Customs import and export declaration method, system, equipment and medium based on intelligent coding

Abstract

The present invention proposes a customs import and export declaration method, system, device, and medium based on intelligent coding. The method includes obtaining customs documents and extracting customs declaration information, configuring declaration rules based on the customs declaration information; obtaining commodity data, building a customs declaration model based on the commodity data and declaration rules, training the customs declaration model through commodity data increments and declaration rule increments, and outputting the customs code and declaration elements corresponding to the target commodity; generating customs declaration documents corresponding to the target commodity based on the customs code and declaration elements; accessing a customs declaration service system, submitting the customs declaration documents through a message queue, and verifying the receipt documents returned by the customs declaration documents. The present invention constructs a declaration rule database, calls and associates coding rules and element rules with commodity data to build a coding model, outputs the customs code and declaration elements of the target commodity based on the trained model, prepares and declares in real time, reduces document preparation errors and document preparation costs, and improves the efficiency and success rate of customs declaration.

Description

Technical Field

[0001] The present invention relates to the technical field of customs clearance data processing, and in particular to a customs import and export declaration method, system, equipment and medium based on intelligent coding. Background Art

[0002] With the rapid development of cross-border e-commerce, foreign trade companies have an increasing demand for customs declarations for import and export goods. Customs declarations involve logistics and transportation, export declarations, and tax refund processing. The declaration elements and customs codes in the commodity data of customs declarations are interrelated. The declaration elements and customs codes are subject to change according to policies. The correctness of the declaration elements and customs codes is the key factor in determining the success of customs declarations.

[0003] The existing customs clearance declaration system has the following defects:

[0004] (1) The data of each business document in the customs declaration document is not related, which makes the generation and verification of documents inefficient and prone to errors, affecting the success rate of customs declaration;

[0005] (2) If the above business documents are sorted manually, the workload will be heavy and prone to manual errors, resulting in low declaration efficiency, untimely responses, and inability to process massive orders. Summary of the Invention

[0006] The technical problem to be solved by the present invention is that, in response to the above-mentioned defects of the prior art, the present invention proposes a customs import and export declaration method, system, equipment and medium based on intelligent coding.

[0007] The first aspect of the present invention discloses a customs import and export declaration method based on intelligent coding, comprising the following steps:

[0008] S1: Obtain customs documents and extract customs declaration information, and configure declaration rules based on the customs declaration information. The declaration rules include customs coding rules and declaration element rules;

[0009] S2: Obtain commodity data, build a customs declaration model based on the commodity data and the declaration rules, train the customs declaration model based on the NLP algorithm with incremental commodity data and incremental declaration rules, and output the customs code and declaration elements corresponding to the target commodity based on the intelligent matching algorithm;

[0010] S3: Generate customs declaration documents corresponding to the target commodity based on the customs code and declaration elements output by the customs declaration model, encrypt and store the customs declaration documents, and add data signatures to the customs declaration documents. The customs declaration documents include export orders, payment orders, logistics orders, and declaration lists.

[0011] S4: Access the customs declaration service system, submit the customs declaration documents through the message queue, and verify the receipt document returned by the customs declaration documents.

[0012] In an optional embodiment, the customs declaration model is trained based on an NLP algorithm by increasing commodity data and increasing declaration rules, including the following steps:

[0013] S201: Read historical declaration rules and historical commodity data;

[0014] S202: Read the declaration rule increment and the commodity data increment;

[0015] S203: Training the customs declaration model to obtain a first representation model that converts the commodity text representation into a mathematical vector, where the commodity text representation is commodity data that meets the declaration rules;

[0016] S204: Train the first representation model based on the fasttext model, perform embedding processing on each word in the first representation model and then average them to obtain a document vector, predict the document vector through a neural network, obtain all declaration rule labels and customs code labels, and output the trained customs declaration model.

[0017] In an optional embodiment, outputting the customs code and declaration elements corresponding to the target commodity based on the intelligent matching algorithm includes the following steps:

[0018] S205: Calling the product information of the target product and generating a second representation model of the product information;

[0019] S206: Call the trained customs declaration model, perform embedding processing on the second representation model, calculate the distance between the target commodity after embedding processing and each declaration rule label and each customs code label under the same category of the target commodity, and output the declaration elements and customs code corresponding to the target commodity based on the condition of minimum distance.

[0020] In an optional embodiment, the data in the first characterization model and the second characterization model are correlated with each other, and the data in the characterization model include the Chinese and English names of the product, the final level components of the product, the 1st to 5th level components of the product, and classification elements.

[0021] In an optional embodiment, obtaining customs documents and extracting customs declaration information, and configuring declaration rules based on the customs declaration information, includes the following steps:

[0022] S101: Obtain commodity code documents and customs declaration agreement templates from customs documents;

[0023] S102: Parse the commodity coding document, extract the customs coding rules, import them into the database, and build a customs coding rules database;

[0024] S103: Parse the customs declaration agreement template, extract the declaration element rules and import them into the data, and build a declaration element rule database;

[0025] S104: Regularly check for customs document updates. If new coding rules or new declaration element rules are found, the incremental coding rules or incremental declaration element rules are imported into the customs coding rule database or the element declaration rule database, and database updates are performed regularly.

[0026] In an optional embodiment, the customs coding rules database includes classification rules corresponding to commodity codes, and the categories of the classification rules include commodity composition, commodity appearance, commodity specifications and commodity models. The customs coding rules database can perform batch deactivation / activation of customs coding rules, import / export of customs coding rules and testing of customs coding rules.

[0027] In an optional embodiment, the declaration element rule database includes declaration element specifications of commodity data, and the declaration element specifications include commodity data source specifications and commodity value logic specifications. The declaration element rule database can execute batch deactivation / activation of declaration element rules, copying / addition of declaration element rules, and testing of declaration element rules.

[0028] In an optional embodiment, generating customs declaration documents corresponding to the commodity according to the customs code and declaration elements output by the customs declaration model includes the following steps:

[0029] S301: Obtain the customs declaration agreement template;

[0030] S302: Confirming the filling elements and customs code of the target commodity information in the customs declaration document based on the declaration elements and customs code;

[0031] S303: Output the complete customs declaration data and fill it into the customs declaration agreement template to generate a customs declaration document.

[0032] In an optional embodiment, encrypting and storing the customs declaration document and signing the data includes:

[0033] S304: Detecting sensitive data in the customs declaration document, and encrypting and storing the sensitive data using the AES algorithm;

[0034] S305: After encryption and storage, the customs declaration document is signed by encryption hardware.

[0035] In an optional embodiment, the transmission method for submitting the customs declaration documents through the message queue includes HTTP request and RabbitMQ communication. The RabbitMQ communication temporarily stores the customs declaration documents in the RabbitMQ server, captures the customs declaration documents through the customs declaration system, and then returns the corresponding receipt file.

[0036] The second aspect of the present invention discloses a customs import and export declaration system based on intelligent coding, the system comprising:

[0037] A rule configuration module is used to obtain customs documents and extract customs declaration information, and configure declaration rules based on the customs declaration information. The declaration rules include customs coding rules and declaration element rules;

[0038] A rule matching module is used to obtain commodity data, build a customs declaration model based on the commodity data and the declaration rules, train the customs declaration model based on incremental commodity data and incremental declaration rules using an NLP algorithm, and output the customs code and declaration elements corresponding to the target commodity based on an intelligent matching algorithm;

[0039] A document preparation module is used to generate customs declaration documents corresponding to the target commodity based on the customs code and declaration elements output by the customs declaration model, and to encrypt, store and sign the customs declaration documents. The customs declaration documents include export orders, payment orders, logistics orders and declaration lists;

[0040] The document declaration module is used to access the customs declaration service system, submit the customs declaration documents through the message queue, and verify the receipt documents returned by the customs declaration documents.

[0041] In an optional embodiment, the rule matching module includes:

[0042] A machine learning module is configured to read historical declaration rules and historical commodity data; read increments of the declaration rules and the commodity data; train the customs declaration model to obtain a first representation model that converts the commodity text representation into a mathematical vector, where the commodity text representation is commodity data that satisfies the declaration rules; train the first representation model based on a fasttext model, embed each word in the first representation model and then average it to obtain a document vector, predict the document vector through a neural network, obtain all declaration rule labels and customs code labels, and output the trained customs declaration model.

[0043] In an optional embodiment, the rule matching module further includes:

[0044] The intelligent matching module is configured to call the commodity information of the target commodity and generate a second representation model of the commodity information; call the trained customs declaration model, perform embedding processing on the second representation model, calculate the distance between the target commodity after embedding processing and each declaration rule label and each customs code label under the same category of the target commodity, and output the declaration elements and customs code corresponding to the target commodity based on the condition of minimum distance.

[0045] In an optional embodiment, the rule configuration module includes:

[0046] The rule database module is used to obtain commodity coding documents and customs declaration agreement templates in customs documents; parse the commodity coding documents, extract customs coding rules, import them into the database, and build a customs coding rules database; parse the customs declaration agreement template, extract declaration element rules, import them into the data, and build a declaration element rules database; regularly detect customs document updates, and if new coding rules or new declaration element rules are parsed, import the coding rule increments or declaration element increments into the customs coding rules database or the element declaration rules database, and regularly perform database updates.

[0047] In an optional embodiment, the document preparation module includes:

[0048] The customs declaration document generation module is used to obtain a customs declaration agreement template; confirm the filling elements and customs codes of the target commodity information in the customs declaration document based on the declaration elements and customs codes; output complete customs declaration data and fill it into the customs declaration agreement template to generate a customs declaration document.

[0049] In an optional embodiment, the document preparation module further includes:

[0050] The security encryption module is used to detect sensitive data in the customs declaration document and encrypt and store the sensitive data using the AES algorithm; after encryption and storage, the customs declaration document is signed by the encryption machine hardware.

[0051] The third aspect of the present invention discloses a customs import and export declaration device based on intelligent coding, comprising:

[0052] at least one processor, and

[0053] a memory communicatively connected to the at least one processor; wherein,

[0054] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the customs import and export declaration method based on intelligent coding as disclosed in any one of the first aspects of the present invention.

[0055] The fourth aspect of the present invention discloses a computer-readable storage medium, which stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the customs import and export declaration method based on intelligent coding as described in any one of the items disclosed in the first aspect of the present invention.

[0056] Compared with the prior art, the present invention has the following advantages:

[0057] (1) The present invention ensures that customs declaration documents are configured according to real-time rules and regularly update declaration rules through rule configuration. Customs declaration documents are real-time, which greatly reduces the error rate in making customs declaration documents and improves the accuracy of customs declaration documents.

[0058] (2) The present invention trains a customs declaration model by associating declaration element rules with customs code rules, outputting customs codes and declaration elements corresponding to the declared goods, thereby avoiding customs clearance failures caused by inaccurate declaration elements or customs codes in the prepared customs declaration documents, reducing document preparation errors, and improving the customs declaration success rate;

[0059] (3) The present invention uses the declaration system to create and push documents to the customs system in real time, integrating multiple businesses such as logistics transportation, export declaration, and tax refund processing in customs declaration. There is no need for manual sorting, which lowers the declaration threshold and significantly improves the declaration efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0061] Figure 1 This is a flow chart of the customs import and export declaration method based on intelligent coding of the present invention;

[0062] Figure 2 This is a schematic diagram of the customs import and export declaration system based on intelligent coding of the present invention. DETAILED DESCRIPTION

[0063] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0064] In the description of the present invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely to facilitate and simplify the description of the present invention. They are not intended to indicate or imply that the devices or components referred to must have, be constructed, or operate in a specific orientation, and therefore should not be construed as limiting the present invention. Furthermore, the terms "first," "second," "third," and "fourth," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0065] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed, detachable, or integral connections; mechanical or electrical connections; direct or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.

[0066] Example 1

[0067] See also Figure 1 The embodiment of the present invention discloses a customs import and export declaration method based on intelligent coding, comprising the following steps:

[0068] S1: Obtain customs documents and extract customs declaration information, and configure declaration rules based on the customs declaration information. The declaration rules include customs coding rules and declaration element rules;

[0069] S2: Obtain commodity data, build a customs declaration model based on the commodity data and the declaration rules, train the customs declaration model based on the NLP algorithm with incremental commodity data and incremental declaration rules, and output the customs code and declaration elements corresponding to the target commodity based on the intelligent matching algorithm;

[0070] S3: Generate customs declaration documents corresponding to the target commodity based on the customs code and declaration elements output by the customs declaration model, encrypt and store the customs declaration documents, and add data signatures to the customs declaration documents. The customs declaration documents include export orders, payment orders, logistics orders, and declaration lists.

[0071] S4: Access the customs declaration service system, submit the customs declaration documents through the message queue, and verify the receipt document returned by the customs declaration documents.

[0072] It should be noted that the data of declaration rule configuration in the declaration method, the data of the customs declaration model and the data of the historical commodity declaration process are all shared in the declaration system. Based on business needs, the configured declaration rule data, the commodity declaration elements (i.e. commodity data specifications and how to fill in customs declaration document specifications) and commodity customs codes that have been matched according to the declaration rules in the customs declaration model data can be flexibly called. After the customs updates the customs coding rules and element declaration rules in the declaration rules, the rule configuration is updated in real time, and the newly added customs coding rule increments and element declaration rule increments are imported into the training process of the customs declaration model, thereby solving the problem of customs declaration document preparation in real time.

[0073] In an optional embodiment, the customs declaration model is trained based on an NLP algorithm by increasing commodity data and increasing declaration rules, including the following steps:

[0074] S201: Read historical declaration rules and historical commodity data;

[0075] S202: Read the declaration rule increment and the commodity data increment;

[0076] S203: Training the customs declaration model to obtain a first representation model that converts the commodity text representation into a mathematical vector, where the commodity text representation is commodity data that meets the declaration rules;

[0077] S204: Train the first representation model based on the fasttext model, perform embedding processing on each word in the first representation model and then average them to obtain a document vector, predict the document vector through a neural network, obtain all declaration rule labels and customs code labels, and output the trained customs declaration model.

[0078] In an optional embodiment, outputting the customs code and declaration elements corresponding to the target commodity based on the intelligent matching algorithm includes the following steps:

[0079] S205: Calling the product information of the target product and generating a second representation model of the product information;

[0080] S206: Call the trained customs declaration model, perform embedding processing on the second representation model, calculate the distance between the target commodity after embedding processing and each declaration rule label and each customs code label under the same category of the target commodity, and output the declaration elements and customs code corresponding to the target commodity based on the condition of minimum distance.

[0081] In an optional embodiment, the data in the first characterization model and the second characterization model are correlated with each other, and the data in the characterization model include the Chinese and English names of the product, the final level components of the product, the 1st to 5th level components of the product, and classification elements.

[0082] In an optional embodiment, obtaining customs documents and extracting customs declaration information, and configuring declaration rules based on the customs declaration information, includes the following steps:

[0083] S101: Obtain commodity code documents and customs declaration agreement templates from customs documents;

[0084] S102: Parse the commodity coding document, extract the customs coding rules, import them into the database, and build a customs coding rules database;

[0085] S103: Parse the customs declaration agreement template, extract the declaration element rules and import them into the data, and build a declaration element rule database;

[0086] S104: Regularly check for customs document updates. If new coding rules or new declaration element rules are found, the incremental coding rules or incremental declaration element rules are imported into the customs coding rule database or the element declaration rule database, and database updates are performed regularly.

[0087] It should be noted that the customs declaration documents include four types of documents (export orders, payment orders, logistics waybills and declaration lists). All customs declaration documents can be in XML format, which is an extensible markup language used to mark data and define data types. For example, the information in the export order customs declaration document includes orderer information, order number, and product information. This part of information is saved in the database for easy call. When generating the order customs declaration document, this part of information data is called, combined with the data in the configured declaration rule database, and the customs code and declaration elements matched with the goods output after the customs declaration model training. The customs code is used to match the goods and make a unique identification, and the declaration elements involve the data, data specifications and declaration specifications that need to be filled in the order customs declaration document. The declaration elements of different goods are different, and the content of the documents filled in in different customs declaration documents is different. Therefore, it is necessary to determine the customs code and declaration elements matching the goods before declaration, so as to generate a customs declaration document that meets the customs declaration requirements.

[0088] More specifically, the export orders in the four types of customs declaration documents are export orders generated by the enterprise background after the cross-border e-commerce service platform customers place an order. The export order customs declaration document contains an order header node, an order product body node and an information transmission node. The above-mentioned order header node needs to generate a GUID, which is used to include the product name, product description and product price of the order product in the above-mentioned order header node. The GUID is a serial number used to uniquely identify the order, which is composed of uppercase letters, numbers and "-".

[0089] More specifically, the payment slip in the four types of customs declaration documents is generated by the e-commerce company's backend after the customer completes the payment on the cross-border e-commerce service platform. The payment slip information includes the payer information, payment amount, order number and payment slip number, etc. The payment slip customs declaration file contains the payment slip header node and the transmission information node. The above-mentioned payment slip header node also needs to generate a GUID. The payment slip information will involve the tax refund processing business in the customs business. If there is an error in the payment slip customs declaration file, it will affect the subsequent business development.

[0090] More specifically, the logistics waybills in the four types of customs declaration documents transmit information such as product attributes and product destination to the logistics system through the e-commerce service platform. The logistics system allocates logistics waybills to the corresponding products. The logistics waybills include consignee information, order number, logistics waybill number, product information, etc. The logistics waybill customs declaration file contains the logistics waybill header node and information transmission node, and GUID also needs to be generated.

[0091] In an optional embodiment, the customs coding rules database includes classification rules corresponding to commodity codes, and the categories of the classification rules include commodity composition, commodity appearance, commodity specifications and commodity models. The customs coding rules database can perform batch deactivation / activation of customs coding rules, import / export of customs coding rules and testing of customs coding rules.

[0092] In an optional embodiment, the declaration element rule database includes declaration element specifications of commodity data, and the declaration element specifications include commodity data source specifications and commodity value logic specifications. The declaration element rule database can execute batch deactivation / activation of declaration element rules, copying / addition of declaration element rules, and testing of declaration element rules.

[0093] It should be noted that the declaration element rules and customs code rules can be turned on and off according to actual business needs, and the declaration element rule database and the customs code database are linked to the customs code model in real time. When it is necessary to output the declaration elements and customs codes corresponding to the goods through the customs code model, the customs code model will be restricted by the declaration element rule database and the rules that have been turned on in the customs code database. By controlling the turning on and off of the declaration rules, the entire customs declaration process can be fully controlled. The declaration element rules and customs code rules in the declaration rules can be batch tested in the database to ensure that the declaration rules that need to be turned on and off during the declaration process are accurate, thereby improving the accuracy of customs declaration document preparation and the success rate of customs clearance.

[0094] In an optional embodiment, generating customs declaration documents corresponding to the commodity according to the customs code and declaration elements output by the customs declaration model includes the following steps:

[0095] S301: Obtain the customs declaration agreement template;

[0096] S302: Confirming the filling elements and customs code of the target commodity information in the customs declaration document based on the declaration elements and customs code;

[0097] S303: Output the complete customs declaration data and fill it into the customs declaration agreement template to generate a customs declaration document. In an optional embodiment, the customs declaration document is encrypted and stored and the data is signed, including:

[0098] S304: Detecting sensitive data in the customs declaration document, and encrypting and storing the sensitive data using the AES algorithm;

[0099] S305: After encryption and storage, the customs declaration document is signed by encryption hardware.

[0100] It should be noted that

[0101] In an optional embodiment, the transmission method for submitting the customs declaration documents through the message queue includes HTTP request and RabbitMQ communication. The RabbitMQ communication temporarily stores the customs declaration documents in the RabbitMQ server, captures the customs declaration documents through the customs declaration system, and then returns the corresponding receipt file.

[0102] The present invention differs from other existing customs electronic declaration technologies, such as intelligent commodity classification, text anomaly detection, and cargo information identification and verification. These existing technologies are unable to intelligently generate declaration documents (four types of documents) with data associations under multiple declaration rules. Furthermore, they are unable to flexibly adjust the filling of declaration documents in real time according to business needs and the requirements of the customs electronic declaration system, completing the full-process intelligent declaration of customs declaration business. The present invention configures and regularly updates declaration rules to ensure that customs declaration documents are configured according to real-time rules, making customs declaration documents real-time, greatly reducing the error rate in customs declaration document preparation and improving the accuracy of customs declaration documents. The present invention trains a customs declaration model by associating declaration element rules with customs code rules, outputting the customs codes and declaration elements corresponding to the declared goods, avoiding customs clearance failures caused by inaccurate declaration elements or customs codes in the prepared declaration documents, reducing document preparation errors, and improving the customs declaration success rate. The present invention uses the declaration system to generate and push documents to the customs system in real time, integrating multiple customs declaration operations such as logistics and transportation, export declaration, and tax refund processing, eliminating the need for manual processing, lowering the declaration threshold, and significantly improving declaration efficiency and accuracy.

[0103] like Figure 2 As shown, the second aspect of the present invention discloses a customs import and export declaration system based on intelligent coding, the system comprising:

[0104] A rule configuration module is used to obtain customs documents and extract customs declaration information, and configure declaration rules based on the customs declaration information. The declaration rules include customs coding rules and declaration element rules;

[0105] A rule matching module is used to obtain commodity data, build a customs declaration model based on the commodity data and the declaration rules, train the customs declaration model based on incremental commodity data and incremental declaration rules using an NLP algorithm, and output the customs code and declaration elements corresponding to the target commodity based on an intelligent matching algorithm;

[0106] A document preparation module is used to generate customs declaration documents corresponding to the target commodity based on the customs code and declaration elements output by the customs declaration model, and to encrypt, store and sign the customs declaration documents. The customs declaration documents include export orders, payment orders, logistics orders and declaration lists;

[0107] The document declaration module is used to access the customs declaration service system, submit the customs declaration documents through the message queue, and verify the receipt documents returned by the customs declaration documents.

[0108] In an optional embodiment, the rule matching module includes:

[0109] A machine learning module is configured to read historical declaration rules and historical commodity data; read increments of the declaration rules and the commodity data; train the customs declaration model to obtain a first representation model that converts the commodity text representation into a mathematical vector, where the commodity text representation is commodity data that satisfies the declaration rules; train the first representation model based on a fasttext model, embed each word in the first representation model and then average it to obtain a document vector, predict the document vector through a neural network, obtain all declaration rule labels and customs code labels, and output the trained customs declaration model.

[0110] In an optional embodiment, the rule matching module further includes:

[0111] The intelligent matching module is configured to call the commodity information of the target commodity and generate a second representation model of the commodity information; call the trained customs declaration model, perform embedding processing on the second representation model, calculate the distance between the target commodity after embedding processing and each declaration rule label and each customs code label under the same category of the target commodity, and output the declaration elements and customs code corresponding to the target commodity based on the condition of minimum distance.

[0112] In an optional embodiment, the rule configuration module includes:

[0113] The rule database module is used to obtain commodity coding documents and customs declaration agreement templates in customs documents; parse the commodity coding documents, extract customs coding rules, import them into the database, and build a customs coding rules database; parse the customs declaration agreement template, extract declaration element rules, import them into the data, and build a declaration element rules database; regularly detect customs document updates, and if new coding rules or new declaration element rules are parsed, import the coding rule increments or declaration element increments into the customs coding rules database or the element declaration rules database, and regularly perform database updates.

[0114] In an optional embodiment, the document preparation module includes:

[0115] The customs declaration document generation module is used to obtain a customs declaration agreement template; confirm the filling elements and customs codes of the target commodity information in the customs declaration document based on the declaration elements and customs codes; output complete customs declaration data and fill it into the customs declaration agreement template to generate a customs declaration document.

[0116] In an optional embodiment, the document preparation module further includes:

[0117] The security encryption module is used to detect sensitive data in the customs declaration document and encrypt and store the sensitive data using the AES algorithm; after encryption and storage, the customs declaration document is signed by the encryption machine hardware.

[0118] The third aspect of the present invention discloses a customs import and export declaration device based on intelligent coding, comprising:

[0119] at least one processor, and

[0120] a memory communicatively connected to the at least one processor; wherein,

[0121] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the customs import and export declaration method based on intelligent coding as disclosed in any one of the first aspects of the present invention.

[0122] The fourth aspect of the present invention discloses a computer-readable storage medium, which stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the customs import and export declaration method based on intelligent coding as described in any one of the items disclosed in the first aspect of the present invention.

[0123] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A customs import and export declaration method based on intelligent coding, characterized in that: The method comprises the following steps: S1: Obtain customs documents and extract customs declaration information, and configure declaration rules based on the customs declaration information. The declaration rules include customs coding rules and declaration element rules; S2: Obtain commodity data, build a customs declaration model based on the commodity data and the declaration rules, train the customs declaration model based on the NLP algorithm with incremental commodity data and incremental declaration rules, and output the customs code and declaration elements corresponding to the target commodity based on the intelligent matching algorithm; S3: Generate customs declaration documents corresponding to the target commodity based on the customs code and declaration elements output by the customs declaration model, encrypt and store the customs declaration documents, and add data signatures to the customs declaration documents. The customs declaration documents include export orders, payment orders, logistics orders, and declaration lists. S4: Access the customs declaration service system, submit the customs declaration documents through the message queue, and verify the receipt document returned by the customs declaration documents; The steps of obtaining customs documents and extracting customs declaration information, and configuring declaration rules based on the customs declaration information, include the following steps: S101: Obtain commodity code documents and customs declaration agreement templates from customs documents; S102: Parse the commodity coding document, extract the customs coding rules, import them into the database, and build a customs coding rules database; S103: Parse the customs declaration agreement template, extract the declaration element rules and import them into the data, and build a declaration element rule database; S104: Regularly check for customs document updates. If new coding rules or new declaration element rules are found, the incremental coding rules or declaration element increments are imported into the customs coding rule database or the declaration element rule database, and database updates are performed regularly. The NLP algorithm-based training of the customs declaration model through incremental commodity data and incremental declaration rules includes the following steps: S201: Read historical declaration rules and historical commodity data; S202: Read the declaration rule increment and the commodity data increment; S203: Training the customs declaration model to obtain a first representation model that converts the commodity text representation into a mathematical vector, where the commodity text representation is commodity data that meets the declaration rules; S204: Training the first representation model based on the FastText model, embedding each word in the first representation model and averaging them to obtain a document vector, predicting the document vector using a neural network, obtaining all declaration rule labels and customs code labels, and outputting the trained customs declaration model; Generate customs declaration documents corresponding to the goods based on the customs code and declaration elements output by the customs declaration model, including the following steps: S301: Obtain the customs declaration agreement template; S302: Confirming the filling elements and customs code of the target commodity information in the customs declaration document based on the declaration elements and customs code; S303: Output the complete customs declaration data and fill it into the customs declaration agreement template to generate a customs declaration document.

2. The method for customs import and export declaration based on intelligent coding according to claim 1, characterized in that: Outputting the customs code and declaration elements corresponding to the target commodity based on the intelligent matching algorithm includes the following steps: S205: Calling the product information of the target product and generating a second representation model of the product information; S206: Call the trained customs declaration model, perform embedding processing on the second representation model, calculate the distance between the target commodity after embedding processing and each declaration rule label and each customs code label under the same category of the target commodity, and output the declaration elements and customs code corresponding to the target commodity based on the condition of minimum distance.

3. The method for customs import and export declaration based on intelligent coding according to claim 2, characterized in that: The data in the first characterization model and the second characterization model are related to each other. The data in the characterization model include the Chinese and English names of the product, the final ingredients of the product, the 1st to 5th level ingredients of the product, and classification elements.

4. The method for customs import and export declaration based on intelligent coding according to claim 1, characterized in that: The customs coding rules database includes classification rules corresponding to commodity codes. The categories of the classification rules include commodity composition, commodity appearance, commodity specifications and commodity models. The customs coding rules database can execute batch deactivation / activation of customs coding rules, import / export of customs coding rules and testing of customs coding rules.

5. The method for customs import and export declaration based on intelligent coding according to claim 1, characterized in that: The declaration element rule database includes declaration element specifications for commodity data, and the declaration element specifications include commodity data source specifications and commodity value logic specifications. The declaration element rule database can execute batch deactivation / activation of declaration element rules, copying / adding of declaration element rules, and testing of declaration element rules.

6. The method for customs import and export declaration based on intelligent coding according to claim 1, characterized in that: Encrypted storage and data signing of the customs declaration documents include: S304: Detecting sensitive data in the customs declaration document, and encrypting and storing the sensitive data using the AES algorithm; S305: After encryption and storage, the customs declaration document is signed by encryption hardware.

7. The method for customs import and export declaration based on intelligent coding according to claim 1, characterized in that: The transmission method for submitting the customs declaration documents through the message queue includes HTTP request and RabbitMQ communication. The RabbitMQ communication temporarily stores the customs declaration documents in the RabbitMQ server, captures the customs declaration documents through the customs declaration system, and then returns the corresponding receipt file.

8. A customs import and export declaration system based on intelligent coding, characterized by: The system comprises: A rule configuration module is used to obtain customs documents and extract customs declaration information, and configure declaration rules based on the customs declaration information. The declaration rules include customs coding rules and declaration element rules; A rule matching module is used to obtain commodity data, build a customs declaration model based on the commodity data and the declaration rules, train the customs declaration model based on incremental commodity data and incremental declaration rules using an NLP algorithm, and output the customs code and declaration elements corresponding to the target commodity based on an intelligent matching algorithm; A document preparation module is used to generate customs declaration documents corresponding to the target commodity based on the customs code and declaration elements output by the customs declaration model, and to encrypt, store and sign the customs declaration documents. The customs declaration documents include export orders, payment orders, logistics orders and declaration lists; The document declaration module is used to access the customs declaration service system, submit the customs declaration documents through the message queue, and verify the receipt documents returned by the customs declaration documents; The rule configuration module includes: The rule database module is used to obtain commodity coding documents and customs declaration agreement templates from customs documents; parse the commodity coding documents, extract customs coding rules, import them into the database, and build a customs coding rules database; parse the customs declaration agreement template, extract declaration element rules, import them into the data, and build a declaration element rules database; regularly detect customs document updates, and if new coding rules or new declaration element rules are parsed, import the incremental coding rules or incremental declaration elements into the customs coding rules database or the element declaration rules database, and regularly perform database updates; The rule matching module includes: A machine learning module is configured to read historical declaration rules and historical commodity data; read incremental declaration rule data and incremental commodity data; train the customs declaration model to obtain a first representation model that converts the commodity text representation into a mathematical vector, where the commodity text representation is commodity data that satisfies the declaration rules; train the first representation model based on a FastText model, embed each word in the first representation model and then average it to obtain a document vector; predict the document vector using a neural network to obtain all declaration rule labels and customs code labels; and output the trained customs declaration model; The document preparation module includes: The customs declaration document generation module is used to obtain a customs declaration agreement template; confirm the filling elements and customs codes of the target commodity information in the customs declaration document based on the declaration elements and customs codes; output complete customs declaration data and fill it into the customs declaration agreement template to generate a customs declaration document.

9. The customs import and export declaration system based on intelligent coding according to claim 8 is characterized in that: The rule matching module also includes: The intelligent matching module is configured to call the commodity information of the target commodity and generate a second representation model of the commodity information; call the trained customs declaration model, perform embedding processing on the second representation model, calculate the distance between the target commodity after embedding processing and each declaration rule label and each customs code label under the same category of the target commodity, and output the declaration elements and customs code corresponding to the target commodity based on the condition of minimum distance.

10. The customs import and export declaration system based on intelligent coding according to claim 8 is characterized in that: The document preparation module also includes: The security encryption module is used to detect sensitive data in the customs declaration document and encrypt and store the sensitive data using the AES algorithm; after encryption and storage, the customs declaration document is signed by the encryption machine hardware.

11. A customs import and export declaration device based on intelligent coding, characterized in that: include: at least one processor, and a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the customs import and export declaration method based on intelligent coding as described in any one of claims 1 to 7.

12. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the customs import and export declaration method based on intelligent coding as described in any one of claims 1 to 7.

Images