Express delivery receiving and sending standard identification method, device and equipment and storage medium

Abstract

The invention discloses an express delivery receiving and sending standard identification method, device and equipment and a storage medium, and the method comprises the steps: obtaining the name of a to-be-queried item, carrying out the feature extraction of the name of the to-be-queried item, obtaining an embedded vector of the name of the to-be-queried item, and carrying out the query from a preset delivery library of the to-be-queried item according to the embedded vector and the name of the to-be-queried item. The method comprises the following steps: obtaining a target posted object name and a customs code corresponding to the target posted object name, querying a policy file in a preset policy database by using the customs code to obtain a target policy file, sorting the target policy file to obtain a posted object receiving and sending standard represented by the to-be-queried posted object name, and sending the posted object to the to-be-queried posted object name to the to-be-queried posted object name. According to the method, express related policies can be inquired by using the name of the express delivery object, and the international express delivery receiving and sending standard can be quickly obtained.

Description

Technical Field

[0001] The present invention relates to the technical field of natural language processing, and in particular, to a method, device, equipment, and storage medium for identifying express delivery collection and delivery standards. Background Art

[0002] With the acceleration of globalization, more and more express delivery companies have begun to attempt to expand their business overseas. However, in the international market, they face the problem of being difficult to cope with the different and changing express delivery collection and delivery standards of various countries and regions. In the international market, the entry regulations and customs inspection procedures of different countries and regions vary. For example, in the United States, when it comes to items such as food and health products, an FDA form needs to be filled out. In Australia, all incoming packages need to be reviewed by the Australian customs, while in Europe, relevant regulations such as the EU CE certification need to be followed. If express delivery companies cannot accurately master these standards, it may lead to additional costs such as transportation failures and return shipping fees.

[0003] Currently, in order to maintain and update these standards, express delivery companies need to establish a team dedicated to collecting, sorting, interpreting, and updating the latest express delivery collection and delivery policies of various countries and regions around the world. However, relying on manual labor to maintain and update the standards easily results in a relatively low efficiency in identifying express delivery collection and delivery standards. Summary of the Invention

[0004] To solve the above technical problems, embodiments of the present invention provide an express delivery collection and delivery standard identification method and system, which realizes querying express delivery-related policies using the name of the express consignment, and can quickly obtain international express delivery collection and delivery standards.

[0005] A first aspect of embodiments of the present invention provides an express delivery collection and delivery standard identification method, and the method includes:

[0006] Obtain the name of the consignment to be queried;

[0007] Extract features from the name of the consignment to be queried to obtain an embedding vector of the name of the consignment to be queried. Query from a preset consignment delivery library according to the embedding vector and the name of the consignment to be queried to obtain a target consignment name and the customs code corresponding to the target consignment name;

[0008] Query a policy document in a preset policy database using the customs code to obtain a target policy document, and organize the target policy document to obtain the collection and delivery standards of the consignment represented by the name of the consignment to be queried.

[0009] In a possible implementation manner of the first aspect, querying from a preset consignment delivery library according to the embedding vector to obtain a target consignment name and the customs code corresponding to the target consignment name includes:

[0010] Query from a preset consignment delivery database according to the embedding vector and the name of the consignment to be queried, and obtain initial target consignment information, where the initial target consignment information includes the initial target consignment name and the customs code corresponding to the initial target consignment name;

[0011] Judge whether the attribute of the initial target consignment name is consistent with that of the consignment to be queried. If it is consistent, determine the initial target consignment name as the target consignment name and the customs code corresponding to the initial target consignment name as the customs code corresponding to the target consignment name.

[0012] In a possible implementation manner of the first aspect, query from a preset consignment delivery database according to the embedding vector and the name of the consignment to be queried, and obtain the initial target consignment information, including:

[0013] Calculate the coincidence degree between the name of the consignment to be queried and each consignment name in the preset consignment delivery database to obtain multiple coincidence degree values;

[0014] Determine the consignment name corresponding to the coincidence degree value reaching the preset value as the first target consignment name;

[0015] Judge whether the name of the first target consignment is the same as the name of the consignment to be queried. If they are equal, determine the information corresponding to the first target consignment name as the initial target consignment information; if they are not equal, calculate the similarity between the embedding vector of the consignment to be queried and the embedding vectors of each consignment name in the preset consignment delivery database to obtain multiple similarity values, and determine the consignment information corresponding to the similarity value greater than the preset similarity value as the initial target consignment information.

[0016] In a possible implementation manner of the first aspect, before obtaining the name of the consignment to be queried, the method further includes:

[0017] Construct a customs code interpretation database according to the import and export tariff policy text. The customs code interpretation database includes customs codes and the interpretations of customs codes;

[0018] Obtain the names and customs codes of historical consignments, query the customs code interpretations from the customs code interpretation database according to the customs codes of historical consignments, and obtain the customs code interpretations corresponding to the customs codes of historical consignments;

[0019] Judge whether the name of the historical consignment is consistent with the customs code interpretation corresponding to the customs code of the historical consignment. If it is consistent, extract the name of the historical consignment from the customs code interpretation;

[0020] Construct a preset consignment delivery database according to the names of historical consignments.

[0021] In a possible implementation of the first aspect, a preset consignment delivery library is constructed according to historical consignment names, including:

[0022] Determine the embedding vector of the historical consignment name;

[0023] A preset consignment delivery library is constructed according to the historical consignment name, the historical consignment customs code, and the embedding vector of the historical consignment name.

[0024] In a possible implementation of the first aspect, before obtaining the consignment name to be queried, the method further includes:

[0025] Extract the embedding vector of the goods name or goods description from the policy and regulation documents, and use the embedding vector of the goods name or goods description to query multiple consignment information to be judged from the preset consignment delivery library;

[0026] Perform attribute judgment on each consignment information to be judged and the goods name and goods description. If they are of the same attribute, store the information of the consignment information to be judged and the policy and regulation document corresponding to the goods name in the preset policy database. If they are not of the same attribute, delete the consignment information to be judged.

[0027] The second aspect of the embodiments of the present invention provides an express consignment and receipt standard recognition device, and the system includes:

[0028] An acquisition module for acquiring the consignment name to be queried;

[0029] A first query module for extracting features from the consignment name to be queried to obtain the embedding vector of the consignment name to be queried, and querying from the preset consignment delivery library according to the embedding vector and the consignment name to be queried to obtain the target consignment name and the customs code corresponding to the target consignment name;

[0030] A second query module for querying the policy document in the preset policy database using the customs code to obtain the target policy document, and sorting out the consignment and receipt standard of the consignment represented by the consignment name to be queried from the target policy document.

[0031] In a possible implementation of the second aspect, the first query module includes an extraction unit and a judgment unit,

[0032] Among them, the extraction unit is used to query from the preset consignment delivery library according to the embedding vector and the consignment name to be queried to obtain the initial target consignment information, where the initial target consignment information includes the initial target consignment name and the customs code corresponding to the initial target consignment name;

[0033] The determination unit is used to determine whether the initial target consignment name is consistent with the attributes of the consignment to be queried. If they are consistent, the initial target consignment name is determined as the target consignment name, and the customs code corresponding to the initial target consignment name is the customs code corresponding to the target consignment name.

[0034] The third aspect of the embodiments of the present invention provides a computer device, including:

[0035] A memory for storing a computer program;

[0036] A processor for implementing the express delivery receiving and sending standard recognition method as in the first aspect when executing the computer program.

[0037] The fourth aspect of the embodiments of the present invention provides a storage medium, on which a computer program is stored, and the steps of the express delivery receiving and sending standard recognition method as in the first aspect are implemented when the computer program is executed by a processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 : A flowchart of an embodiment of the express delivery receiving and sending standard recognition method provided by the present invention;

[0039] Figure 2 : A flowchart of the express delivery receiving and sending standard recognition process of an embodiment of the express delivery receiving and sending standard recognition method provided by the present invention;

[0040] Figure 3 : A schematic diagram of the system structure of another embodiment of the express delivery receiving and sending standard recognition method provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0041] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0042] Please refer to Figure 1 , a flowchart of an embodiment of the express delivery receiving and sending standard recognition method provided by the embodiments of the present invention, including steps S101 to S103, and the specific steps are as follows:

[0043] S101. Obtain the name of the consignment to be queried.

[0044] It should be noted that the name of the consignment to be queried is the name of the consignment that the user needs to query.

[0045] In this embodiment, in this step, an operation of inputting the name of the consignment to be queried is required. The user inputs the name of the consignment to be queried through the query interface, and the background receives the name of the consignment to be queried input by the user.

[0046] S102. Extract features from the name of the consignment to be queried to obtain the embedding vector of the name of the consignment to be queried. Query from the preset consignment delivery library according to the embedding vector and the name of the consignment to be queried to obtain the target consignment name and the customs code corresponding to the target consignment name.

[0047] In this embodiment, in this step, an operation of querying the name of the consignment to be queried is required. Use the large language model technology to extract the embedding of the consignment name, and query for similar consignment names from the consignment delivery library query interface, that is, HS-INST-API, based on Chinese character matching and semantic matching of the embedding. After querying for similar consignment names, continue to use the large language model technology to determine whether the attributes of the similar consignment name and the input consignment name are the same. If they are the same, determine that the customs code corresponding to the similar consignment name is the customs code corresponding to the consignment to be queried.

[0048] It should be noted that the large language model technology also refers to the LLM technology, which is a model based on machine learning and natural language processing technologies. Embedding refers to the embedding vector, which is a multi-dimensional real-valued vector, referring to the process of mapping high-dimensional data to a low-dimensional space.

[0049] S103. Query the policy document in the preset policy database using the customs code to obtain the target policy document, and organize the target policy document to obtain the collection and delivery standards of the consignment represented by the name of the consignment to be queried.

[0050] In this embodiment, in this step, query operations and summarization operations are required. As Figure 2 shown, after obtaining the customs code of the similar consignment, determine that the customs code of the similar consignment is the customs code of the consignment to be queried. Use this customs code to query the corresponding policy document from the preset policy data, and finally use the large language model technology to summarize the policy and output it, so as to obtain the collection and delivery standards of the consignment to be queried.

[0051] Implementing this embodiment, by obtaining the name of the consignment to be queried, extracting features from the name of the consignment to be queried, obtaining the embedding vector of the name of the consignment to be queried, querying from the preset consignment delivery library according to the embedding vector and the name of the consignment to be queried, obtaining the target consignment name and the customs code corresponding to the target consignment name, querying the policy document in the preset policy database using the customs code, obtaining the target policy document, and sorting out the target policy document to obtain the collection and delivery standards of the consignment represented by the name of the consignment to be queried. This method realizes querying express-related policies using the name of the express consignment and can quickly obtain the international express collection and delivery standards.

[0052] In some embodiments, it includes: querying from the preset consignment delivery library according to the embedding vector and the name of the consignment to be queried to obtain the initial target consignment information, where the initial target consignment information includes the initial target consignment name and the customs code corresponding to the initial target consignment name; determining whether the attributes of the initial target consignment name and the consignment to be queried are consistent. If they are consistent, then determine the initial target consignment name as the target consignment name and the customs code corresponding to the initial target consignment name as the customs code corresponding to the target consignment name.

[0053] In this embodiment, use the name of the consignment to be queried and the embedding vector to query for similar consignment names from the consignment delivery library. The data fields in the consignment delivery library include <consignment name, hscode, embedding>. There are two index fields in the database, one is the consignment name and the other is the embedding, that is, the embedding vector. The consignment name is used for character matching, and the embedding is used for vector matching.

[0054] After obtaining the similar consignment names, determine whether the attributes of each similar consignment name and the consignment to be queried are consistent. If they are consistent, then determine the customs code corresponding to the similar consignment name as the customs code of the consignment to be queried.

[0055] In some embodiments, it includes: calculating the coincidence degree between the name of the consignment to be queried and each consignment name in the preset consignment delivery library to obtain multiple coincidence degree values; determining the consignment name corresponding to the coincidence degree value reaching the preset value as the first target consignment name; determining whether the name of the first target consignment name is the same as the name of the consignment to be queried. If they are equal, then determine the information corresponding to the first target consignment name as the initial target consignment information; if they are not equal, then calculate the similarity between the embedding vector of the consignment to be queried and the embedding vectors of each consignment name in the preset consignment delivery library to obtain multiple similarity values, and determine the consignment information corresponding to the similarity value greater than the preset similarity value as the initial target consignment information.

[0056] In this embodiment, similar consignment names are queried from the consignment delivery library query interface, i.e., HS-INST-API, using Chinese character matching and embedding semantic matching. The specific process is as follows:

[0057] First, character matching is used. Character matching refers to the number of hit characters. The user inputs the consignment name to be queried, and this consignment name to be queried is traversed and compared with the consignment names in the consignment delivery library to calculate the overlap degree between the two and determine which group has a higher overlap degree. For example, if the consignment name A to be queried is "milk" and the similar consignment name B is "milk powder", then the overlap degree between A and B is 1 / 2, that is, out of the 2 characters in A, 1 is found in B. To judge the accuracy rate, the character matching threshold is set to 100%, that is, only when all the characters in A are found in B is B considered to meet the requirements.

[0058] However, when using character matching, a one-to-many situation may occur, that is, one consignment name to be queried gets multiple similar consignment names. For example, if the consignment name A is "milk" and "milk powder" and "milk" are queried, at this time, it is judged whether there is a situation where the consignment name A and the similar consignment B are exactly the same. At this time, "milk" appears in B and is exactly the same, so "milk takes precedence". If "milk powder" and "cow's milk" are queried, and at this time "milk" does not appear in B, in this case, embedding needs to be used for further judgment.

[0059] When performing embedding judgment, first use the large-scale language model technology to extract the embedding-A of the consignment to be queried, and then traverse the consignment delivery library, compare the cosine similarity with each embedding, and take out the consignment name with the highest similarity. For example, if the similarity threshold is set to 0.9, the consignment name with a similarity greater than 0.9 is selected as the target consignment name of the consignment name to be queried.

[0060] In some embodiments, it includes: before obtaining the consignment name to be queried, the method further includes: constructing a customs code interpretation database according to the import and export tariff policy text, where the customs code interpretation database includes customs codes and the interpretations of customs codes; obtaining the names and customs codes of historical consignments, querying the interpretations of customs codes from the customs code interpretation database according to the customs codes of historical consignments, and obtaining the interpretations of customs codes corresponding to the customs codes of historical consignments; judging whether the name of the historical consignment is consistent with the interpretation of the customs code corresponding to the customs code of the historical consignment. If they are consistent, extract the historical consignment name from the interpretation of the customs code; construct a preset consignment delivery library according to the historical consignment names.

[0061] In this embodiment, it should be noted that the preset consignment item delivery library refers to a database including information such as customs codes and consignment item names, and also refers to the consignment item delivery library mentioned above. The import and export tariff policy text refers to policy documents on import and export related information, including but not limited to laws, regulations, work reports, and planning documents.

[0062] The preset consignment item delivery library is constructed by using information such as policy and regulation documents and customs codes. The specific construction method is as follows: Obtain relevant national policy documents, and use the relevant national policy documents to construct an interpretation library of customs codes, with fields including <hscode, hscode interpretation>, and construct a query interface for the customs code interpretation library, namely HS-LIB-API. Obtain the historical consignment items and their customs codes of express delivery enterprises, use the query interface of the customs code interpretation library to query the corresponding interpretations of the customs codes, and use large language model technology to judge whether the consignment item name is consistent with its corresponding customs code interpretation. If they are consistent, use large language model technology to extract the embedding vector of the consignment item name, construct a consignment item delivery library based on customs codes <consignment item name, hscode, embedding>, and construct a query interface for the consignment item delivery library, namely HS-INST-API. It should be noted that hscode refers to the customs code, and embedding refers to the embedding vector.

[0063] In some embodiments, it includes: determining the embedding vector of the historical consignment item name; constructing the preset consignment item delivery library based on the historical consignment item name, historical consignment item customs code, and the embedding vector of the historical consignment item name.

[0064] In this embodiment, construct a consignment item delivery library based on customs codes according to the historical consignment item name, historical consignment item customs code, and the embedding vector of the historical consignment item name. The data in the consignment item delivery library includes <consignment item name, hscode, embedding>, and construct a query interface for the consignment item delivery library, namely HS-INST-API.

[0065] In some embodiments, it includes: extracting the embedding vector of the goods name or goods description from the policy and regulation documents, and using the embedding vector of the goods name or goods description to query multiple consignment items to be judged from the preset consignment item delivery library; performing attribute judgment on each consignment item to be judged with the goods name and goods description. If they are of the same attribute, store the information of the consignment item to be judged and the policy and regulation document corresponding to the goods name in the preset policy database. If they are not of the same attribute, delete the information of the consignment item to be judged.

[0066] In this embodiment, policy and regulation documents are obtained, and key information therein, such as country, goods, and import and export requirements, is extracted. Then, the large language model technology is used to extract the embedding of the goods name or goods description in the policy and regulation documents, and similar consignment names are queried from the consignment delivery library query interface using Chinese character matching and semantic matching based on embedding. It is determined whether the multiple similar consignment names retrieved are items of the same attribute as the "goods name / description". Since there may be a one-to-many relationship between the goods description and the consignment name, the similar consignment names of different attributes are judged and excluded in turn to obtain the similar consignment names of the same attribute. For example, "equidae and their related products" and "donkey meat" are of the same attribute, while "equidae and their related products" and "vest" are of different attributes. The similar consignment names and corresponding information such as customs codes, customs code interpretations, and policy information are stored in the policy database. The data in the policy database includes <customs code, code description, goods name, goods description, document title, document content, document link, document release time>, and a policy database query interface, i.e., POLICY-LIB-API, is constructed.

[0067] It should be noted that the policy database can be updated according to domestic and foreign legal policy documents to ensure the timeliness of the preset policy database.

[0068] This method effectively solves the pain point of difficult standard collection in international express delivery services and improves the efficiency and accuracy of standard collection in international express delivery services by using large language model technology as the connection hub for consignments, policy and regulations, and customs codes, and extracting and analyzing consignments to obtain effective policy documents.

[0069] In some embodiments, as Figure 3 shown, it shows a block diagram of an express delivery receiving and dispatching standard recognition device 300 provided by an embodiment of the present application, including: an acquisition module 301, a first query module 302, and a second query module 303, where:

[0070] The acquisition module 301 is used to acquire the consignment name to be queried;

[0071] The first query module 302 is used to extract the features of the consignment name to be queried to obtain the embedding vector of the consignment name to be queried, and query from the preset consignment delivery library according to the embedding vector and the consignment name to be queried to obtain the target consignment name and the customs code corresponding to the target consignment name;

[0072] The second query module 303 is used to query policy documents in the preset policy database using the customs code to obtain the target policy document, and organize the target policy document to obtain the receiving and dispatching standard of the consignment represented by the consignment name to be queried.

[0073] In some embodiments, the first query module 302 includes an extraction unit and a judgment unit.

[0074] Among them, the extraction unit is used to query from a preset consignment delivery library according to the embedding vector and the name of the consignment to be queried, and obtain initial target consignment information, where the initial target consignment information includes the initial target consignment name and the customs code corresponding to the initial target consignment name.

[0075] The judgment unit is used to judge whether the attribute of the initial target consignment name is consistent with that of the consignment to be queried. If they are consistent, the initial target consignment name is determined as the target consignment name, and the customs code corresponding to the initial target consignment name is determined as the customs code corresponding to the target consignment name. In some embodiments, querying from a preset consignment delivery library according to the embedding vector and the name of the consignment to be queried to obtain the initial target consignment information includes:

[0076] Calculating the coincidence degree between the name of the consignment to be queried and each consignment name in the preset consignment delivery library to obtain multiple coincidence degree values;

[0077] Determining the consignment name corresponding to the coincidence degree value reaching the preset value as the first target consignment name;

[0078] Judging whether the name of the first target consignment is the same as the name of the consignment to be queried. If they are equal, determining the information corresponding to the first target consignment name as the initial target consignment information; if they are not equal, calculating the similarity between the embedding vector of the consignment to be queried and the embedding vectors of each consignment name in the preset consignment delivery library to obtain multiple similarity values, and determining the consignment information corresponding to the similarity value greater than the preset similarity value as the initial target consignment information.

[0079] In some embodiments, before obtaining the name of the consignment to be queried, the method further includes:

[0080] Constructing a customs code interpretation database according to the import and export tariff policy text, where the customs code interpretation database includes customs codes and the interpretations of customs codes;

[0081] Obtaining the names and customs codes of historical consignments, querying the interpretations of customs codes from the customs code interpretation database according to the customs codes of historical consignments, and obtaining the interpretations of customs codes corresponding to the customs codes of historical consignments;

[0082] Judging whether the name of the historical consignment is consistent with the interpretation of the customs code corresponding to the customs code of the historical consignment. If they are consistent, extracting the name of the historical consignment from the interpretation of the customs code;

[0083] Constructing a preset consignment delivery library according to the names of historical consignments.

[0084] In some embodiments, a preset consignment delivery library is constructed according to historical consignment names, including:

[0085] Determine the embedding vector of the historical consignment name;

[0086] A preset consignment delivery library is constructed according to the historical consignment name, the historical consignment customs code, and the embedding vector of the historical consignment name.

[0087] In some embodiments, before obtaining the consignment name to be queried, the method further includes:

[0088] Extract the embedding vector of the goods name or goods description from the policy and regulation documents, and use the embedding vector of the goods name or goods description to query multiple consignment information to be judged from the preset consignment delivery library;

[0089] Perform attribute judgment on each consignment information to be judged and the goods name and goods description. If they are of the same attribute, store the information of the consignment information to be judged and the policy and regulation documents corresponding to the goods name in the preset policy database. If they are not of the same attribute, delete the consignment information to be judged.

[0090] In an embodiment of the present application, a computer device is provided. The computer device includes a memory and a processor. A computer program is stored in the memory. When the processor executes the computer program, the above steps are implemented. For the computer device provided in this embodiment, its implementation principle and technical effects are similar to those of the above method embodiment, and will not be elaborated here.

[0091] In an embodiment of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the above steps are implemented; for the computer-readable storage medium provided in this embodiment, its implementation principle and technical effects are similar to those of the above method embodiment, and will not be elaborated here.

[0092] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0093] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered that the scope described in this specification.

[0094] The specific embodiments described above further elaborate on the purpose, technical solutions, and beneficial effects of the present invention. It should be understood that the above are only specific embodiments of the present invention and are not used to limit the protection scope of the present invention. It is particularly pointed out that for those skilled in the art, any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for identifying express delivery collection and delivery standards, characterized in that, Including: Obtain the name of the consignment to be queried; Extract features from the name of the consignment to be queried to obtain the embedding vector of the name of the consignment to be queried. Query from the preset consignment delivery library according to the embedding vector and the name of the consignment to be queried to obtain the target consignment name and the customs code corresponding to the target consignment name; Use the customs code to query policy documents in the preset policy database to obtain the target policy document, and organize the target policy document to obtain the receiving and sending standards of the consignment represented by the name of the consignment to be queried.

2. The method for identifying express delivery collection and delivery standards according to claim 1, characterized in that, The querying from the preset consignment delivery library according to the embedding vector to obtain the target consignment name and the customs code corresponding to the target consignment name includes: Query from the preset consignment delivery library according to the embedding vector and the name of the consignment to be queried to obtain the initial target consignment information, where the initial target consignment information includes the initial target consignment name and the customs code corresponding to the initial target consignment name; Judge whether the attribute of the initial target consignment name is consistent with that of the consignment to be queried. If it is consistent, determine the initial target consignment name as the target consignment name and the customs code corresponding to the initial target consignment name as the customs code corresponding to the target consignment name.

3. The method for identifying express delivery collection and delivery standards according to claim 2, characterized in that, The querying from the preset consignment delivery library according to the embedding vector and the name of the consignment to be queried to obtain the initial target consignment information includes: Calculate the coincidence degree between the name of the consignment to be queried and each consignment name in the preset consignment delivery library to obtain multiple coincidence degree values; Determine the consignment name corresponding to the coincidence degree value reaching the preset value as the first target consignment name; Judge whether the name of the first target consignment name is the same as the name of the consignment to be queried. If they are equal, determine the information corresponding to the first target consignment name as the initial target consignment information; if they are not equal, calculate the similarity between the embedding vector of the name of the consignment to be queried and the embedding vectors of each consignment name in the preset consignment delivery library to obtain multiple similarity values, and determine the consignment information corresponding to the similarity value greater than the preset similarity value as the initial target consignment information.

4. The method for identifying express delivery collection and delivery standards according to claim 1, characterized in that, Before the obtaining the name of the consignment to be queried, the method further includes: Construct a customs code interpretation database according to the import and export tariff policy text, where the customs code interpretation database includes customs codes and the interpretations of customs codes; Obtain the names and customs codes of historical consignments, and query the interpretations of customs codes from the customs code interpretation database according to the customs codes of the historical consignments to obtain the interpretations of customs codes corresponding to the customs codes of the historical consignments; Judge whether the name of the historical consignment is consistent with the interpretation of the customs code corresponding to the customs code of the historical consignment. If it is consistent, extract the name of the historical consignment from the interpretation of the customs code; Construct a preset consignment delivery library according to the names of the historical consignments.

5. The method for identifying express delivery collection and delivery standards according to claim 4, characterized in that, The constructing a preset consignment delivery library according to the names of the historical consignments includes: Determine the embedding vector of the historical consignment name; Construct a preset consignment delivery library based on the historical consignment name, the historical consignment customs code, and the embedding vector of the historical consignment name.

6. The method for identifying express delivery collection and delivery standards according to claim 1, characterized in that, Before obtaining the consignment name to be queried, the method further includes: Extract the embedding vector of the goods name or goods description from the policy and regulation documents, and use the embedding vector of the goods name or goods description to query multiple consignment information to be judged from the preset consignment delivery library; Perform attribute judgment on each of the consignment information to be judged and the goods name and the goods description. If they are of the same attribute, store the consignment information to be judged and the information of the policy and regulation document corresponding to the goods name in the preset policy database. If they are not of the same attribute, delete the consignment information to be judged.

7. An apparatus for identifying express delivery collection and delivery standards, characterized in that, Includes: An acquisition module for acquiring the consignment name to be queried; A first query module for extracting features of the consignment name to be queried to obtain the embedding vector of the consignment name to be queried, and querying from the preset consignment delivery library according to the embedding vector and the consignment name to be queried to obtain the target consignment name and the customs code corresponding to the target consignment name; A second query module for querying a policy document in the preset policy database using the customs code to obtain a target policy document, and sorting out the target policy document to obtain the collection and delivery standards of the consignment represented by the consignment name to be queried.

8. The apparatus for identifying express delivery collection and delivery standards according to claim 7, characterized in that, The first query module includes an extraction unit and a judgment unit, wherein the extraction unit is used to query from the preset consignment delivery library according to the embedding vector and the consignment name to be queried to obtain initial target consignment information, where the initial target consignment information includes the initial target consignment name and the customs code corresponding to the initial target consignment name; The judgment unit is used to judge whether the attribute of the initial target consignment name is consistent with that of the consignment name to be queried. If they are consistent, determine the initial target consignment name as the target consignment name, and the customs code corresponding to the initial target consignment name as the customs code corresponding to the target consignment name.

9. A computer device, characterized in that, Includes: A memory for storing a computer program; A processor for implementing the express delivery collection and delivery standard recognition method according to any one of claims 1 to 6 when executing the computer program.

10. A storage medium, characterized in that, A computer program is stored on the storage medium, and when the computer program is executed by the processor, the steps of the express delivery collection and delivery standard recognition method according to any one of claims 1 to 6 are implemented.

Images