Cross-border commodity labeling method and device, electronic equipment and storage medium

Abstract

The invention relates to the field of label generation, in particular to a cross-border commodity labeling method and device, electronic equipment and a storage medium. The method comprises the steps that complete commodity information including basic information, production information and marketing information is systematically acquired, a multi-dimensional commodity data system is constructed, a data foundation is laid for accurate labeling, then labels are divided into basic labels, production labels, traceability labels and marketing labels, and the labels are classified into four types, namely, the basic labels, the production labels, the traceability labels and the marketing labels. Structured presentation of commodity information is achieved, the basic attribute display requirement is met, value-added information such as production traceability and marketing promotion is covered, meanwhile, differentiated query permissions are configured for different labeling plates, commercial sensitive information is effectively protected while information transparency is guaranteed, and the user experience is improved. The requirements of information sharing and data security are balanced, and the differentiated requirements of supervision mechanisms, consumers and supply chains are met.

Description

Technical Field

[0001] This application relates to the field of label generation, and in particular to a method, apparatus, electronic device and storage medium for labeling cross-border goods. Background Technology

[0002] Cross-border goods refer to products circulated between different countries or regions through cross-border e-commerce platforms, covering various types such as consumer goods and industrial products. In cross-border trade, product labeling and information management are the core hubs connecting the production, distribution, and consumption ends, and their standardization and digitalization directly affect trade efficiency, compliance, and consumer trust.

[0003] In existing technologies, product identification typically involves extracting information such as the product's name, specifications, etc., generating a label from this extracted information, and then affixing the label to the corresponding product. However, traditional labels usually only contain basic product information (such as name and specifications), making it difficult to meet the diverse needs of regulatory agencies, consumers, and various parties in the supply chain. Summary of the Invention

[0004] In order to enable cross-border product labeling to meet differentiated needs, this application provides a cross-border product labeling method, apparatus, electronic device and storage medium.

[0005] Firstly, this application provides a method for labeling cross-border goods, employing the following technical solution:

[0006] A method for labeling cross-border goods, comprising:

[0007] Obtain product information of the product to be bid on, the product information including basic information, production information and marketing information, the basic information being used to characterize the product attributes of the product to be bid on;

[0008] Generate product tags corresponding to the product information, including basic tags, production tags, traceability tags, and marketing tags;

[0009] Identify the labeling section corresponding to each product label and determine the query permissions for each labeling section;

[0010] Based on the labeling section and the corresponding query permissions, the labeling process for the trademark to be labeled is completed.

[0011] By adopting the above technical solutions, a multi-dimensional commodity data system was constructed by systematically acquiring complete commodity information, including basic, production, and marketing information. This laid the data foundation for accurate labeling. The labels were then divided into four categories: basic labels, production labels, traceability labels, and marketing labels, achieving a structured presentation of commodity information. This not only met the needs of displaying basic attributes but also covered value-added information such as production traceability and marketing promotion. Furthermore, by configuring differentiated query permissions for different labeling sections, commercially sensitive information was effectively protected while ensuring information transparency, balancing the needs of information sharing and data security. The standardized labeling process not only improved the uniformity and standardization of cross-border commodity identification but also optimized the efficiency of traditional labeling operations through digital means, meeting the differentiated needs of regulatory agencies, consumers, and all parties in the supply chain, and further enhancing the intelligence level and operational efficiency of cross-border commodity information management.

[0012] In one possible implementation, generating product tags corresponding to product information includes:

[0013] Based on the aforementioned basic information, the product category corresponding to the product to be bid on is determined;

[0014] Based on the product type, determine the keywords corresponding to the basic information, and determine the extraction model corresponding to the basic information;

[0015] The product information is extracted based on each extraction model to obtain initial extraction words;

[0016] The initial extracted words are subjected to data checks to obtain extraction accuracy and extraction completeness;

[0017] Based on the extraction accuracy and extraction completeness, the basic information is iteratively extracted and checked until the preset conditions are met to obtain the target extraction words;

[0018] Each target word is used as a basic tag, and the corresponding 3D model of the product to be tagged is obtained;

[0019] Based on the basic tags and the 3D model of the product, product tags corresponding to the product information are generated.

[0020] In one possible implementation, determining the extraction model corresponding to the basic information includes:

[0021] Determine whether image sub-information exists in the basic information;

[0022] If the basic information contains image sub-information, then the extraction model corresponding to the basic information is determined to include an image extraction model and a text extraction model;

[0023] If there is no image sub-information in the basic information, then the extraction model corresponding to the basic information is determined to be a text extraction model.

[0024] In one possible implementation, based on the basic tag and the product 3D model, a product tag corresponding to the product information is generated, including:

[0025] The production information is input into the Embedding model and the sparse coding model respectively, and the text vector output by the Embedding model and the feature representation output by the sparse coding model are obtained.

[0026] Obtain the feature vector constructed based on artificial features corresponding to the decision tree model, and concatenate the text vector with the feature vector to form a first comprehensive feature vector. Concatenate the feature representation with the feature vector to form a second comprehensive feature vector.

[0027] The first comprehensive feature vector and the second comprehensive feature vector are respectively input into the decision tree model, and the first extracted data and the second extracted data output by the decision tree model are obtained.

[0028] The first and second extracted data are transformed to obtain multiple extracted words;

[0029] Each extracted word is identified as a production tag, and based on the basic tag, the production tag, and the product 3D model, a product tag corresponding to the product information is generated.

[0030] In one possible implementation, based on the base tag, the production tag, and the product 3D model, a product tag corresponding to the product information is generated, including:

[0031] Extract the production process and traceability results of the product to be labeled from the production information;

[0032] Determine the production node corresponding to the traceability result, and determine the representation format of the traceability result;

[0033] Based on the production process, the traceability results, the production nodes corresponding to the traceability results, and the presentation format, a traceability tag corresponding to the product information is generated.

[0034] Identify the marketing information to determine the marketing scope corresponding to the product to be marketed;

[0035] Filter the downstream industries corresponding to the marketing scope;

[0036] Based on the marketing scope and the downstream industries, determine the marketing tags corresponding to the product information;

[0037] Based on the basic label, the production label, the traceability label, the marketing label, and the product 3D model, product labels corresponding to the product information are generated.

[0038] In one possible implementation, the query permissions for each labeling section are determined, including:

[0039] Based on the product information, the importance of the product to be bid on is determined;

[0040] Determine the permission level corresponding to each labeling section and determine the dynamic adjustment mechanism corresponding to the labeling section. The dynamic adjustment mechanism is used to update query permissions in real time.

[0041] Based on the importance level and the permission level, and in conjunction with the dynamic adjustment mechanism, the query permissions corresponding to each labeling section are determined.

[0042] In one possible implementation, determining the importance of the product to be labeled based on the product information includes:

[0043] Determine the first importance level corresponding to the basic information, the second importance level corresponding to the production information, and the third importance level corresponding to the marketing information;

[0044] Based on the first importance level, the second importance level, and the third importance level, the initial importance value of the product to be bid on is obtained by weighted calculation;

[0045] Obtain the historical circulation data of the product to be bid on, and correct the initial importance value based on the historical circulation data to obtain the final importance value;

[0046] The final importance value is compared with a preset importance threshold range to determine the importance of the product to be labeled.

[0047] Secondly, this application provides a cross-border commodity labeling device, which adopts the following technical solution:

[0048] A cross-border commodity labeling device, comprising:

[0049] The acquisition module is used to acquire product information of the product to be bid on. The product information includes basic information, production information and marketing information. The basic information is used to characterize the product attributes of the product to be bid on.

[0050] The generation module is used to generate product tags corresponding to the product information. The product tags include basic tags, production tags, traceability tags, and marketing tags.

[0051] The determination module is used to determine the labeling section corresponding to each product label and to determine the query permissions corresponding to each labeling section;

[0052] The completion module is used to complete the labeling process of the trademark to be labeled based on the labeling section and the corresponding query permissions.

[0053] Thirdly, this application provides an electronic device that adopts the following technical solution:

[0054] An electronic device comprising:

[0055] At least one processor;

[0056] Memory;

[0057] At least one application, wherein the at least one application is stored in memory and configured to be executed by at least one processor, the at least one application being configured to: perform the cross-border commodity labeling method described in any of the preceding claims.

[0058] Fourthly, this application provides a computer-readable storage medium, which adopts the following technical solution:

[0059] A computer-readable storage medium includes: a computer program stored thereon that can be loaded by a processor and execute the cross-border commodity labeling method described in any of the preceding claims.

[0060] In summary, this application includes the following beneficial technical effects:

[0061] By systematically acquiring complete product information, including basic, production, and marketing information, a multi-dimensional product data system was constructed, laying the data foundation for accurate labeling. Labels were then categorized into four main types: basic labels, production labels, traceability labels, and marketing labels, achieving a structured presentation of product information. This not only met the needs for displaying basic attributes but also covered value-added information such as production traceability and marketing promotion. Furthermore, by configuring differentiated query permissions for different labeling sections, information transparency was ensured while effectively protecting commercially sensitive information, balancing the needs of information sharing and data security. The standardized labeling process not only improved the uniformity and standardization of cross-border product identification but also optimized the efficiency of traditional labeling operations through digital means, meeting the differentiated needs of regulatory agencies, consumers, and all parties in the supply chain, and further enhancing the intelligence level and operational efficiency of cross-border product information management. Attached Figure Description

[0062] Figure 1 This is a flowchart illustrating a cross-border commodity labeling method provided in an embodiment of this application;

[0063] Figure 2This is a block diagram of a cross-border commodity labeling device provided in an embodiment of this application;

[0064] Figure 3 This is a schematic diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0065] The following is in conjunction with the appendix Figure 1 -Appendix Figure 3 This application will be described in further detail.

[0066] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, 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, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0067] This application provides a method for labeling cross-border goods, such as... Figure 1 As shown, the method provided in this application embodiment is executed by an electronic device, which can be a server or a terminal device. The server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smartphone, tablet, laptop, desktop computer, etc., but is not limited to these. The terminal device and the server can be directly or indirectly connected via wired or wireless communication. This application embodiment does not impose any limitations on this connection. The method includes steps S101-S104, wherein:

[0068] Step S101: Obtain product information of the product to be bid on.

[0069] The product information includes basic information, production information, and marketing information. The basic information is used to characterize the product attributes of the product to be labeled.

[0070] Among them, the goods awaiting labeling are cross-border goods that have not yet completed the labeling process and are about to enter the labeling process.

[0071] Basic information is data used to describe the basic attributes of goods, including product name, HS code, specifications, country of origin, material composition, applicable standards, etc. This information is a key element for the identification, classification and customs clearance of goods in cross-border trade.

[0072] Production information is data that records the production process of a product, such as the name of the production factory, production date, production batch, source of raw materials, processing technology, and quality inspection report number. It can be used for product quality traceability and production process management.

[0073] Marketing information consists of data related to product sales and promotion, including instructions for use, after-sales service policies, key selling points, and industry type.

[0074] Specifically, the electronic device contains a product database with information on the products to be bid on. The electronic device can directly retrieve this information from the product database. Specifically, the electronic device interfaces with the enterprise's supply chain management system, product database, or external data interfaces (such as product coding databases or supplier information systems) to obtain relevant information about the products to be bid on and stores this information in the product database. For example, after a product is manufactured, the barcode scanner on the production line transmits the product's unique code (such as the Global Trade Item Number (GTIN)) to the electronic device. Based on this code, the electronic device retrieves basic information, production information, and marketing information from the database. For products procured through cross-border channels, the electronic device can also extract product information from electronic documents or data interfaces provided by the supplier.

[0075] Step S102: Generate product tags corresponding to product information.

[0076] Product labels include basic labels, production labels, traceability labels, and marketing labels.

[0077] After obtaining the product information for the products to be labeled, the appropriate label templates can be called for processing based on different types of product information. For basic information, data such as product name, specifications, and brand are automatically filled into the basic label template to generate a concise, clear, and easily identifiable basic label for consumers. For production information, by extracting information such as production batch, production time, and process, and combining it with a production flow chart template, detailed and intuitive production labels are generated. When generating traceability labels, quality inspection data and traceability results from each production stage are integrated, and blockchain technology is used to generate a unique and tamper-proof traceability code for each product, forming a highly credible traceability label. For marketing labels, eye-catching promotional slogans, activity QR codes, and other elements are added based on marketing information such as sales scope and promotional activities to generate attractive marketing labels.

[0078] In this embodiment, generating product tags corresponding to product information includes:

[0079] Based on the basic information, determine the product category corresponding to the product to be bid on;

[0080] Based on the product type, determine the keywords corresponding to the basic information, and determine the extraction model corresponding to the basic information;

[0081] Product information is extracted based on each extraction model to obtain initial extraction words;

[0082] Perform data checks on the initial extracted words to obtain extraction accuracy and completeness;

[0083] Based on the accuracy and completeness of extraction, the basic information is iteratively extracted and checked until the preset conditions are met to obtain the target words.

[0084] Each target word is extracted as a basic tag, and the corresponding 3D model of the product to be tagged is obtained;

[0085] Based on the basic tags and the 3D model of the product, product tags corresponding to the product information are generated.

[0086] Once the basic information of the product to be bid on is obtained, the key attributes in the basic information (such as product name, material composition, etc.) are compared and matched with the category features in the product database. For example, if the basic information shows that the product is a "stainless steel thermos cup", by analyzing the "stainless steel" material and the purpose of "thermos cup", the matching category "kitchenware - thermos cup" is found in the database, thereby determining the product category corresponding to the product to be bid on.

[0087] Furthermore, based on the determined product category, keywords related to that category can be extracted from a pre-defined keyword library. For example, for the category "kitchenware - thermos cups," the keyword library would include keywords such as "insulation performance," "capacity," "material," and "brand." Simultaneously, based on the characteristics of the product category and keywords, a suitable extraction model is selected from several pre-defined models. These extraction models are built upon different algorithms and data processing logics. For instance, for basic information containing a large amount of textual description, a natural language processing-based extraction model might be chosen; for basic information with a large amount of structured data, a rule-matching-based extraction model would be selected. The electronic device will automatically call the most suitable extraction model to process the basic information based on the specific circumstances of the product category and keywords. Further, the selected extraction model is invoked, and the basic information of the product to be labeled is input into the model. The extraction model scans, analyzes, and filters the basic information according to its own set rules and algorithms.

[0088] Furthermore, using pre-defined data checking rules and evaluation algorithms, a comprehensive check is performed on the initially extracted words. For accuracy checks, the electronic device compares the initial extracted words with the original content of the basic information to determine whether the extracted words accurately reflect the true attributes of the product. For example, it checks whether the extracted keyword "material" matches the actual material labeled in the basic information. For completeness checks, based on the keyword library and product category requirements, it determines whether all keywords corresponding to key attributes have been extracted. For example, for a "thermos cup" product, it checks whether necessary keywords such as "capacity" and "insulation time" have been extracted. Through this series of checks and evaluations, the accuracy and completeness values ​​of the initial extracted words can be calculated. Simultaneously, the electronic device connects to a 3D product model database and retrieves the corresponding 3D product model based on the product's category, model, and other information. This 3D model is a digital 3D representation of the product's appearance and structure. The product 3D model is a 3D digital model of the product constructed using computer technology, which can intuitively display the product's three-dimensional shape and details.

[0089] Furthermore, the basic tags are integrated with the product's 3D model. Specifically, based on the obtained extraction accuracy and completeness values, it is determined whether preset conditions are met (e.g., accuracy above 95%, completeness above 90%). If the conditions are not met, the parameters of the extraction model are automatically adjusted or a more suitable extraction model is used to extract the basic information again, obtaining new initial extracted words. Then, the data checking steps are repeated to calculate new extraction accuracy and completeness. This process is iterated and continuously optimized until the extraction accuracy and completeness meet the preset conditions. The final extracted words obtained at this point are the target extracted words. Each target extracted word is treated as an independent information unit, and according to the pre-designed basic tag template format, the target extracted word is filled into the corresponding position to generate a basic tag. The basic tag is then integrated with the product's 3D model to obtain the product tag.

[0090] More specifically, in this embodiment, determining the extraction model corresponding to the basic information includes:

[0091] Determine whether image sub-information exists in the basic information;

[0092] If image sub-information exists in the basic information, then the extraction model corresponding to the basic information is determined to include both image extraction model and text extraction model;

[0093] If there is no image sub-information in the basic information, then the extraction model corresponding to the basic information is determined to be the text extraction model.

[0094] After receiving the basic information of the product, the device can scan the storage format and content type of the information. Specifically, it identifies the data format of the basic information. For example, if the basic information exists in document form, it checks whether the document contains image elements; if the basic information is stored as a database record, it checks whether the corresponding fields contain the storage path of an image file or image encoding data. Further, it can perform preliminary analysis of the information content, such as checking for character combinations similar to image file extensions (.jpg, .png, etc.) or specific image identification codes. Through this series of format recognition and content screening operations, the electronic device determines whether image sub-information exists within the basic information.

[0095] Once image sub-information is detected within the basic information, the corresponding extraction model is retrieved from its built-in model library. For the image portion, an image extraction model can be enabled. This model can parse the image, recognizing features such as product appearance, color, and size markings. Simultaneously, to extract any textual information that may be present in the image (such as product model numbers or parameter descriptions), a text extraction model is invoked. This model, based on Optical Character Recognition (OCR) technology, converts the text in the image into editable text data. Combining these two models allows for the comprehensive extraction of all types of data contained in the basic information, accurately extracting both the image's inherent features and the textual information within it.

[0096] When the basic information does not contain image sub-information, meaning it consists only of text data such as words and numbers, a text extraction model can be directly selected from the model library. This text extraction model can perform semantic analysis, keyword extraction, and data classification on text data. By recognizing words, phrases, and sentence structures in the text, it extracts key information related to product attributes, such as product type, material, and function from the product name and description. Using this model to process the basic information completes the information extraction process, providing data support for the subsequent generation of product labels.

[0097] Furthermore, in this embodiment, based on the basic tags and the 3D model of the product, product tags corresponding to the product information are generated, including:

[0098] The production information is input into the Embedding model and the sparse coding model respectively, and the text vector output by the Embedding model and the feature representation output by the sparse coding model are obtained.

[0099] Obtain the feature vectors constructed based on manually generated features corresponding to the decision tree model, and concatenate the text vectors with the feature vectors to form the first comprehensive feature vector. Concatenate the feature representations with the feature vectors to form the second comprehensive feature vector.

[0100] The first comprehensive feature vector and the second comprehensive feature vector are respectively input into the decision tree model, and the first extracted data and the second extracted data output by the decision tree model are obtained.

[0101] The first and second extracted data are transformed to obtain multiple extracted words;

[0102] Each extracted word is designated as a production tag, and based on the basic tag, production tag, and product 3D model, product tags corresponding to the product information are generated.

[0103] After obtaining the production information of the goods, this information (such as the name of the production factory, production date, production batch, raw material source, etc.) can be preprocessed to remove redundant characters and unify the format before being transmitted to the Embedding model and the sparse coding model respectively. The Embedding model performs semantic analysis on the text content in the production information, mapping each word to a low-dimensional dense vector, thus converting the text information into a numerical form that is easy for computers to process, and outputting a text vector. The sparse coding model processes the production information from another perspective. By analyzing the feature distribution in the data, it extracts representative features and represents these features in the form of sparse vectors, outputting a feature representation. For example, for the information "organic raw material production", the sparse coding model may highlight key features such as "organic" and "raw material" in the form of sparse vectors.

[0104] Furthermore, from a pre-defined rule base, feature vectors constructed by the decision tree model based on manually defined key features (such as industry-standard production standards and enterprise-defined quality indicators) are retrieved. Next, the text vector output by the Embedding model is concatenated with these manually defined feature vectors, forming a first comprehensive feature vector that includes both textual semantic information and manually defined key features. Simultaneously, the feature representation output by the sparse coding model is also concatenated with the manually defined feature vectors in the same way, forming a second comprehensive feature vector that integrates both data-driven key features and manually defined features. For example, if the manually defined feature vector contains the feature "production standard level," and the text vector or feature representation contains details of the production process, the concatenation yields a more comprehensive feature vector reflecting production information.

[0105] The generated first and second comprehensive feature vectors are sequentially input into the decision tree model. The decision tree model analyzes and judges the input comprehensive feature vectors layer by layer according to pre-trained decision rules. Starting from the root node, it branches downwards according to predetermined splitting conditions based on the values ​​of each dimension in the feature vector, eventually reaching the leaf nodes and outputting the corresponding results. For the first comprehensive feature vector, the decision tree model outputs the first extracted data reflecting the fusion of textual semantics and manual features in the production information; for the second comprehensive feature vector, it outputs the second extracted data based on the combination of key data features and manual features. For example, the decision tree model might extract data from the first comprehensive feature vector stating "compliant with XX international production standards," and from the second comprehensive feature vector stating "the key raw material originates from XX region."

[0106] After obtaining the first and second extracted data, these data can be formatted and parsed. For structured data (such as production standard numbers), it can be converted into easily understandable text descriptions; for unstructured text data, operations such as word segmentation and stop word removal will be performed to extract core vocabulary. Each extracted word obtained in the previous step is treated as an independent information unit, and according to the preset template format of the production label, the extracted word is filled into the corresponding position to generate multiple production labels. For example, extracted words such as "ISO9001 certification" and "New Zealand milk source" are made into separate production labels. Then, the previously generated basic labels and the newly generated production labels are combined with the product's 3D model to generate product labels.

[0107] Furthermore, in this embodiment, based on the basic tag, production tag, and product 3D model, a product tag corresponding to the product information is generated, including:

[0108] Extract the production process and traceability results of the product to be labeled from the production information;

[0109] Identify the production node corresponding to the traceability results, and determine the corresponding representation format of the traceability results;

[0110] Based on the production process, traceability results, the production nodes corresponding to the traceability results, and the presentation format, generate traceability tags corresponding to the product information;

[0111] Identify marketing information to determine the marketing scope corresponding to the product to be bid on;

[0112] Filter the downstream industries corresponding to the marketing scope;

[0113] Based on the marketing scope and downstream industries, determine the marketing tags corresponding to the product information;

[0114] Based on the basic label, production label, traceability label, marketing label, and the product's 3D model, product labels corresponding to the product information are generated.

[0115] Upon receiving the production information of the goods, a structured analysis is performed. This production information typically includes records from multiple stages, such as raw material procurement, processing and manufacturing, quality inspection, packaging, and warehousing. By using pre-defined keyword matching and text parsing rules, key information for each stage can be identified, thus extracting the complete production process. For example, phrases like "raw materials were procured from XX farm," "bottling was done at XX factory," and "tested by XX quality inspection agency" can identify the production stages, including raw material procurement, processing and manufacturing, and quality inspection. Furthermore, traceable data regarding the product's origin, flow, and quality status, such as raw material batch numbers, processing timestamps, and logistics records, are retrieved from the product database and integrated into a traceability result.

[0116] Based on the extracted traceability results, analyze the key time points and operational points involved to determine the corresponding production nodes. For example, raw material warehousing time, processing start time, and quality inspection completion time can all be considered production nodes. The format for presenting the traceability results can be chosen based on the company's needs and the habits of the target market. If target market consumers are accustomed to obtaining detailed information by scanning codes, the traceability results can be generated as QR codes or barcodes; if a more intuitive display is required, charts or text lists can be used to clearly present the traceability results.

[0117] The extracted information, including production processes, traceability results, and production milestones, is integrated according to a pre-designed traceability label template. If QR codes are chosen as the presentation format, this information is encoded into corresponding QR codes, with necessary text descriptions added, such as "Scan the code to view product traceability information." If charts are used, the production processes and milestones are transformed into visual flowcharts or timeline charts, accompanied by concise text descriptions of the traceability results. Finally, the generated content is laid out according to label layout requirements to create a complete traceability label.

[0118] The marketing information for products undergoes text analysis and semantic understanding. Specifically, marketing information typically includes details such as the target customer group, sales area, and applicable scenarios. Keyword extraction and semantic association analysis can be used to determine the product's marketing scope. For example, if the marketing information mentions "targeting young and fashionable consumers, sold in first- and second-tier cities," then the target customer group is identified as young and fashionable individuals, and the sales area is first- and second-tier cities, thus defining the marketing scope. Furthermore, the applicable scenarios for the product, such as "suitable for use during sports and fitness," are analyzed to further clarify the marketing scope.

[0119] Based on the defined marketing scope, and referencing industry classification databases and market research data, downstream industries related to that marketing scope are selected. For example, if the product's marketing scope targets sports and fitness enthusiasts, the industry database will filter downstream industries such as sporting goods retail, fitness clubs, and sports event organizations; if the product targets the maternal and infant market, downstream industries such as maternal and infant product stores, pediatric medical institutions, and early childhood education institutions will be selected. In this way, the industries of potential sales channels and partners for the product are identified. Based on the characteristics of the marketing scope and downstream industries, suitable templates are selected from a pre-designed marketing label template library, and the relevant information is populated.

[0120] The generated basic labels, production labels, traceability labels, and marketing labels are integrated and combined with the product's 3D model to obtain the product label.

[0121] Step S103: Determine the labeling section corresponding to each product label and determine the query permissions corresponding to each labeling section.

[0122] The labeling section is a specific area in the cross-border commodity display system used to display labels for different types of commodities. Each section corresponds to one or more types of commodity labels.

[0123] The cross-border product display system can be divided into four main labeling sections: basic information, production information, smart traceability, and product marketing. Basic labels, production labels, traceability labels, and marketing labels correspond to these sections, respectively. After obtaining the product labels, each label is entered into its corresponding labeling section. Furthermore, when determining query permissions, permissions can be set based on internal management regulations, the sensitivity of product information, and relevant laws and regulations. For basic labels, query access is open to all consumers to facilitate their understanding of basic product information; production labels are only accessible to internal management personnel, quality supervision departments, and authorized partners to ensure the security of production information; traceability labels allow consumers to query basic traceability information, while complete and detailed traceability data is only available to the enterprise, regulatory agencies, and specific supply chain participants; marketing labels are open to all potential consumers to achieve product promotion.

[0124] Specifically, in this embodiment, determining the query permissions corresponding to each labeling section includes:

[0125] Based on the product information, determine the importance of the product to be bid on;

[0126] Determine the permission level corresponding to each labeling section and determine the dynamic adjustment mechanism for the labeling section. The dynamic adjustment mechanism is used to update query permissions in real time.

[0127] Based on the importance and the aforementioned permission level, and combined with a dynamic adjustment mechanism, the query permissions corresponding to each labeling section are determined.

[0128] A comprehensive analysis is conducted on the acquired product information (including basic information, production information, and marketing information). For basic information, products belonging to high-value luxury goods, medical devices, or other highly regulated categories, or those with special uses (such as military-to-civilian conversion products), are assigned a higher importance weight. In production information, products involving special processes or stringent quality certifications (such as aerospace-grade material certification) also have increased importance. Regarding marketing information, products that are core brand products, limited editions, or have significant market influence will also have their importance score increased. Based on pre-set scoring rules and weighting systems, each piece of information is quantitatively scored, ultimately resulting in a comprehensive importance level for the product being evaluated, categorized into three levels: "Ordinary," "Important," and "Highly Important."

[0129] Based on internal company regulations and industry practices, different permission levels are pre-set for different labeling sections. For example, basic labels mainly display the product's basic attributes and are intended for personnel in basic distribution channels such as customs and logistics, and are set to a lower "public" permission level; production labels contain sensitive information such as production processes and quality inspections, and are only allowed to be viewed by internal quality management personnel and regulatory agencies, and are set to an "internal" permission level; traceability labels involve data on the entire product traceability process, with some information needing to be disclosed to consumers and others needing to be kept confidential, and are set to a "semi-public" permission level; marketing labels are mainly used to attract consumers and are intended for the general public, and are set to a "fully public" permission level. These pre-set permission levels are bound to the corresponding labeling sections to form a fixed permission mapping relationship. At the same time, the electronic device is equipped with a dynamic adjustment mechanism for each labeling section. This mechanism is linked to the product's real-time data sources (such as sales data systems, policy update platforms, and user feedback channels). When a trigger condition is detected (such as a change in policy related to a labeling section's information, or a quality complaint about the product), the dynamic adjustment mechanism is automatically activated, updating the permission level according to pre-set rules (such as increasing the permission level when policies tighten, or expanding the query scope when the number of complaints reaches a certain level). Among them, the dynamic adjustment mechanism is a rule system that can automatically update the permission level of the labeling section based on real-time data (such as policy changes and market feedback) to ensure that query permissions are synchronized with actual needs.

[0130] The determination of product importance is combined with the set access levels for labeling sections. For products with low importance, even if a labeling section originally had a high preset access level, its access may be appropriately reduced. For example, the production labels of ordinary daily necessities, originally at the "internal level," may be adjusted to allow limited viewing by cooperating suppliers due to the low importance of the product itself. For products with high importance, the preset high access level is strictly enforced. For example, the production labels of high-value medical equipment are prohibited from being viewed by anyone other than company personnel and regulatory agencies. Simultaneously, a dynamic adjustment mechanism is activated to continuously monitor related real-time data. If the product is detected to be under close scrutiny by regulatory agencies due to quality issues (triggering the dynamic adjustment condition), the dynamic adjustment mechanism will adjust the access level of the traceability label section from "tiered public level" to "regulatory priority" (regulatory agencies can view all traceability information) according to the rules. Finally, based on the combined importance, initial access level, and dynamic adjustment results, the specific query permissions for each labeling section for different users (such as consumers, company employees, and regulatory agencies) are determined and stored in the access management system for verification in subsequent information queries.

[0131] More specifically, in this embodiment, determining the importance of the product to be labeled based on product information includes:

[0132] Determine the first importance level for basic information, the second importance level for production information, and the third importance level for marketing information;

[0133] Based on the first, second, and third importance levels, the initial importance value of the product to be bid on is obtained through weighted calculation;

[0134] Obtain historical circulation data of the goods to be bid on, and adjust the initial importance value based on the historical circulation data to obtain the final importance value;

[0135] The final importance value is compared with the preset importance threshold range to determine the importance of the product to be labeled.

[0136] The system acquires pre-defined importance level assessment rules for different types of information. For basic information, importance can be determined based on the product's HS code, country of origin, and material composition. Different HS codes correspond to different scores, and different countries of origin and different material grades also correspond to different scores. The HS code, country of origin, and material composition of the product to be bid on are compared with the pre-defined content scores to obtain the score for each item of the product to be bid on. These scores are then added together to obtain a first sum. This first sum is compared with a first pre-defined level threshold to obtain the first importance level of the product to be bid on. Similarly, for production information, importance can be determined based on the product's production process and raw material source. Different production processes correspond to different scores, and different raw material sources also correspond to different scores. The production process and raw material source of the product to be bid on are compared with the pre-defined content scores to obtain the score for each item of the product to be bid on. These scores are then added together to obtain a second sum. This second sum is compared with a second pre-defined level threshold to obtain the second importance level of the product to be bid on. Similarly, for marketing information, we can analyze whether the product is a core brand product, a limited edition, or part of a marketing campaign with significant market impact. If these conditions are met, the third level of importance will be higher; for products with ordinary promotion, the importance level of marketing information will be lower. Based on these rules, we can evaluate basic information, production information, and marketing information separately to determine their respective importance levels.

[0137] Obtain the preset weight values ​​for basic information, production information, and marketing information. For example, for technology-intensive products, the weight of production information is set to 0.5, the weight of basic information is 0.3, and the weight of marketing information is 0.2. The first, second, and third importance levels can be numerically processed (e.g., "low" corresponds to 1 point, "medium" to 2 points, and "high" to 3 points), then multiplied by their respective weight values, and the products are added together to obtain the initial importance value of the product. For example, if a product's first importance level is "medium" (2 points) with a weight of 0.3; the second importance level is "high" (3 points) with a weight of 0.5; and the third importance level is "low" (1 point) with a weight of 0.2, then the initial importance value = 2 × 0.3 + 3 × 0.5 + 1 × 0.2 = 2.3 points.

[0138] Historical circulation data of the target product is obtained from the product database. This data includes sales volume and sales revenue for each historical period. Time-domain analysis is performed on the historical circulation data to obtain sales volume and sales revenue change curves. Based on these curves, the degree of sales change (positive or negative) is determined. Using the formula: Final Importance Value = (1 + Sales Change Degree) * Initial Importance Ratio, a revised final importance value is obtained and used as the importance of the target product.

[0139] Step S104: Based on the labeling section and the corresponding query permissions, complete the labeling process for the trademark to be labeled.

[0140] After obtaining the labeling section corresponding to each product label and the query permissions corresponding to each labeling section, the basic label, production label, traceability label and marketing label can be filled into the corresponding labeling section to complete the labeling process of the trademark to be labeled.

[0141] This application provides a method for labeling cross-border goods. By systematically acquiring complete product information including basic, production, and marketing information, a multi-dimensional product data system is constructed, laying the data foundation for accurate labeling. The labels are then divided into four categories: basic labels, production labels, traceability labels, and marketing labels, achieving a structured presentation of product information. This satisfies the need for displaying basic attributes while also covering value-added information such as production traceability and marketing promotion. Furthermore, by configuring differentiated query permissions for different labeling sections, information transparency is ensured while effectively protecting commercially sensitive information, balancing the needs of information sharing and data security. The standardized labeling process not only improves the uniformity and standardization of cross-border product identification but also optimizes the efficiency of traditional labeling operations through digital means, meeting the differentiated needs of regulatory agencies, consumers, and all parties in the supply chain, and further enhancing the intelligence level and operational efficiency of cross-border product information management.

[0142] The above embodiments describe a method for labeling cross-border goods from the perspective of process flow. The following embodiments describe an XX device from the perspective of virtual module or virtual unit. For details, please refer to the following embodiments.

[0143] See Figure 2 The cross-border commodity labeling device 20 may specifically include: an acquisition module 201, a generation module 202, a determination module 203, and a completion module 204, wherein:

[0144] A cross-border commodity labeling device 20, comprising:

[0145] The acquisition module 201 is used to acquire product information of the product to be bid on. The product information includes basic information, production information and marketing information. The basic information is used to characterize the product attributes of the product to be bid on.

[0146] The generation module 202 is used to generate product labels corresponding to product information. Product labels include basic labels, production labels, traceability labels, and marketing labels.

[0147] The determination module 203 is used to determine the labeling section corresponding to each product label and to determine the query permissions corresponding to each labeling section;

[0148] Module 204 is used to complete the labeling process of the trademark to be labeled based on the labeling section and the corresponding query permissions.

[0149] In one possible implementation of this application embodiment, when generating product tags corresponding to product information, the generation module 202 is specifically used for:

[0150] Based on the basic information, determine the product category corresponding to the product to be bid on;

[0151] Based on the product type, determine the keywords corresponding to the basic information, and determine the extraction model corresponding to the basic information;

[0152] Product information is extracted based on each extraction model to obtain initial extraction words;

[0153] Perform data checks on the initial extracted words to obtain extraction accuracy and completeness;

[0154] Based on the accuracy and completeness of extraction, the basic information is iteratively extracted and checked until the preset conditions are met to obtain the target words.

[0155] Each target word is extracted as a basic tag, and the corresponding 3D model of the product to be tagged is obtained;

[0156] Based on the basic tags and the 3D model of the product, product tags corresponding to the product information are generated.

[0157] In one possible implementation of this application embodiment, when determining the extraction model corresponding to the basic information, the generation module 202 is specifically used for:

[0158] Determine whether image sub-information exists in the basic information;

[0159] If image sub-information exists in the basic information, then the extraction model corresponding to the basic information is determined to include both image extraction model and text extraction model;

[0160] If there is no image sub-information in the basic information, then the extraction model corresponding to the basic information is determined to be the text extraction model.

[0161] In one possible implementation of this application embodiment, when the generation module 202 generates product tags corresponding to product information based on the basic tags and the product 3D model, it is specifically used for:

[0162] The production information is input into the Embedding model and the sparse coding model respectively, and the text vector output by the Embedding model and the feature representation output by the sparse coding model are obtained.

[0163] Obtain the feature vectors constructed based on manually generated features corresponding to the decision tree model, and concatenate the text vectors with the feature vectors to form the first comprehensive feature vector. Concatenate the feature representations with the feature vectors to form the second comprehensive feature vector.

[0164] The first comprehensive feature vector and the second comprehensive feature vector are respectively input into the decision tree model, and the first extracted data and the second extracted data output by the decision tree model are obtained.

[0165] The first and second extracted data are transformed to obtain multiple extracted words;

[0166] Each extracted word is designated as a production tag, and based on the basic tag, production tag, and product 3D model, product tags corresponding to the product information are generated.

[0167] In one possible implementation of this application embodiment, when the generation module 202 generates product tags corresponding to product information based on basic tags, marketing tags, and the product 3D model, it is specifically used for:

[0168] Extract the production process and traceability results of the product to be labeled from the production information;

[0169] Identify the production node corresponding to the traceability results, and determine the corresponding representation format of the traceability results;

[0170] Based on the production process, traceability results, the production nodes corresponding to the traceability results, and the presentation format, generate traceability tags corresponding to the product information;

[0171] Identify marketing information to determine the marketing scope corresponding to the product to be bid on;

[0172] Filter the downstream industries corresponding to the marketing scope;

[0173] Based on the marketing scope and downstream industries, determine the marketing tags corresponding to the product information;

[0174] Based on basic tags, production tags, traceability tags, marketing tags, and product 3D models, product tags corresponding to product information are generated.

[0175] In one possible implementation of this application embodiment, when determining the query permission corresponding to each labeling section, the determining module 203 is specifically used for:

[0176] Based on the product information, determine the importance of the product to be bid on;

[0177] Determine the permission level corresponding to each labeling section and determine the dynamic adjustment mechanism for the labeling section. The dynamic adjustment mechanism is used to update query permissions in real time.

[0178] Based on the importance and the aforementioned permission level, and combined with a dynamic adjustment mechanism, the query permissions corresponding to each labeling section are determined.

[0179] In one possible implementation of this application embodiment, when determining the importance of the product to be labeled based on product information, the determining module 203 is specifically used for:

[0180] Determine the first importance level for basic information, the second importance level for production information, and the third importance level for marketing information;

[0181] Based on the first, second, and third importance levels, the initial importance value of the product to be bid on is obtained through weighted calculation;

[0182] Obtain historical circulation data of the goods to be bid on, and adjust the initial importance value based on the historical circulation data to obtain the final importance value;

[0183] The final importance value is compared with the preset importance threshold range to determine the importance of the product to be labeled.

[0184] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0185] See Figure 3 This application also describes an electronic device from the perspective of a physical device, such as... Figure 3 As shown, Figure 3 The illustrated electronic device 300 includes a processor 301 and a memory 303. The processor 301 and the memory 303 are connected, for example, via a bus 302. Optionally, the electronic device 300 may also include a transceiver 304. It should be noted that in practical applications, the transceiver 304 is not limited to one type, and the structure of this electronic device 300 does not constitute a limitation on the embodiments of this application.

[0186] Processor 301 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 301 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0187] Bus 302 may include a pathway for transmitting information between the aforementioned components. Bus 302 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Bus 302 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 3 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0188] The memory 303 may be a ROM (Read Only Memory) or other type of static storage device capable of storing static information and instructions, RAM (Random Access Memory) or other type of dynamic storage device capable of storing information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0189] The memory 303 is used to store application code that executes the solution of this application, and its execution is controlled by the processor 301. The processor 301 is used to execute the application code stored in the memory 303 to implement the content shown in the foregoing method embodiments.

[0190] Among them, electronic devices include, but are not limited to: mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), and in-vehicle terminals (such as in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers, and can also be servers, etc. Figure 3 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0191] This application provides a computer-readable storage medium storing a computer program that, when run on a computer, enables the computer to execute the corresponding content in the aforementioned method embodiments.

[0192] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0193] The above are only some embodiments of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for labeling cross-border goods, characterized in that, include: Obtain product information of the product to be bid on, the product information including basic information, production information and marketing information, the basic information being used to characterize the product attributes of the product to be bid on; Generate product tags corresponding to the product information, including basic tags, production tags, traceability tags, and marketing tags; Identify the labeling section corresponding to each product label and determine the query permissions for each labeling section; Based on the labeling section and the corresponding query permissions, the labeling process for the trademark to be labeled is completed.

2. The cross-border commodity labeling method according to claim 1, characterized in that, The product tags corresponding to the generated product information include: Based on the aforementioned basic information, the product category corresponding to the product to be bid on is determined; Based on the product type, determine the keywords corresponding to the basic information, and determine the extraction model corresponding to the basic information; The product information is extracted based on each extraction model to obtain initial extraction words; The initial extracted words are subjected to data checks to obtain extraction accuracy and extraction completeness; Based on the extraction accuracy and extraction completeness, the basic information is iteratively extracted and checked until the preset conditions are met to obtain the target extraction words; Each target word is used as a basic tag, and the corresponding 3D model of the product to be tagged is obtained; Based on the basic tags and the 3D model of the product, product tags corresponding to the product information are generated.

3. The cross-border commodity labeling method according to claim 2, characterized in that, The step of determining the extraction model corresponding to the basic information includes: Determine whether image sub-information exists in the basic information; If the basic information contains image sub-information, then the extraction model corresponding to the basic information is determined to include an image extraction model and a text extraction model; If there is no image sub-information in the basic information, then the extraction model corresponding to the basic information is determined to be a text extraction model.

4. The cross-border commodity labeling method according to claim 2, characterized in that, The step of generating product tags corresponding to the product information based on the basic tags and the product 3D model includes: The production information is input into the Embedding model and the sparse coding model respectively, and the text vector output by the Embedding model and the feature representation output by the sparse coding model are obtained. Obtain the feature vector constructed based on artificial features corresponding to the decision tree model, and concatenate the text vector with the feature vector to form a first comprehensive feature vector. Concatenate the feature representation with the feature vector to form a second comprehensive feature vector. The first comprehensive feature vector and the second comprehensive feature vector are respectively input into the decision tree model, and the first extracted data and the second extracted data output by the decision tree model are obtained. The first and second extracted data are transformed to obtain multiple extracted words; Each extracted word is identified as a production tag, and based on the basic tag, the production tag, and the product 3D model, a product tag corresponding to the product information is generated.

5. The cross-border commodity labeling method according to claim 4, characterized in that, The process of generating product tags corresponding to the product information based on the basic tag, the production tag, and the product 3D model includes: Extract the production process and traceability results of the product to be labeled from the production information; Determine the production node corresponding to the traceability result, and determine the representation format of the traceability result; Based on the production process, the traceability results, the production nodes corresponding to the traceability results, and the presentation format, a traceability tag corresponding to the product information is generated. Identify the marketing information to determine the marketing scope corresponding to the product to be marketed; Filter the downstream industries corresponding to the marketing scope; Based on the marketing scope and the downstream industries, determine the marketing tags corresponding to the product information; Based on the basic label, the production label, the traceability label, the marketing label, and the product 3D model, product labels corresponding to the product information are generated.

6. The method for labeling cross-border goods according to any one of claims 1 to 5, characterized in that, The process of determining the query permissions corresponding to each labeling section includes: Based on the product information, the importance of the product to be bid on is determined; Determine the permission level corresponding to each labeling section and determine the dynamic adjustment mechanism corresponding to the labeling section. The dynamic adjustment mechanism is used to update query permissions in real time. Based on the importance level and the permission level, and in conjunction with the dynamic adjustment mechanism, the query permissions corresponding to each labeling section are determined.

7. The cross-border commodity labeling method according to claim 6, characterized in that, The step of determining the importance of the product to be bid on based on the product information includes: Determine the first importance level corresponding to the basic information, the second importance level corresponding to the production information, and the third importance level corresponding to the marketing information; Based on the first importance level, the second importance level, and the third importance level, the initial importance value of the product to be bid on is obtained by weighted calculation; The historical circulation data of the product to be bid is obtained, the initial importance value is corrected based on the historical circulation data to obtain the final importance value, and the importance of the product to be bid is determined based on the final importance value.

8. A cross-border commodity labeling device, characterized in that, include: The acquisition module is used to acquire product information of the product to be bid on. The product information includes basic information, production information and marketing information. The basic information is used to characterize the product attributes of the product to be bid on. The generation module is used to generate product tags corresponding to the product information. The product tags include basic tags, production tags, traceability tags, and marketing tags. The determination module is used to determine the labeling section corresponding to each product label and to determine the query permissions corresponding to each labeling section; The completion module is used to complete the labeling process of the trademark to be labeled based on the labeling section and the corresponding query permissions.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; Memory; At least one application, wherein the at least one application is stored in memory and configured to be executed by at least one processor, said at least one application being configured to: perform the cross-border commodity labeling method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed in the computer, the computer is instructed to perform the cross-border commodity labeling method according to any one of claims 1 to 7.

Images