to the single certificate retrieval request, and feeding back the target electronic single certificate to the unified authentication trusted account.

[0006] Further, the single document verification module determines the category of the electronic document, comprising: establishing a keyword library; for each unified authentication trusted account, determining templates of electronic documents of multiple categories of the unified authentication trusted account, extracting multiple initial keywords of the templates of electronic documents of each category according to the keyword library, and determining multiple difference keywords of the templates of electronic documents of each category according to the multiple initial keywords of the templates of electronic documents of each category; extracting multiple keywords of the electronic document according to the keyword library; obtaining the multiple difference keywords of the templates of electronic documents of each category of the unified authentication trusted account of the uploaded electronic document; determining the similarity of the electronic document to each category of keywords according to the multiple keywords of the electronic document and the multiple difference keywords of the templates of electronic documents of each category of the unified authentication trusted account of the uploaded electronic document; and determining the category of the electronic document according to the similarity of the electronic document to each category of keywords.

[0007] Further, the single document verification module extracts multiple initial keywords of the templates of electronic documents of each category, comprising: for the templates of electronic documents of each category, pre-processing the templates of electronic documents to generate pre-processed templates of electronic documents, wherein the pre-processing includes word segmentation and stop word removal, extracting multiple keywords of the templates of electronic documents according to the keyword library, determining co-occurrence features of the multiple keywords according to a preset window, establishing a keyword co-occurrence undirected graph according to the co-occurrence features of the multiple keywords, and extracting multiple initial keywords of the templates of electronic documents according to the keyword co-occurrence undirected graph.

[0008] Further, the single document verification module determines multiple difference keywords of the templates of electronic documents of each category, comprising: for the templates of electronic documents of each category, determining node features of each initial keyword based on the keyword co-occurrence undirected graph, determining a recurrence ratio of each initial keyword according to the multiple initial keywords of the templates of electronic documents of each category, taking an initial keyword with a recurrence ratio less than a recurrence ratio threshold as a difference keyword, and for an initial keyword with a recurrence ratio greater than or equal to the recurrence ratio threshold, calculating a recurrence difference value of the initial keyword according to the node features of the initial keyword in the templates of electronic documents of multiple categories, and taking an initial keyword with a recurrence difference value greater than a recurrence difference threshold as a difference keyword.

[0009] Further, the document verification module verifies the electronic document, including: for each unified authentication trusted account, determining a plurality of mandatory fields and a filling rule of each mandatory field according to a template of each category of electronic document of the unified authentication trusted account, and determining a verification priority of each mandatory field according to historical electronic documents of each category of the unified authentication trusted account; and according to the plurality of mandatory fields of the template of the category of the electronic document corresponding to the unified authentication trusted account of the uploaded electronic document, intercepting a plurality of fields from the electronic document, and verifying the plurality of fields according to the filling rule of the plurality of mandatory fields of the template of the category of the electronic document corresponding to the unified authentication trusted account of the uploaded electronic document and the verification priority of each mandatory field.

[0010] Further, the document verification module determines the secret level of the electronic document, including: querying whether the secret level of the electronic document is indicated in the upload request of the unified authentication trusted account, and if yes, determining the secret level of the electronic document according to the knowledge of the upload request of the unified authentication trusted account; and if no, determining a first similar electronic document of the unified authentication trusted account according to the category of the electronic document and the plurality of fields intercepted from the electronic document according to the plurality of mandatory fields; and determining the secret level of the electronic document according to the first similar electronic document of the unified authentication trusted account.

[0011] Further, the document management module stores the electronic document into the blockchain through the smart contract according to the category and the secret level of the electronic document, including: performing sharding processing on the electronic document according to the category and the secret level of the electronic document to generate a plurality of electronic document segments of the electronic document; calculating a hash value of each electronic document segment of the electronic document; determining a storage node of each electronic document segment of the electronic document; establishing a document index of the electronic document based on the hash value and the storage node of each electronic document segment of the electronic document; determining a search tag of the electronic document based on the plurality of fields intercepted from the electronic document according to the plurality of mandatory fields; storing the document index and the search tag of the electronic document on the blockchain through the smart contract; and distributing the plurality of electronic document segments of the electronic document and storing the plurality of electronic document segments in a plurality of nodes through the smart contract.

[0012] Further, the document management module performs sharding processing on the electronic document according to the category and the secret level of the electronic document to generate a plurality of electronic document segments of the electronic document, including: determining a second similar electronic document of the electronic document according to the category and the secret level of the electronic document; and performing sharding processing on the electronic document according to a cycle search frequency of the second similar electronic document of the electronic document to generate the plurality of electronic document segments of the electronic document.

[0013] Further, the single document retrieval module queries the target electronic single document according to the single document retrieval request, and feeds back the target electronic single document to the unified authentication trusted account, comprising: determining a plurality of candidate stored single documents according to the authority of the unified authentication trusted account initiating the single document retrieval request to the plurality of stored single documents; determining the target electronic single document according to the retrieval keywords of the single document retrieval request and the retrieval labels of the plurality of candidate stored single documents; acquiring a plurality of electronic single document segments of the target electronic single document according to the single document index of the target electronic single document; splicing the plurality of electronic single document segments of the target electronic single document to generate the target electronic single document, and feeding back to the unified authentication trusted account initiating the single document retrieval request.

[0014] The application provides a blockchain-based cross-border trade electronic single document storage and verification method, comprising: establishing a plurality of unified authentication trusted accounts through a plurality of blockchain nodes, and managing the authority of the plurality of unified authentication trusted accounts; receiving an upload request of a unified authentication trusted account, and determining the category of an electronic single document according to the uploaded electronic single document, and verifying the electronic single document to determine the secret level of the electronic single document; storing the electronic single document to the blockchain through a smart contract according to the category and secret level of the electronic single document; receiving a single document retrieval request initiated by a unified authentication trusted account, and querying a target electronic single document according to the single document retrieval request, and feeding back the target electronic single document to the unified authentication trusted account.

[0015] Compared with the prior art, the blockchain-based cross-border trade electronic single document storage and verification method and system provided by the present specification has at least the following beneficial effects:

[0016] 1. The decentralization, tamper resistance and transparency characteristics of the blockchain are used to ensure the integrity and authenticity of the electronic single document data. All on-chain single document information is encrypted to prevent data tampering or forgery, improving data credibility; a unified authentication trusted account is established through a plurality of blockchain nodes to ensure the authenticity and uniqueness of the account identity. The account authority is finely managed to further enhance security; automatically receive upload requests, determine the category of the electronic single document, and perform verification and secret level determination. This process reduces manual intervention, improves processing efficiency, and reduces the risk of human error; according to the category and secret level of the electronic single document, the single document is stored to the blockchain through a smart contract. The automatic execution characteristics of the smart contract ensure the transparency and tamper resistance of the storage process, and optimize the allocation and utilization of storage resources; query the target electronic single document according to the request. Through technical means such as single document index and retrieval label, the retrieval efficiency is improved to ensure that users can quickly obtain the required single document.

[0017] 2. By establishing a keyword database and extracting initial and differential keywords from each category of electronic document templates, the category of electronic documents can be identified more accurately. The introduction of differential keywords enables the system to distinguish subtle differences between different categories of documents, thereby improving classification accuracy. Analyzing the co-occurrence characteristics of initial keywords using an undirected keyword co-occurrence graph helps to discover potential associations and patterns between keywords. This analysis method can capture the complex relationships between keywords in document templates, further improving classification accuracy.

[0018] 3. By defining the required fields and filling rules for various categories of electronic document templates for each unified authentication trusted account, the requirements and format for each field are clearly defined, thereby improving the accuracy of verification. Simultaneously, these clear rules reduce the workload of manual verification and improve efficiency. Determining the verification priority for each required field of each category based on historical electronic documents allows the system to prioritize the verification of critical and error-prone fields. This prioritization helps to quickly identify and correct errors in documents, further improving verification efficiency.

[0019] 4. Fragmenting electronic documents based on their category, security level, and the periodic retrieval frequency of similar documents optimizes storage resource allocation, improves storage efficiency, and distributes fragments across multiple nodes, reducing the risk of single points of failure. This ensures that even if some nodes experience data leakage, the complete electronic document cannot be restored, thus enhancing data security. A hash value is calculated for each electronic document fragment to ensure data integrity and immutability. The hash value serves as a unique identifier for the data, verifying its authenticity. The storage process is automatically executed via smart contracts, reducing human intervention and the risk of data tampering or misoperation. A unique fingerprint is generated by combining structured data (such as bill of lading numbers and invoice numbers) with file hashes; the on-chain document stores the fingerprint instead of the full text. This reduces on-chain storage pressure (saving 90% of space) while ensuring immutability. Attached Figure Description

[0020] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein:

[0021] Figure 1 This is a block diagram of a blockchain-based cross-border trade electronic document storage and verification system shown in one embodiment of this application.

[0022] Figure 2 This is an architecture diagram of a blockchain-based cross-border trade electronic document storage and verification system shown in one embodiment of this application.

[0023] Figure 3 is a schematic diagram of a keyword co-occurrence undirected graph shown in an embodiment of the present application;

[0024] Figure 4 is a flowchart of a blockchain-based cross-border trade electronic document storage and verification method shown in an embodiment of the present application. DETAILED DESCRIPTION

[0025] In order to more clearly illustrate the technical solutions of the embodiments of the present specification, the drawings needed in the embodiment description will be briefly introduced as follows.

[0026] Figure 1 is a module diagram of a blockchain-based cross-border trade electronic document storage and verification system shown in an embodiment of the present application, as shown in Figure 1 The blockchain-based cross-border trade electronic document storage and verification system can include an account management module, a document checking module, a document management module, and a document retrieval module, based on blockchain technology, an electronic document storage system is constructed, and through system construction, the functions of collection, archiving, querying, and sharing of enterprise documents are realized, unified management of various documents is realized, repeated input of documents is effectively avoided, one-window storage and one-window retrieval are realized. Meet the relevant provisions and requirements of the regulatory units on the storage and storage of the documents of foreign trade enterprises, provide data support for customs clearance, foreign exchange, tax refund and other businesses.

[0027] Figure 2 is an architecture diagram of a blockchain-based cross-border trade electronic document storage and verification system shown in an embodiment of the present application, as shown in Figure 2 The blockchain-based cross-border trade electronic document storage and verification system adopts a five-layer architecture design, which is a user layer, an application layer, a support service layer, a data layer, and an infrastructure layer.

[0028] The user layer is for export tax refund enterprises, tax authorities, and system operation and management personnel.

[0029] The application layer is the core link of the specific business processing of the system, mainly composed of functions such as document filing, document management, online checking of filing, statistical analysis, etc.

[0030] The business support layer includes document storage services, basic services, blockchain technology services, etc.

[0031] The data layer includes storage of document data, distributed ledger, etc.

[0032] The infrastructure layer is based on cloud platform computing resources, storage resources, network resources, security resources, etc.

[0033] The account management module is used to establish a plurality of unified authentication trusted accounts through a plurality of blockchain nodes, and manage the permissions of the plurality of unified authentication trusted accounts.

[0034] Specifically, by establishing a unified authentication trusted account, the single certificate data ownership is returned to the enterprise and the account, and the single certificate data can be authorized and traced, forming a mechanism for jointly serving multiple business systems. The single certificate data of multiple parties can be written into a unified evidence storage system, and the single certificate data can be provided for multiple business systems to use. Taking the enterprise as the primary key, the single window realizes unified storage of single certificates in each business system, unified use of single certificates, and one-window storage and one-window retrieval.

[0035] The account management module works collaboratively through multiple blockchain nodes to ensure the transparency and security of the account creation process. Each node verifies the validity of the account creation request to prevent the creation of malicious accounts. A multi-factor authentication mechanism, such as digital certificates and biometric identification, is used to ensure the true identity of the account creator. The account information, such as the enterprise name, unified social credit code, and account category, is encrypted and stored on the blockchain to ensure the information's immutability and traceability. The account information can be modified and updated, and all operations are recorded on the blockchain to ensure traceability. The account status, such as account activity, login time, and operation records, is monitored in real-time to detect and handle abnormal behavior in a timely manner. The account cancellation function is also recorded on the blockchain to ensure the integrity and security of the account information.

[0036] The account management module defines different roles (such as enterprise users, tax authorities, and system maintenance personnel) and assigns appropriate permissions to each role. Based on the role definition, each unified authentication trusted account is assigned appropriate operation permissions, such as single certificate verification, management, and inquiry. When the account performs operations, the system verifies the account's permissions in real-time to ensure that only authorized operations can be executed.

[0037] A two-factor authentication mechanism, such as password + SMS verification code and password + biometric identification, is used to improve the security of the account. By analyzing the behavior patterns of the account, abnormal behavior such as frequent login failures and abnormal operations is identified, and measures are taken to prevent the account from being misused. All account operation logs, including login time, operation content, and IP address, are recorded to provide data support for subsequent security audits.

[0038] The single certificate verification module is used to receive the upload request of the unified authentication trusted account and determine the category of the electronic single certificate based on the uploaded electronic single certificate. The electronic single certificate is verified to determine its security level.

[0039] The single certificate data is uploaded and stored to ensure the validity and non-repudiation of the single certificate data of the business system, and a secure, available, and reliable single certificate resource is constructed. Enterprises can manually upload their own single certificates or use the single certificate interface service to collect single certificates uniformly, forming a complete enterprise single certificate storage library to meet the requirements of relevant departments for single certificate retention.

[0040] In some embodiments, the document verification module determines the category of the electronic document, comprising:

[0041] A keyword library is established, which contains professional terms, industry jargons and specific words that may appear in the electronic document of each category. These keywords will be used as the basis for subsequent document classification;

[0042] For each unified authentication trusted account, determine the templates of electronic documents of multiple categories of the unified authentication trusted account, extract multiple initial keywords of the templates of electronic documents of each category according to the keyword library, and determine multiple difference keywords of the templates of electronic documents of each category according to the multiple initial keywords of the templates of electronic documents of each category;

[0043] According to the keyword library, multiple keywords of the electronic document are extracted;

[0044] Obtain the multiple difference keywords of the templates of electronic documents of each category of the unified authentication trusted account for the uploaded electronic document;

[0045] According to the multiple keywords of the electronic document and the multiple difference keywords of the templates of electronic documents of each category of the unified authentication trusted account for the uploaded electronic document, determine the similarity of the electronic document to each category of keywords;

[0046] According to the similarity of the electronic document to each category of keywords, determine the category of the electronic document.

[0047] In some embodiments, the document verification module extracts multiple initial keywords of the templates of electronic documents of each category, comprising:

[0048] For the template of electronic documents of each category, pre-process the template of electronic documents to generate a pre-processed template of electronic documents, wherein the pre-processing includes word segmentation and stop word removal, extract multiple keywords of the template of electronic documents according to the keyword library, determine co-occurrence features of the multiple keywords according to a preset window, establish a keyword co-occurrence undirected graph according to the multiple keyword co-occurrence features, and extract multiple initial keywords of the template of electronic documents according to the keyword co-occurrence undirected graph.

[0049] Specifically, word segmentation is to divide the template text of the electronic document into individual word units, i.e. word items. A word segmentation algorithm in natural language processing is adopted, such as rule-based word segmentation, statistical-based word segmentation or a hybrid word segmentation method combining the two. The purpose of removing stop words is to filter out those words in the text that have no actual meaning for document classification and verification, such as auxiliary words, conjunctions and prepositions. A stop word list can be established, and the text after word segmentation is compared with the stop word list to remove the words appearing in the stop word list, reduce data redundancy, and improve the accuracy and efficiency of keyword extraction.

[0050] The preprocessed text is compared with the keyword library to match the keywords appearing in the text.

[0051] A fixed-size text window (such as a few words or a few sentences) is set, and the co-occurrence of keywords in the window is analyzed, the co-occurring keyword pairs or keyword groups in the preset window and their occurrence times are counted, i.e. the co-occurrence characteristics of multiple keywords. Figure 3 is a schematic diagram of the keyword co-occurrence undirected graph shown in an embodiment of the present application, as shown in Figure 3 The keyword co-occurrence undirected graph consists of nodes and edges, the nodes represent keywords, and the edges represent the co-occurrence relationship between keywords. Each keyword is taken as a node in the graph, and the co-occurring keyword pairs are connected by edges. The weight of the edge can represent the number of co-occurrences.

[0052] For each keyword, the node degree of the node corresponding to the keyword and the sum of the weights of the edges can be determined based on the keyword co-occurrence undirected graph, the keyword weight corresponding to the keyword is calculated according to the node degree of the node corresponding to the keyword and the sum of the weights of the edges, and the keyword whose keyword weight is greater than the keyword weight threshold (for example, 0.1) is taken as the initial keyword. Wherein, for each node in the keyword co-occurrence undirected graph, all adjacent edges are traversed, and the number of edges directly connected to the node is counted, which is the node degree of the node. The keyword weight threshold can be determined according to artificial experience or experimental data.

[0053] For example, the keyword weight corresponding to the keyword can be calculated according to the following formula:

[0054]

[0055] Wherein, is the keyword weight corresponding to the i-th keyword, n i is the node degree of the i-th keyword, N is the total number of edges, w j is the sum of the weights of the edges of the j-th keyword, w i is the sum of the weights of the edges of the i-th keyword, a 11 and a 12is a weight, a 11 and a 12 is greater than 0, a 11 + a 12 = 1, for example, a 11 is 0.5, a 12 is 0.5.

[0056] It can be understood that the above formula calculates the keyword weight by combining the node degree and the sum of edge weights, which not only considers the co-occurrence number of keywords and other keywords (node degree), but also considers the closeness of co-occurrence (sum of edge weights). By calculating the keyword weight, the importance and representativeness of each keyword in the keyword co-occurrence undirected graph can be quantified. The higher the keyword weight, the more likely it is a core or important keyword in the text, ensuring the effectiveness of the initial keywords screened.

[0057] In some embodiments, the single document verification module determines a plurality of difference keywords of the template of each category of electronic single document, including:

[0058] For the template of each category of electronic single document, based on the keyword co-occurrence undirected graph, the node feature of each initial keyword is determined, the recurrence proportion of each initial keyword is determined according to the plurality of initial keywords of the template of each category of electronic single document, the initial keyword whose recurrence proportion is less than the recurrence proportion threshold (0.5) is taken as the difference keyword, and for the initial keyword whose recurrence proportion is greater than or equal to the recurrence proportion threshold, the recurrence difference value of the initial keyword is calculated according to the node feature of the initial keyword in the template of the plurality of categories of electronic single document, and the initial keyword whose recurrence difference value is greater than the recurrence difference threshold (for example, 0.3) is taken as the difference keyword, wherein the recurrence proportion threshold and the recurrence difference threshold can be determined according to artificial experience or experimental data.

[0059] Specifically, the node feature of the initial keyword can be the node degree of the corresponding node and the sum of the weights of the edges.

[0060] For each initial keyword, the proportion of the number of categories taking the keyword as the initial keyword to the number of categories is calculated as the recurrence proportion of the initial keyword.

[0061] The recurrence difference value of the initial keyword can be calculated according to the following formula:

[0062]

[0063] wherein D i is the recurrence difference value of the i-th initial keyword of the current category, n ik is the node degree of the i-th initial keyword of the k-th category, K is the total number of categories, w ikthe sum of the weights of the edges of the i-th initial keyword of the k-th category, b 11 and b 12 is the weight, b 11 and b 12 is greater than 0, b 11 + b 12 = 1, for example, b 11 is 0.5, b 12 is 0.5.

[0064] It can be understood that the above formula quantifies the difference degree of the occurrence frequency and co-occurrence relationship of the keyword in different categories by calculating the difference of the node degree and the sum of the weights of the edges of the keyword in different categories of electronic document templates. By calculating the difference of the occurrence frequency and co-occurrence relationship, the keywords with significant differences can be identified. These keywords have significant differences in the occurrence frequency and co-occurrence relationship in different categories of electronic document templates, and are therefore considered to have uniqueness and difference, providing strong support for subsequent document classification.

[0065] For each category of electronic document template, the system compares the keywords of the uploaded electronic document with the difference keywords of the category, and calculates the similarity between them. The similarity can be calculated by various methods, such as, for example, cosine similarity, Jaccard similarity calculation algorithm, etc. according to the difference keywords in the node degree and the sum of the weights of the edges of the uploaded electronic document and the template. Traverse all categories of electronic document templates, calculate the keyword similarity between the uploaded electronic document and each category, and take the category with the largest keyword similarity as the category of the uploaded electronic document.

[0066] In some embodiments, the document verification module performs verification of the electronic document, including:

[0067] For each unified authentication trusted account, according to the templates of the electronic documents of various categories of the unified authentication trusted account, determine the multiple mandatory fields of each category and the filling rules of each mandatory field, and according to the historical electronic documents of each category of the unified authentication trusted account, determine the verification priority of each mandatory field of each category;

[0068] According to the multiple mandatory fields of the template of the category of the electronic document corresponding to the uploaded electronic document of the unified authentication trusted account, cut multiple fields from the electronic document, and according to the filling rules and the verification priority of each mandatory field of the multiple mandatory fields of the template of the category of the electronic document corresponding to the uploaded electronic document of the unified authentication trusted account, verify the multiple fields.

[0069] Specifically, for each unified authentication trusted account, the system first determines the multiple mandatory fields in each category of electronic document according to the template of the multiple categories of electronic documents involved in the account. These mandatory fields are parts of the electronic document that must be filled in and cannot be empty, and are crucial for the completeness and validity of the document. For each mandatory field, the system formulates detailed filling rules. These rules may include data categories (such as text, numbers, dates, etc.), data formats (such as length restrictions, character set requirements, etc.), filling content (such as specific codes, options, etc.), and whether to allow empty, etc.

[0070] According to each category of historical electronic document of the unified authentication trusted account, the filling of the mandatory fields in these documents is analyzed. Through statistical analysis, it can be understood which mandatory fields are more prone to error or more critical in actual operation, so as to determine the verification priority of each mandatory field. Mandatory fields with high priority will be subjected to more stringent checks during the verification process to ensure the accuracy and completeness of the data in these fields.

[0071] When the user uploads an electronic document, the system first determines the multiple mandatory fields that need to be verified according to the electronic document category and template corresponding to the unified authentication trusted account of the uploaded electronic document. Then, the fields corresponding to these mandatory fields are extracted from the uploaded electronic document, and the extracted fields are verified according to the previously determined filling rules and verification priorities. The verification process may include data category verification, data format verification, content verification, and null verification, etc. After verification, the system will feedback the verification result to the user. If the verification is passed, subsequent operations will be performed; if the verification is not passed, the specific unqualified fields and reasons will be pointed out, and the user will be required to modify and re-upload.

[0072] Through the above steps, the document checking module can effectively verify the uploaded electronic document, ensuring the completeness and validity of the document. This process not only improves the automation level of document processing, but also reduces the risk of human error, providing strong support for the storage and verification of cross-border trade electronic documents. At the same time, by analyzing the historical electronic documents to determine the verification priority, it can more accurately identify and handle critical mandatory fields, further improving the efficiency and accuracy of verification.

[0073] In some embodiments, the document checking module determines the classification level of the electronic document, comprising:

[0074] Query whether the classification level of the electronic document is indicated in the upload request of the unified authentication trusted account, if so, determine the classification level of the electronic document according to the knowledge of the upload request of the unified authentication trusted account;

[0075] If not, according to the category of the electronic document and according to a plurality of mandatory fields, a plurality of fields intercepted from the electronic document are determined to be the first similar electronic document of the uniform authentication trusted account;

[0076] According to the first similar electronic document of the uniform authentication trusted account, the security level of the electronic document is determined.

[0077] Specifically, if the security level of the electronic document is not indicated in the upload request of the uniform authentication trusted account, a plurality of fields are intercepted from the uploaded electronic document according to the category of the electronic document and a plurality of mandatory fields. These fields are key information in the document, which can reflect the main content and characteristics of the document, and these fields are used to search the first similar electronic document of the uniform authentication trusted account in the database. Similar electronic documents refer to those highly similar to the currently uploaded electronic document in terms of category, content or structure.

[0078] The security level of the electronic document is usually classified according to the sensitivity and importance of its content, such as public, internal, confidential, top secret, etc. Different security levels of the document have different security requirements in use, storage and transmission. In the process of determining the security level based on similar electronic documents, similarity analysis is the key. The system compares the field values of the uploaded electronic document with the existing documents in the database, calculates the similarity between them, and finds the most similar electronic document, and the security level of the similar electronic document is used as the security level of the currently uploaded electronic document. The security level determination process follows strict security principles to ensure that only authorized users can access and process high security level electronic documents.

[0079] The document management module is used to store the electronic document to the blockchain through the smart contract according to the category and security level of the electronic document.

[0080] In some embodiments, the document management module stores the electronic document to the blockchain through the smart contract according to the category and security level of the electronic document, including:

[0081] According to the category and security level of the electronic document, the electronic document is fragmented to generate a plurality of electronic document fragments of the electronic document;

[0082] The hash value of each electronic document fragment of the electronic document is calculated;

[0083] The storage node of each electronic document fragment of the electronic document is determined;

[0084] Based on the hash value and the storage node of each electronic document fragment of the electronic document, a document index of the electronic document is established;

[0085] Based on the plurality of fields intercepted from the electronic document according to a plurality of mandatory fields, a search tag of the electronic document is determined;

[0086] The document index and the retrieval tag of the electronic document are stored on the blockchain through the smart contract.

[0087] The plurality of electronic document fragments of the electronic document are distributed stored in the plurality of nodes through the smart contract.

[0088] In some embodiments, the document management module performs sharding processing on the electronic document according to the category and the security level of the electronic document, to generate a plurality of electronic document fragments of the electronic document, including:

[0089] According to the category and the security level of the electronic document, a second similar electronic document of the electronic document is determined.

[0090] According to the cycle retrieval frequency of the second similar electronic document of the electronic document, the electronic document is sharded to generate a plurality of electronic document fragments of the electronic document.

[0091] Specifically, in order to perform reasonable sharding processing on the electronic document, it is necessary to first determine other electronic documents similar to the current electronic document in terms of category and security level, i.e., the second similar electronic document. Similarity analysis helps to identify electronic documents with similar storage requirements and access patterns, thereby providing a basis for the development of sharding strategies. The similarity between electronic documents can be calculated according to the category label, security level identifier, and content features (such as keywords, topics, etc.) of the electronic document. By comparing these features of the current electronic document with the existing documents in the database, the most similar electronic document can be found. The cycle retrieval frequency refers to the number of times an electronic document is retrieved or accessed within a period of time. By analyzing the cycle retrieval frequency of the second similar electronic document, the access pattern and hot data of this type of electronic document can be understood. According to the cycle retrieval frequency of the second similar electronic document, the system can develop different sharding strategies. For example, for electronic documents with high retrieval frequency, they can be divided into smaller fragments to facilitate quick positioning and access; for electronic documents with low retrieval frequency, they can be divided into larger fragments to reduce storage overhead. Through sharding processing, the electronic document can be stored in multiple storage nodes, thereby balancing the storage load and improving the overall performance of the storage system.

[0092] A secure hash algorithm such as SHA-256 can be used to calculate the hash value of each electronic document fragment. According to the security level and fragmentation strategy of the electronic document, the storage node of each electronic document fragment is determined. For high security level document fragments, multiple nodes need to be selected for redundant storage to improve the reliability and security of the data. Distributed storage means that electronic document fragments are stored on multiple nodes rather than centralized storage on a single server. This can reduce the risk of single point failure and improve the overall availability of the system. Based on the hash value of each electronic document fragment and the storage node, a document index of the electronic document is established. The index should include the hash value of each fragment, the storage node address, and the location information of the fragment in the electronic document. The document index is used to quickly locate and retrieve electronic document fragments. When accessing a certain electronic document is needed, the system can find the storage location of all related fragments by querying the index.

[0093] Based on the multiple fields extracted from the electronic document according to the multiple mandatory areas, the retrieval tags of the electronic document are determined. These fields usually contain key information of the electronic document, such as document number, date, amount, etc. Retrieval tags are used to improve the retrieval efficiency of electronic documents. Users can quickly locate the required electronic document by inputting tag information.

[0094] A unique fingerprint can be generated by combining structured data (such as bill of lading number, invoice number, etc.) with file hash, and storing the fingerprint instead of the full text of the document on the chain. This reduces the storage pressure on the chain (saving 90% space) while ensuring tamper resistance.

[0095] Through the smart contract, the document index and retrieval tags of the electronic document are stored on the blockchain. At the same time, multiple electronic document fragments of the electronic document are distributed and stored on multiple nodes. The smart contract ensures the transparency and tamper resistance of the storage process.

[0096] The document retrieval module is used to receive a document retrieval request initiated by a unified authentication trusted account, and according to the document retrieval request, query the target electronic document, and feed back the target electronic document to the unified authentication trusted account.

[0097] Specifically includes:

[0098] According to the authority of the unified authentication trusted account initiating the document retrieval request to the multiple stored single documents, a plurality of candidate stored single documents are determined;

[0099] According to the retrieval tags of the multiple candidate stored single documents and the retrieval keywords of the document retrieval request, a target electronic document is determined;

[0100] According to the document index of the target electronic document, multiple electronic document fragments of the target electronic document are obtained;

[0101] The multiple electronic document fragments of the target electronic document are spliced to generate the target electronic document, which is fed back to the unified authentication trusted account that initiates the document retrieval request.

[0102] Specifically, when the unified authentication trusted account initiates the document retrieval request, the document retrieval module first verifies the access rights of the account to the stored electronic documents. This step ensures that only accounts with corresponding rights can retrieve specific electronic documents, thereby protecting the security and privacy of data. Based on the rights of the account, the module filters out multiple candidate stored electronic documents. These documents are accessible to the account and may meet the conditions of the retrieval request.

[0103] The retrieval keywords in the document retrieval request are matched with the retrieval labels of the candidate documents. The matching process may involve string comparison, fuzzy matching or semantic analysis techniques to ensure that the relevant documents can be accurately found.

[0104] According to the matching result, the module determines the target electronic document that best meets the retrieval request. If there are multiple matching documents, the module may sort them according to the matching degree or other standards (such as timestamp, version number, etc.) and select the most relevant one or more as the target.

[0105] According to the document index of the target electronic document, the module obtains all fragments of the document from the corresponding node on the blockchain. Since the document is stored in fragments, this step needs to ensure that all fragments can be obtained completely and accurately.

[0106] After obtaining all fragments of the target electronic document, the module splices these fragments in the order recorded in the document index to restore the complete electronic document, and feeds back the spliced target electronic document to the unified authentication trusted account that initiates the document retrieval request. The user can view or download the document on the account interface to meet their business needs.

[0107] The storage, chaining, authorization, calling and other use process information logs of the document data are recorded on the chain, and all operation links can be checked and verified on the chain to ensure that the data is real and reliable, and the document data is traceable. Users can use unified accounting smart contracts and hash calculation technologies to store, chain, query and use trusted data and stored data of third parties in a unified and data standard-compliant manner.

[0108] The system can also have report and analysis tools that can statistically analyze the documents and generate various forms of reports.

[0109] Figure 4 is the flowchart of the blockchain-based cross-border trade electronic document storage and verification method shown in an embodiment of the present application, as Figure 4As shown, the blockchain-based cross-border trade electronic document storage and verification method can include the following steps.

[0110] A plurality of unified authentication trusted accounts are established by a plurality of blockchain nodes, and the permissions of the plurality of unified authentication trusted accounts are managed;

[0111] An upload request of the unified authentication trusted account is received, and the category of the electronic document is determined according to the uploaded electronic document, and the electronic document is verified to determine the secret level of the electronic document;

[0112] According to the category and secret level of the electronic document, the electronic document is stored to the blockchain through the smart contract;

[0113] The document retrieval request initiated by the unified authentication trusted account is received, and the target electronic document is queried according to the document retrieval request, and the target electronic document is fed back to the unified authentication trusted account.

[0114] The blockchain-based cross-border trade electronic document storage and verification method can be applied to the blockchain-based cross-border trade electronic document storage and verification system described above, which will not be described here.

[0115] Finally, it should be understood that the embodiments described in the specification are only used to illustrate the principles of the embodiments of the specification. Other variations can also belong to the scope of the specification. Therefore, as an example but not limitation, alternative configurations of the embodiments of the specification can be considered consistent with the teachings of the specification. Accordingly, the embodiments of the specification are not limited to the embodiments explicitly introduced and described in the specification.

Claims

1. A blockchain-based cross-border trade electronic document storage and verification system, characterized in that, include: The account management module is used to establish multiple unified authentication trusted accounts through multiple blockchain nodes and manage the permissions of multiple unified authentication trusted accounts; The document verification module is used to receive upload requests from unified authentication trusted accounts, determine the category of the electronic document based on the uploaded electronic document, verify the electronic document, and determine the security level of the electronic document. The document management module is used to store electronic documents on the blockchain through smart contracts, based on the category and security level of the electronic documents. The document retrieval module is used to receive document retrieval requests initiated by the unified authentication trusted account, query the target electronic document according to the document retrieval request, and return the target electronic document to the unified authentication trusted account. The document verification module determines the category of electronic documents, including: Establish a keyword database; For each unified authentication trusted account, determine the templates of multiple categories of electronic documents for the unified authentication trusted account. Based on the keyword library, extract multiple initial keywords for the templates of each category of electronic documents. Based on the multiple initial keywords for the templates of each category of electronic documents, determine multiple differential keywords for the templates of each category of electronic documents. Extract multiple keywords from the keyword database for electronic documents; Obtain multiple different keywords for the template of each type of electronic document from the unified authentication trusted account of the uploaded electronic documents; Based on multiple keywords of electronic documents and multiple different keywords of the templates of electronic documents for each category of uploaded electronic documents from the unified authentication trusted account, the similarity between electronic documents and keywords of each category is determined. The category of the electronic document is determined based on the keyword similarity between the electronic document and each category; The document verification module determines multiple distinguishing keywords for the templates of each category of electronic documents, including: For each type of electronic document template, based on the keyword co-occurrence undirected graph, the node characteristics of each initial keyword are determined. Based on multiple initial keywords in each type of electronic document template, the recurrence ratio of each initial keyword is determined. Initial keywords with a recurrence ratio less than a recurrence ratio threshold are designated as differential keywords. For initial keywords with a recurrence ratio greater than or equal to the recurrence ratio threshold, the recurrence difference value of the initial keyword is calculated based on the node characteristics of the initial keyword in multiple types of electronic document templates. Initial keywords with a recurrence difference value greater than the recurrence difference threshold are designated as differential keywords. The keyword co-occurrence undirected graph consists of nodes and edges. Nodes represent keywords, and edges represent the co-occurrence relationships between keywords. Each keyword is treated as a node in the graph, and co-occurring keyword pairs are connected by edges. The edge weight represents the number of times the keywords co-occur. The node characteristic of the initial keyword is the sum of the node degree and the edge weight. By calculating the differences in node degree and the sum of edge weights of the initial keyword in different types of electronic document templates, the degree of difference in the frequency of occurrence and co-occurrence relationships of the initial keyword in different categories is quantified, and the recurrence difference value is calculated.

2. The blockchain-based cross-border trade electronic document storage and verification system according to claim 1, characterized in that, The document verification module extracts multiple initial keywords from the templates of each type of electronic document, including: For each type of electronic document template, the template is preprocessed to generate a preprocessed electronic document template. The preprocessing includes word segmentation and stop word removal. Based on a keyword library, multiple keywords of the electronic document template are extracted. Based on a preset window, the co-occurrence features of multiple keywords are determined. Based on the co-occurrence features of multiple keywords, an undirected keyword co-occurrence graph is established. Based on the undirected keyword co-occurrence graph, multiple initial keywords of the electronic document template are extracted.

3. The blockchain-based cross-border trade electronic document storage and verification system according to claim 1, characterized in that, The document verification module verifies electronic documents, including: For each unified authentication trusted account, based on the templates of various types of electronic documents of the unified authentication trusted account, determine multiple required fields for each category and the filling rules for each required field. Based on the historical electronic documents of each category of the unified authentication trusted account, determine the verification priority of each required field for each category. Based on the multiple required fields of the template of the electronic document corresponding to the unified authentication trusted account of the uploaded electronic document, multiple fields are extracted from the electronic document. According to the filling rules of the multiple required fields of the template of the electronic document corresponding to the unified authentication trusted account of the uploaded electronic document and the verification priority of each required field, multiple fields are verified.

4. The blockchain-based cross-border trade electronic document storage and verification system according to claim 3, characterized in that, The document verification module determines the security level of electronic documents, including: Check if the upload request of the unified authentication trusted account indicates the security level of the electronic document. If so, determine the security level of the electronic document based on the knowledge of the upload request of the unified authentication trusted account. If not, determine the first similar electronic document for the unified authentication trusted account based on the type of electronic document and multiple fields extracted from the electronic document based on multiple required fields; The security level of the electronic document is determined based on the first similar electronic document of the unified authentication trusted account.

5. The blockchain-based cross-border trade electronic document storage and verification system according to claim 4, characterized in that, The document management module stores electronic documents on the blockchain via smart contracts based on the category and security level of the electronic documents, including: Based on the category and security level of the electronic document, the electronic document is segmented to generate multiple electronic document fragments. Calculate the hash value of each electronic document segment; Determine the storage node for each electronic document segment; Based on the hash value and storage node of each electronic document fragment, a document index is established for the electronic document. The search tags for the electronic document are determined based on multiple fields extracted from the electronic document according to multiple required fields; The document index and retrieval tags of the electronic documents are stored on the blockchain through smart contracts; Smart contracts are used to distribute and store multiple electronic document fragments across multiple nodes.

6. The blockchain-based cross-border trade electronic document storage and verification system according to claim 5, characterized in that, The document management module segments electronic documents according to their category and security level, generating multiple electronic document fragments, including: Based on the category and security level of the electronic document, identify the second similar electronic document; Based on the periodic retrieval frequency of the second similar electronic document, the electronic document is segmented to generate multiple electronic document fragments.

7. The blockchain-based cross-border trade electronic document storage and verification system according to claim 5, characterized in that, The document retrieval module queries the target electronic document according to the document retrieval request and sends the target electronic document back to the unified authentication trusted account, including: Based on the permissions of the unified authentication trusted account that initiated the document retrieval request to multiple stored document documents, multiple candidate stored document documents are determined. The target electronic document is determined based on the search tags of multiple candidate stored electronic documents and the search keywords of the document search request. Based on the document index of the target electronic document, obtain multiple electronic document fragments of the target electronic document; Multiple electronic document fragments of the target electronic document are spliced ​​together to generate the target electronic document, which is then sent back to the unified authentication trusted account that initiated the document retrieval request.

8. A blockchain-based method for storing and verifying electronic documents in cross-border trade, characterized in that: The blockchain-based cross-border trade electronic document storage and verification system described in claim 1 includes: Establish multiple unified authentication trusted accounts through multiple blockchain nodes, and manage the permissions of multiple unified authentication trusted accounts; Receive upload requests from unified authentication trusted accounts, determine the category of electronic documents based on the uploaded electronic documents, verify the electronic documents, and determine the security level of the electronic documents; Based on the category and security level of the electronic documents, the electronic documents are stored on the blockchain through smart contracts; It receives document retrieval requests initiated by a unified authentication trusted account, queries the target electronic document according to the document retrieval request, and sends the target electronic document back to the unified authentication trusted account.

Images