Method for classifying goods and electronic device

Abstract

This application discloses a commodity classification processing method and electronic device. The method includes: receiving a request for customs classification of a target commodity, the request carrying the URL information of a commodity details page associated with the target commodity and destination country information; obtaining detailed description information associated with the target commodity based on the URL information, and processing the detailed description information using a large-scale parameter model of artificial intelligence (AI) to generate commodity elements; initiating a customs code query request to an associated export customs declaration service system based on the commodity name, so as to determine the customs code corresponding to the target commodity in the exporting country based on the information returned by the customs declaration service system; determining candidate customs codes associated with the internationally common part of the customs code corresponding to the target commodity in the exporting country based on the internationally common part of the customs code, and matching the key-value pairs in the commodity elements with the key-value pairs in the classification elements corresponding to the candidate customs codes to determine the complete customs code corresponding to the target commodity in the destination country. This application embodiment can improve the efficiency and accuracy of customs classification.

Description

Technical Field

[0001] This application relates to the field of commodity information processing technology, and in particular to commodity classification processing methods and electronic devices. Background Technology

[0002] In cross-border e-commerce platforms and other product information service systems, buyers and sellers come from different countries. After a transaction is completed, cross-border transportation of goods is involved. This process involves costs beyond just the goods themselves and shipping, often including customs duties, which may be borne by the buyer. The specific customs rates and taxes are usually determined by the destination country's customs system upon clearance, based on the goods' customs classification code and applicable tax rules. The buyer then makes the payment. Because cross-border transportation can be lengthy, the time from initial transaction completion to customs clearance and receipt of goods can be considerable. While the buyer may have paid upon completion, additional costs may arise while waiting for delivery. These extra costs can negatively impact the user experience, and the uncertainty can influence purchasing decisions. Furthermore, new users may experience disputes with sellers due to these additional costs.

[0003] To address the above issues, some cross-border e-commerce platforms can disclose information such as potential tariff rates and taxes to buyers in advance on the product information page. This allows buyers to obtain more definitive information about tariffs before or even when placing an order, helping them make informed decisions.

[0004] However, since different goods correspond to different customs codes (HS codes) in the customs system, and the corresponding tax rates are also different, and the definitions of customs codes in the customs tariffs of different countries are also different, in order to reveal tariff information to users in advance, it is necessary to classify and tax specific goods in the cross-border e-commerce system. That is, it is necessary to determine the customs code of specific goods in the destination country before it is possible to query specific tariff rates and other information.

[0005] Of course, besides providing buyers with specific customs and tariff information, customs classification of goods is also crucial for sellers (i.e., exporters). Specifically, the customs code for a product primarily affects the seller's tax refunds, export licenses, and inspection and quarantine. Therefore, customs classification of goods on cross-border e-commerce platforms is essential for improving user experience.

[0006] To determine the corresponding customs codes for specific goods in the destination country from e-commerce platforms, current solutions typically rely on classification experts learning and understanding massive amounts of customs tariff data. Furthermore, they need to learn and understand the complex product category classification systems of cross-border e-commerce platforms before establishing a mapping relationship between the e-commerce platform's product categories and the HS codes defined in the customs tariff. However, since different e-commerce platforms have their own distinct product category systems, each platform needs to establish its own mapping relationship with the HS codes of multiple destination countries, resulting in a very large workload. In addition, due to the individual differences in the cultural background, language skills, and business proficiency of classification experts, the accuracy of classification varies significantly when dealing with goods from different exporting countries, and there is currently no universal solution for a quick resolution. Summary of the Invention

[0007] This application provides a commodity classification processing method and electronic equipment that can improve the efficiency and accuracy of customs classification.

[0008] This application provides the following solution:

[0009] A commodity classification processing method, comprising:

[0010] Receive a request for customs classification of a target product, the request carrying the URL of the product details page associated with the target product, and the destination country information;

[0011] The detailed description information associated with the target product is obtained based on the URL information, and the detailed description information is processed by an artificial intelligence (AI) large-scale parameter model to generate product elements. The product elements include the product name and key-value pairs consisting of multiple attributes of the product and their corresponding attribute values.

[0012] Based on the product name, a customs code query request is initiated to the associated export customs declaration service system. The export customs declaration service system is used to provide export customs declaration services for exporting users in the same exporting country. Its database records historical data generated during the provision of export customs declaration services. The historical data includes the product names of multiple products and the corresponding customs codes of the exporting country, so as to determine the customs code corresponding to the target product in the exporting country based on the information returned by the customs declaration service system.

[0013] Based on the internationally recognized portion of the customs code corresponding to the target commodity in the exporting country, candidate customs codes associated with the internationally recognized portion are determined for the destination country. The key-value pairs in the commodity elements are matched with the key-value pairs in the classification elements corresponding to the candidate customs codes to determine the complete customs code corresponding to the target commodity in the destination country. The classification elements are generated in advance by an AI large-scale parameter model through understanding the customs taxation rules of the destination country, and the key-value pairs include commodity attributes and attribute values.

[0014] The process of initiating a customs code query request to the associated export customs declaration service system based on the commodity name also includes:

[0015] The product names are normalized to generate standardized product names, so that a customs code query request can be initiated to the export customs declaration service system based on the standardized product names.

[0016] If the export customs declaration service system returns multiple customs codes for the target commodity in the exporting country, the customs taxation rules corresponding to these multiple customs codes in the exporting country and the commodity elements of the target commodity are input into the AI ​​large-scale parameter model so that the AI ​​large-scale parameter model can select one of the customs codes as the customs code corresponding to the target commodity in the exporting country.

[0017] Specifically, when matching the key-value pairs in the commodity elements of the target commodity with the key-value pairs in the classification elements corresponding to multiple customs codes of the destination country, the internationally common part of the customs code corresponding to the target commodity in the exporting country, the key-value pairs in the commodity elements, and the key-value pairs in the classification elements corresponding to the multiple customs codes related to the internationally common part of the destination country are input into the AI ​​large-scale parameter model for matching and judgment.

[0018] This also includes:

[0019] Based on the complete customs code corresponding to the target product in the destination country and the customs taxation rules of the destination country, the tax rate and / or tax information of the target product when sold in the destination country is determined for display to buyer users in the cross-border commodity information service system.

[0020] The step of receiving a request for customs classification of the target goods includes:

[0021] Receive requests from sellers in the cross-border commodity information service system to classify new commodities for customs purposes after the new commodities have been published.

[0022] The step of receiving a request for customs classification of the target goods includes:

[0023] Receives customs classification requests for multiple existing commodities initiated by scheduled tasks set in the cross-border commodity information service system.

[0024] A commodity classification processing method, comprising:

[0025] Receive a request for customs classification of a target product, the request carrying the URL of the product details page associated with the target product, and the destination country information;

[0026] The detailed description information associated with the target product is obtained based on the URL information, and the detailed description information is processed by an artificial intelligence (AI) large-scale parameter model to generate product elements. The product elements include the product name and key-value pairs consisting of multiple attributes of the product and their corresponding attribute values.

[0027] The key-value pairs in the commodity elements are input into the key-value pairs in the classification elements corresponding to multiple customs codes associated with the destination country, and then matched and judged by an AI large-scale parameter model to determine the complete customs code corresponding to the target commodity in the destination country. The classification elements are generated in advance by the AI ​​large-scale parameter model through understanding the customs taxation rules of the destination country, and the key-value pairs of the classification elements include commodity attributes and attribute values.

[0028] A computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the steps of any of the preceding methods.

[0029] An electronic device, comprising:

[0030] One or more processors; and

[0031] A memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform the steps of the method described in any of the preceding descriptions.

[0032] A computer program product includes a computer program / computer executable instructions that, when executed by a processor in an electronic device, implement the steps of any of the preceding methods.

[0033] According to the specific embodiments provided in this application, the following technical effects are disclosed:

[0034] Through the embodiments of this application, customs classification can be performed at the commodity level. Specifically, the detailed description information of the target commodity can be processed by an AI model to generate commodity elements, which may include key-value pairs consisting of commodity name, attributes, and attribute values. Then, a query can be initiated to the export customs declaration service system based on the commodity name. With the help of the data already accumulated in the associated export customs declaration service system, a "coarse classification" can be performed to determine the customs code corresponding to the specific target commodity in the customs tariff of a certain country. Based on the internationally recognized part of the customs code, multiple candidate customs codes related to the internationally recognized part in the destination country can be determined. Then, the commodity elements can be matched and judged with the classification elements corresponding to these candidate customs codes to determine the complete customs code corresponding to the target commodity in the destination country, thereby completing the "refined classification". This approach leverages data accumulated within the export customs declaration service system and the multimodal data processing and understanding capabilities of AI models to automate customs classification of goods, eliminating reliance on expert knowledge or experience. The two-step process of "coarse classification" and "refined classification" improves efficiency, avoiding the need for tedious comparisons between commodity elements and numerous customs code classification elements. Furthermore, the high accuracy of the data accumulated in the export customs declaration service system contributes to the accuracy of the classification results. Additionally, because classification is based on the goods' own information, rather than category information from a commodity information service system, this solution offers greater versatility, providing a unified customs classification service for different commodity information service systems.

[0035] Of course, any product implementing this application does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a schematic diagram of the system architecture provided in the embodiments of this application;

[0038] Figure 2 This is a flowchart of the first method provided in the embodiments of this application;

[0039] Figure 3 This is a flowchart of the second method provided in the embodiments of this application;

[0040] Figure 4This is a schematic diagram of the electronic device provided in the embodiments of this application. Detailed Implementation

[0041] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0042] In this embodiment, a solution is provided to offer a more universal classification and taxation scheme while improving its accuracy. This scheme no longer matches the category system of e-commerce platforms with the customs codes in the customs tariffs of various destination countries; instead, it performs classification and taxation at the commodity level. Although different e-commerce platforms differ in their category system construction, they all contain descriptions of specific commodities. Therefore, commodity-based classification and taxation services are more universal and can provide classification and taxation services for commodities in multiple different commodity information service systems. Furthermore, during the classification and taxation process, the multimodal information processing capabilities of large-scale AI (artificial intelligence) parameter models (hereinafter referred to as AI large models) can be utilized to achieve more accurate and efficient classification and taxation. In this context, while customs tariffs differ between countries, resulting in different customs codes for the same commodity, these codes typically include internationally recognized components (e.g., the first four or six digits are internationally common, meaning the customs codes for the same commodity are identical across countries, with subsequent digits defined individually by each country). Therefore, during the classification and taxation process, a mapping from one country's customs code to others can be implemented, improving efficiency and accuracy. To achieve this, a "rough classification" can be performed using existing data from export customs declaration systems to determine the internationally recognized portion of the commodity's customs code in the destination country. Then, an AI model, building upon the "rough classification," performs a "refined classification" based on the commodity's characteristics and the classification elements generated from the destination country's customs tariff, determining the complete customs code for the specific commodity in the destination country.

[0043] The aforementioned export customs declaration service system can be another service system related to the specific commodity information service system. This export customs declaration service system mainly provides export customs declaration services to exporting users, typically for users within the same exporting country. That is, if a merchant needs to export a certain commodity, they can obtain specific export customs declaration services through this system. Since using this service usually requires the exporter to provide detailed commodity information, including the commodity name, various commodity attribute descriptions, etc., the system will determine the corresponding customs code for the specific commodity in the exporting country and then submit the declaration to the customs department of that exporting country. To ensure successful customs clearance, the customs code assigned to the goods in the system must be accurate. Therefore, the system typically guides users to fill in information such as the product name and attributes. Regarding product attributes, the system can recommend some attributes for users to select, or users can add less common attributes, including the product's purpose, size, specifications, etc. The export customs clearance service system can integrate this information and classify the goods based on rules configured by classification experts in the exporting country, determining the corresponding customs code. It can also perform risk control scanning and manual review to ensure the accuracy of the classification results. During this process, the export customs clearance service system can accumulate a large amount of data, including product names, attributes, and corresponding customs codes in the specific exporting country, and this customs code information has sufficient accuracy. Based on these characteristics, this application embodiment can utilize the data accumulated in this export customs clearance service system to query the customs code corresponding to a specific product in the exporting country, and then map it to the customs codes of multiple destination countries.

[0044] In practical implementation, from a system architecture perspective, such as Figure 1As shown, this application embodiment can provide a related commodity classification and taxation service for a cross-border commodity information service system. This service can provide relevant service interfaces for the cross-border commodity information service system to call. Specifically, when a seller user publishes a new commodity, this interface can be called to initiate a classification and taxation request for the new commodity. Alternatively, for existing commodities in the cross-border commodity information service system, the service interface of the aforementioned commodity classification and taxation service can be called by setting scheduled tasks in the system background. This service can execute these scheduled tasks offline to complete the classification and taxation service for existing commodities. During the classification and taxation process, the specific request can carry the URL information of the details page of the specific commodity, as well as the required destination country information. This then involves calling the AI ​​model and the relevant service interfaces of the associated export customs declaration service system. Specifically, the AI ​​model (here referred to as the "commodity element extraction model") can first process the commodity details information to generate commodity elements, which may include the commodity name, multiple attributes of the commodity, and multiple key-value pairs composed of attribute values, etc. Of course, as an optional method, during the generation of commodity elements, commodity names, attributes, etc., can be normalized to achieve standardized expression of commodity names, attributes, and other information. Then, the relevant service interface of the export customs declaration service system can be called to query the customs code of the specific commodity name associated with the exporting country (e.g., country A) in that export customs declaration service system. If the export customs declaration service system returns multiple customs codes, these customs codes can be used as candidates, and the commodity elements, the tax calculation rules corresponding to each customs code, and other information can be input into the AI ​​model (which can be called a "category recognition model"). The AI ​​model will then select the most matching customs code. After determining the customs code corresponding to the current commodity in the aforementioned exporting country (country A), the internationally recognized portion (e.g., the first 4 or 6 digits) can be extracted. Then, it is matched with the key-value pairs in the commodity elements and the key-value pairs in the classification elements corresponding to the customs code related to the internationally recognized portion, generated according to the customs tax calculation rules of the destination country (e.g., country B), to determine the complete customs code corresponding to the target commodity in the destination country.

[0045] For example, assuming the target product's customs code in the aforementioned exporting country is code 'a', after extracting the first four or six digits, three customs codes related to this internationally accepted part of the target country can be identified: codes b, c, and d. Then, the classification elements corresponding to codes b, c, and d can be determined. These classification elements can be pre-generated by an AI model based on the customs tariff of the aforementioned target country. Classification elements refer to the attributes a product must possess and their values ​​to be classified under a specific customs code. Then, the AI ​​model matches the product elements of the target product—that is, the actual attributes and attribute values ​​of the product—with the classification elements of each of the aforementioned customs codes b, c, and d. The customs code with the highest matching degree determines the target product's classification.

[0046] In this way, the target goods are "coarsely classified" based on data previously accumulated in the export customs declaration service system of a specific exporting country. Then, the AI ​​model, building on this coarse classification, performs "refined classification" according to the classification elements corresponding to the specific customs codes of the destination country. This makes classification via the AI ​​model possible and improves efficiency (without the coarse classification, the AI ​​model would need to match and calculate the classification elements corresponding to all customs codes of the destination country one by one to select the matching customs code, a very large workload even for an AI model and requiring significant time). After the "refined classification," the goods can be taxed according to the tariffs of the destination country, predicting the specific tax rate and / or fees. Finally, the classification and tax calculation results can be returned to the user or a scheduled task.

[0047] The specific implementation schemes provided in the embodiments of this application will be described in detail below.

[0048] Example 1

[0049] First, Embodiment 1 of this application provides a product classification processing method, see [link to embodiment]. Figure 2 The method may include:

[0050] S201: Receive a request for customs classification of the target product, the request carrying the URL information of the product details page associated with the target product, and the destination country information.

[0051] Specifically, the request for customs classification of target goods can be initiated by the seller user in the cross-border commodity information service system, or by the back-end staff of the cross-border commodity information service system through methods such as setting up scheduled tasks. The former is mainly used for customs classification of newly released goods, while the latter is mainly used for customs classification of existing goods in the system (goods that have not been classified before the service provided in this application embodiment).

[0052] Specifically, for the former, it could be a request initiated by a seller in a cross-border commodity information service system to classify a new commodity for customs after the seller has published it. That is, during the process of publishing a new commodity, the seller needs to provide information such as the commodity's name and attributes. Furthermore, since it's a cross-border scenario, specific information such as the destination country can also be specified. After configuring the above information, the commodity can be published in the system, which can then assign a unique identifier such as a commodity ID to the new commodity and generate a details page link so that other users can access the commodity's details. After completing the above process, the seller's interface can provide an option to initiate customs classification for the new commodity. Then, the service interface provided in this application embodiment can be called, and the call request can carry the URL of the new commodity's details page and the destination country information.

[0053] For the latter scenario, which involves customs classification of existing goods in the system, given the large number of goods, technical personnel can set up scheduled tasks to initiate customs classification requests for multiple existing goods. These scheduled tasks can be configured with the URL of each product's details page and its country of destination.

[0054] S202: Obtain the detailed description information associated with the target product based on the URL information, and process the detailed description information through an artificial intelligence (AI) large-scale parameter model to generate product elements. The product elements include the product name and key-value pairs consisting of multiple attributes of the product and their corresponding attribute values.

[0055] For specific classification services, upon receiving a specific request, the detailed description information associated with the target product can be obtained based on the URL information. Then, the detailed description information can be processed using an AI model to generate product elements. These product elements can include the product name and key-value pairs consisting of multiple product attributes and their corresponding values. The product's detailed description information can include data of various modalities, such as text data, image / video data, etc. In this embodiment, the multimodal data processing capabilities of the AI ​​model can be leveraged to extract and generate product elements from this multimodal data. This approach allows for the effective utilization and mining of information contained in images / videos, improving the accuracy of the classification results.

[0056] In practice, during the extraction of product elements based on the product's detailed description information, the extracted product names and attributes can be normalized to achieve standardized expressions of specific product names and attribute values, thereby improving the matching rate and accuracy of subsequent steps. For example, suppose the AI ​​model generates a product name of "blue dress" based on the product details information, but it can be normalized to the standardized product name "blue dress," and so on. The above normalization process can also be performed by querying the aforementioned export customs declaration service system.

[0057] S203: Initiate a customs code query request to the associated export customs declaration service system based on the product name. The export customs declaration service system is used to provide export customs declaration services for exporting users in the same exporting country. Its database records historical data generated during the provision of export customs declaration services. The historical data includes the product names of multiple products and the corresponding customs codes of the exporting country, so as to determine the customs code corresponding to the target product in the exporting country based on the information returned by the customs declaration service system.

[0058] After extracting the commodity elements, a customs code query request can be initiated to the associated export customs declaration service system based on the commodity name. As mentioned earlier, this export customs declaration service system provides export customs declaration services to exporters in the same exporting country (e.g., China). Its database records historical data generated during the provision of export customs declaration services. This historical data includes the commodity names of multiple commodities and their corresponding customs codes for the exporting country. In other words, a large amount of historical data is accumulated in the export customs declaration service system, including commodity names and the corresponding customs codes after classifying the commodity names according to the aforementioned customs tariffs of the exporting country. Furthermore, although the customs codes are generated by the export customs declaration service system, they must be accurate because they need to be declared to the customs department; otherwise, the declaration cannot be successfully submitted to the customs department. Therefore, the accuracy of the customs codes corresponding to specific commodity names in the historical data of the export customs declaration service system is very high. This application embodiment can utilize this historical data to perform a "rough classification" of commodities. Specifically, a query can be initiated to the export customs declaration service system using the commodity name as the input parameter, and the export customs declaration service system can return the matching customs codes accordingly.

[0059] If the export customs declaration service system returns multiple customs codes, the customs duty calculation rules corresponding to these multiple customs codes in the exporting country, as well as the commodity elements of the target product, can be input into the AI ​​model. The AI ​​model will then "select" one of the customs codes as the corresponding customs code for the target product in the exporting country. The AI ​​model can understand and match the customs duty calculation rules corresponding to the multiple customs codes and the commodity elements of the target product, selecting the most matching customs code as the current customs code for the target product.

[0060] S204: Based on the internationally recognized portion of the customs code corresponding to the target commodity in the exporting country, determine the candidate customs codes associated with the internationally recognized portion in the destination country, and match the key-value pairs in the commodity elements with the key-value pairs in the classification elements corresponding to the candidate customs codes to determine the complete customs code corresponding to the target commodity in the destination country; wherein, the classification elements are generated in advance by an AI large-scale parameter model through understanding the customs taxation rules of the destination country, and the key-value pairs include commodity attributes and attribute values.

[0061] The customs code determined by querying the export customs declaration service system is the customs code corresponding to the target commodity after classifying it according to the customs tariff of the exporting country. However, what is actually needed is the customs code corresponding to the target commodity after classifying it according to the customs tariff of the destination country. Therefore, it is also possible to achieve the mapping from the customs code of the exporting country to the customs code of the destination country.

[0062] Since different countries' customs codes contain internationally recognized portions, we can first determine the relevant customs codes (usually multiple) in the destination country's customs tariff based on the internationally recognized portion of the target commodity's customs code in the exporting country's customs code. Additionally, an AI model can pre-interpret the destination country's customs tariff to generate classification elements corresponding to each customs code. That is, to classify a commodity to a specific customs code, the commodity needs to possess certain attributes and attribute values. Therefore, classification elements are key-value pairs composed of attributes and attribute values. Each customs code can have multiple key-value pairs, and classification to that customs code is only completed when all key-value pairs match successfully. Having pre-generated the classification elements for each customs code, we can then match the key-value pairs in the commodity elements with the key-value pairs in the classification elements of the relevant internationally recognized customs codes in the destination country, thus completing the "precise classification" process. Of course, since the expression of attributes and attribute values ​​by product elements and classification elements may not be completely consistent, that is, even if it is the same pair of attributes and attribute values, the expression in product elements may be inconsistent with the expression in classification elements. It is difficult to directly compare whether the two match literally. Therefore, the above-mentioned "refinement" process can also be completed by AI big data models. By combining the cases of multiple key-value pairs, AI big data models can provide more accurate matching results.

[0063] After completing the "refined return" process, since buyers are more concerned with the specific taxes and fees involved, the complete customs code for the target product in the destination country can be determined. Based on this code and the destination country's customs tax rules, the tax rate and / or tax information for the target product when sold in the destination country can then be calculated. In other words, the returned information may include the aforementioned customs code, as well as tax rates and / or tax information. The specific customs code is primarily provided to the seller for subsequent tax refunds; the tax rate and / or tax information is primarily provided to the buyer. For example, this information can be displayed on the product details page, allowing buyers to obtain more definitive tax information in advance to aid their purchasing decisions, reduce transaction disputes, and further improve transaction certainty and service quality.

[0064] In summary, through the embodiments of this application, customs classification can be performed at the commodity level. Specifically, the detailed description information of the target commodity can be processed by an AI large model to generate commodity elements, which may include commodity name and key-value pairs consisting of attributes and attribute values. Then, a query can be initiated to the export customs declaration service system based on the commodity name. With the help of the data already accumulated in the associated export customs declaration service system, a "coarse classification" can be performed to determine the customs code corresponding to the specific target commodity in the customs tariff of a certain country. Based on the internationally recognized part of the customs code, multiple candidate customs codes related to the internationally recognized part in the destination country can be determined. Then, the commodity elements can be matched and judged with the classification elements corresponding to these candidate customs codes to determine the complete customs code corresponding to the target commodity in the destination country, thereby completing the "refined classification". This approach leverages data accumulated within the export customs declaration service system and the multimodal data processing and understanding capabilities of AI models to automate customs classification of goods, eliminating reliance on expert knowledge or experience. The two-step process of "coarse classification" and "refined classification" improves efficiency, avoiding the need for tedious comparisons between commodity elements and numerous customs code classification elements. Furthermore, the high accuracy of the data accumulated in the export customs declaration service system contributes to the accuracy of the classification results. Additionally, because classification is based on the goods' own information, rather than category information from a commodity information service system, this solution offers greater versatility, providing a unified customs classification service for different commodity information service systems.

[0065] Example 2

[0066] In this second embodiment, customs classification based on commodity elements can be directly achieved using a large AI model, eliminating the need for the two steps of "coarse classification" and "refined classification." Specifically, this second embodiment provides a commodity classification processing method, see [link to relevant documentation]. Figure 3 The method may specifically include:

[0067] S301: Receive a request for customs classification of the target product, the request carrying the URL information of the product details page associated with the target product, and the destination country information.

[0068] S302: Obtain the detailed description information associated with the target product based on the URL information, and process the detailed description information through an artificial intelligence (AI) large-scale parameter model to generate product elements. The product elements include the product name and key-value pairs consisting of multiple attributes of the product and their corresponding attribute values.

[0069] S303: Input the key-value pairs in the commodity element and the key-value pairs in the classification element corresponding to multiple customs codes associated with the destination country into the AI ​​large-scale parameter model for matching and judgment, so as to determine the complete customs code corresponding to the target commodity in the destination country; wherein, the classification element is generated in advance by the AI ​​large-scale parameter model through understanding the customs taxation rules of the destination country, and the key-value pairs of the classification element include commodity attributes and attribute values.

[0070] For the parts of this embodiment that are not described in detail, please refer to the description in embodiment one and other parts of this specification, which will not be repeated here.

[0071] It should be noted that the embodiments of this application may involve the use of user data. In practical applications, user-specific personal data may be used in the scheme described herein within the scope permitted by applicable laws and regulations, provided that it complies with the applicable laws and regulations of the country (e.g., with the user's explicit consent, with the user being properly notified, etc.).

[0072] Corresponding to the aforementioned Embodiment 1, this application also provides a commodity classification processing device, which may include:

[0073] The request receiving unit is used to receive a request for customs classification of a target product, the request carrying the URL information of the product details page associated with the target product, and the destination country information;

[0074] The product element generation unit is used to obtain the detailed description information associated with the target product based on the URL information, and process the detailed description information through an artificial intelligence (AI) large-scale parameter model to generate product elements. The product elements include the product name and key-value pairs consisting of multiple attributes of the product and their corresponding attribute values.

[0075] The query request initiation unit is used to initiate a customs code query request to the associated export customs declaration service system based on the product name. The export customs declaration service system is used to provide export customs declaration services for exporter users in the same exporting country. Its database records historical data generated during the provision of export customs declaration services. The historical data includes the product names of multiple products and the corresponding customs codes of the exporting country, so as to determine the customs code corresponding to the target product in the exporting country based on the information returned by the customs declaration service system.

[0076] The matching and judgment unit is used to determine the candidate customs codes associated with the internationally common part of the customs code corresponding to the target commodity in the exporting country, and to match and judge the key-value pairs in the commodity element with the key-value pairs in the classification element corresponding to the candidate customs code, so as to determine the complete customs code corresponding to the target commodity in the destination country; wherein, the classification element is generated in advance by an AI large-scale parameter model through understanding the customs taxation rules of the destination country, and the key-value pairs include commodity attributes and attribute values.

[0077] In a specific implementation, the device may further include:

[0078] The normalization processing unit is used to normalize the commodity name before initiating a customs code query request to the associated export customs declaration service system based on the commodity name, so as to generate a standardized commodity name, so as to initiate a customs code query request to the export customs declaration service system based on the standardized commodity name.

[0079] If the export customs declaration service system returns multiple customs codes for the target commodity in the exporting country, the customs taxation rules corresponding to these multiple customs codes in the exporting country and the commodity elements of the target commodity are input into the AI ​​large-scale parameter model so that the AI ​​large-scale parameter model can select one of the customs codes as the customs code corresponding to the target commodity in the exporting country.

[0080] Specifically, when matching the key-value pairs in the commodity elements of the target commodity with the key-value pairs in the classification elements corresponding to multiple customs codes of the destination country, the internationally common part of the customs code corresponding to the target commodity in the exporting country, the key-value pairs in the commodity elements, and the key-value pairs in the classification elements corresponding to the multiple customs codes related to the internationally common part of the destination country are input into the AI ​​large-scale parameter model for matching and judgment.

[0081] Additionally, the device may also include:

[0082] The tax rate and / or tax information providing unit is used to determine the tax rate and / or tax information of the target goods when they are sold to the destination country based on the complete customs code corresponding to the target goods in the destination country and the customs taxation rules of the destination country, so as to display it to the buyer users in the cross-border commodity information service system.

[0083] Specifically, the request receiving unit can be used for:

[0084] Receive requests from sellers in the cross-border commodity information service system to classify new commodities for customs purposes after the new commodities have been published.

[0085] Alternatively, the request receiving unit may specifically be used for:

[0086] Receives customs classification requests for multiple existing commodities initiated by scheduled tasks set in the cross-border commodity information service system.

[0087] Corresponding to Embodiment 2, this application also provides a commodity classification processing device, which may include:

[0088] The request receiving unit is used to receive a request for customs classification of a target product, the request carrying the URL information of the product details page associated with the target product, and the destination country information;

[0089] The product element generation unit is used to obtain the detailed description information associated with the target product based on the URL information, and process the detailed description information through an artificial intelligence (AI) large-scale parameter model to generate product elements. The product elements include the product name and key-value pairs consisting of multiple attributes of the product and their corresponding attribute values.

[0090] The matching and judgment unit is used to input the key-value pairs in the commodity elements and the key-value pairs in the classification elements corresponding to multiple customs codes associated with the destination country into the AI ​​large-scale parameter model for matching and judgment, so as to determine the complete customs code corresponding to the target commodity in the destination country; wherein, the classification elements are generated in advance by the AI ​​large-scale parameter model through understanding the customs taxation rules of the destination country, and the key-value pairs of the classification elements include commodity attributes and attribute values.

[0091] In addition, embodiments of this application also provide a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method described in any of the foregoing method embodiments.

[0092] And an electronic device, comprising:

[0093] One or more processors; and

[0094] A memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform the steps of the method described in any of the foregoing method embodiments.

[0095] A computer program product includes a computer program / computer executable instructions that, when executed by a processor in an electronic device, implement the steps of the method described in the foregoing method embodiments.

[0096] in, Figure 4 An exemplary architecture of an electronic device is shown, which may include a processor 410, a video display adapter 411, a disk drive 412, an input / output interface 413, a network interface 414, and a memory 420. The processor 410, video display adapter 411, disk drive 412, input / output interface 413, network interface 414, and memory 420 can communicate with each other via a communication bus 430.

[0097] The processor 410 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solution provided in this application.

[0098] The memory 420 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 420 can store the operating system 421 for controlling the operation of the electronic device 400, and the basic input / output system (BIOS) 422 for controlling the low-level operations of the electronic device 400. Additionally, it can store a web browser 423, a data storage management system 424, and a product classification processing system 425, etc. The aforementioned product classification processing system 425 can be the application program that specifically implements the aforementioned steps in this embodiment. In summary, when the technical solution provided in this application is implemented through software or firmware, the relevant program code is stored in the memory 420 and is called and executed by the processor 410.

[0099] Input / output interface 413 is used to connect input / output modules to realize information input and output. Input / output modules can be configured as components in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touch screens, microphones, various sensors, etc., and output devices may include displays, speakers, vibrators, indicator lights, etc.

[0100] Network interface 414 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0101] Bus 430 includes a pathway for transmitting information between various components of the device, such as processor 410, video display adapter 411, disk drive 412, input / output interface 413, network interface 414, and memory 420.

[0102] It should be noted that although the above-described device only shows the processor 410, video display adapter 411, disk drive 412, input / output interface 413, network interface 414, memory 420, bus 430, etc., in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the solution of this application, and does not necessarily include all the components shown in the figures.

[0103] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of this application.

[0104] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for system or system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and relevant parts can be referred to the descriptions in the method embodiments. The systems and system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0105] The commodity classification processing method and electronic device provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the method and its core ideas. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for classifying and processing commodities, characterized in that, include: Receive a request for customs classification of a target product, the request carrying the URL of the product details page associated with the target product, and the destination country information; The detailed description information associated with the target product is obtained based on the URL information, and the detailed description information is processed by an artificial intelligence (AI) large-scale parameter model to generate product elements. The product elements include the product name and key-value pairs consisting of multiple attributes of the product and their corresponding attribute values. Based on the product name, a customs code query request is initiated to the associated export customs declaration service system. The export customs declaration service system is used to provide export customs declaration services for exporting users in the same exporting country. Its database records historical data generated during the provision of export customs declaration services. The historical data includes the product names of multiple products and the corresponding customs codes of the exporting country, so as to determine the customs code corresponding to the target product in the exporting country based on the information returned by the customs declaration service system. Based on the internationally recognized portion of the customs code corresponding to the target commodity in the exporting country, candidate customs codes associated with the internationally recognized portion are determined for the destination country. The key-value pairs in the commodity elements are matched with the key-value pairs in the classification elements corresponding to the candidate customs codes to determine the complete customs code corresponding to the target commodity in the destination country. The classification elements are generated in advance by an AI large-scale parameter model through understanding the customs taxation rules of the destination country, and the key-value pairs include commodity attributes and attribute values.

2. The method according to claim 1, characterized in that, Before initiating a customs code query request to the associated export customs declaration service system based on the commodity name, the process also includes: The product names are normalized to generate standardized product names, so that a customs code query request can be initiated to the export customs declaration service system based on the standardized product names.

3. The method according to claim 1, characterized in that, If the export customs declaration service system returns multiple customs codes for the target commodity in the exporting country, then the customs taxation rules corresponding to these multiple customs codes in the exporting country and the commodity elements of the target commodity are input into the AI ​​large-scale parameter model, so that the AI ​​large-scale parameter model can select one of the customs codes as the customs code corresponding to the target commodity in the exporting country.

4. The method according to claim 1, characterized in that, When matching the key-value pairs in the commodity elements of the target commodity with the key-value pairs in the classification elements corresponding to multiple customs codes of the destination country, the internationally common part of the customs code corresponding to the target commodity in the exporting country, the key-value pairs in the commodity elements, and the key-value pairs in the classification elements corresponding to the multiple customs codes related to the internationally common part of the destination country are input into the AI ​​large-scale parameter model for matching and judgment.

5. The method according to claim 1, characterized in that, Also includes: Based on the complete customs code corresponding to the target product in the destination country and the customs taxation rules of the destination country, the tax rate and / or tax information of the target product when sold in the destination country is determined for display to buyer users in the cross-border commodity information service system.

6. The method according to any one of claims 1 to 5, characterized in that, Receiving a request for customs classification of the target goods includes: Receive requests from sellers in the cross-border commodity information service system to classify new commodities for customs purposes after the new commodities have been published.

7. The method according to any one of claims 1 to 5, characterized in that, Receiving a request for customs classification of the target goods includes: Receives customs classification requests for multiple existing commodities initiated by scheduled tasks set in the cross-border commodity information service system.

8. A method for classifying and processing commodities, characterized in that, include: Receive a request for customs classification of a target product, the request carrying the URL of the product details page associated with the target product, and the destination country information; The detailed description information associated with the target product is obtained based on the URL information, and the detailed description information is processed by an artificial intelligence (AI) large-scale parameter model to generate product elements. The product elements include the product name and key-value pairs consisting of multiple attributes of the product and their corresponding attribute values. The key-value pairs in the commodity elements are input into the key-value pairs in the classification elements corresponding to multiple customs codes associated with the destination country, and then matched and judged by an AI large-scale parameter model to determine the complete customs code corresponding to the target commodity in the destination country. The classification elements are generated in advance by the AI ​​large-scale parameter model through understanding the customs taxation rules of the destination country, and the key-value pairs of the classification elements include commodity attributes and attribute values.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method described in any one of claims 1 to 8.

10. An electronic device, characterized in that, include: One or more processors; as well as A memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform the steps of the method according to any one of claims 1 to 8.

11. A computer program product comprising a computer program / computer-executable instructions, characterized in that, When the computer program / computer-executable instructions are executed by a processor in an electronic device, they implement the steps of the method according to any one of claims 1 to 8.

Images