An electric commodity name auditing method and device, electronic equipment and storage medium

Abstract

This application provides a method, apparatus, electronic device, and storage medium for reviewing e-commerce product names, belonging to the field of artificial intelligence technology. The method includes: obtaining an e-commerce product name to be reviewed; performing a target review step on the e-commerce product name to be reviewed to obtain a review result, the review result including: the e-commerce product name to be reviewed is compliant, or the e-commerce product name to be reviewed is non-compliant and the reason for non-compliance; wherein, the target review step includes: obtaining a first product name category to which the e-commerce product name to be reviewed belongs; determining a first product name filling specification corresponding to the first product name category based on the first product name category and the correspondence between product name categories and product name filling specifications; comparing the e-commerce product name to be reviewed with the first product name filling specification to obtain a review result for the e-commerce product name to be reviewed. The e-commerce product name review method, apparatus, electronic device, and storage medium provided in this application can improve the efficiency and accuracy of e-commerce product name review.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to a method, apparatus, electronic device and storage medium for verifying e-commerce product names. Background Technology

[0002] Because e-commerce product names directly affect the taxation, logistics efficiency, regulatory measures, merchant reputation, and customs clearance efficiency and costs of e-commerce products, the review of e-commerce product names is of paramount importance in the customs business process, especially in the customs business process of cross-border e-commerce.

[0003] Currently, most online product name reviews rely on manual verification by customs officials. However, the specifications for filling out online product names are frequently updated, and manual verification by customs officials alone cannot guarantee the efficiency and accuracy of the review process. Therefore, improving the efficiency and accuracy of online product name reviews has become an urgent problem to be solved. Summary of the Invention

[0004] The main objective of this application is to provide a method, apparatus, electronic device, and storage medium for verifying e-commerce product names, aiming to improve the efficiency and accuracy of e-commerce product name verification.

[0005] To achieve the above objectives, a first aspect of this application proposes a method for reviewing e-commerce product names, comprising: obtaining an e-commerce product name to be reviewed; performing a target review step on the e-commerce product name to be reviewed to obtain a review result of the e-commerce product name to be reviewed, wherein the review result includes: the e-commerce product name to be reviewed is compliant, or the e-commerce product name to be reviewed is non-compliant and the reason for non-compliance; wherein the target review step includes: obtaining a first product name category to which the e-commerce product name to be reviewed belongs; determining a first product name filling specification corresponding to the first product name category based on the first product name category and the correspondence between product name categories and product name filling specifications; comparing the e-commerce product name to be reviewed with the first product name filling specification to obtain a review result of the e-commerce product name to be reviewed.

[0006] In some embodiments, before performing the target review step on the e-commerce product name to be reviewed and obtaining the review result of the e-commerce product name to be reviewed, the method further includes: matching the e-commerce product name to be reviewed with historical review information in the historical review database; if there is no target historical review information matching the e-commerce product name to be reviewed in the historical review database, then performing the target review step on the e-commerce product name to be reviewed and obtaining the review result of the e-commerce product name to be reviewed; after performing the target review step on the e-commerce product name to be reviewed and obtaining the review result of the e-commerce product name to be reviewed, the method further includes: adding the e-commerce product name to be reviewed and the review result of the e-commerce product name to be reviewed as the target historical review information to the historical review database.

[0007] In some embodiments, after matching the product name to be reviewed with historical review information in the historical review database, the method further includes: if there is target historical review information in the historical review database that matches the product name to be reviewed, then determining the review result of the product name to be reviewed based on the target historical review information.

[0008] In some embodiments, the step of performing a target review step on the product name to be reviewed to obtain a review result for the product name to be reviewed includes: inputting the product name to be reviewed into a trained product name review model, and performing the target review step on the product name to be reviewed through the product name review model to obtain a review result for the product name to be reviewed; before inputting the product name to be reviewed into the trained product name review model and performing the target review step on the product name to be reviewed through the product name review model to obtain a review result for the product name to be reviewed, the method further includes: obtaining a training dataset, wherein the training dataset includes multiple... Training samples, each training sample including an e-commerce product name sample and the corresponding tag review result of the e-commerce product name sample; using the training dataset and a preset e-commerce product name review process, a preset e-commerce product name review model is trained until the training stop condition is met, resulting in a trained e-commerce product name review model; wherein, the preset e-commerce product name review process includes: obtaining the second product name category to which the e-commerce product name sample belongs; determining the second product name filling specification corresponding to the second product name category based on the second product name category and the correspondence between product name categories and product name filling specifications; comparing the e-commerce product name sample with the second product name filling specification to obtain the predicted review result of the e-commerce product name sample.

[0009] In some embodiments, training the e-commerce product name verification model using the training dataset until the training stopping condition is met to obtain a fully trained e-commerce product name verification model includes: for each training sample, performing the following steps: inputting the training sample and the preset e-commerce product name verification process into the e-commerce product name verification model; verifying the training sample using the e-commerce product name verification model based on the preset e-commerce product name verification process to obtain the predicted verification result of the training sample; determining the loss function value of the e-commerce product name verification model based on the predicted verification result and the tag verification result; if the loss function value does not meet the training stopping condition, adjusting the model parameters of the e-commerce product name verification model to obtain an updated e-commerce product name verification model, and training the updated e-commerce product name verification model using the next training sample until the training stopping condition is met to obtain a fully trained e-commerce product name verification model.

[0010] In some embodiments, if the product name sample in the training sample is a single product name, the product name to be reviewed is a single product name; if the product name sample in the training sample is a combined product name obtained by concatenating characters, the product name to be reviewed is a combined product name.

[0011] In some embodiments, before determining the first product name filling specification corresponding to the first product name category based on the first product name category and the correspondence between product name categories and product name filling specifications, the method further includes: obtaining a product name category library; the product name category library includes a set of product names of at least one product name category; determining the correspondence between the product name category to which the e-commerce product name belongs and the product name filling specification based on the e-commerce product names in the product name set; wherein, the e-commerce product names include standardized product names and non-standard product names.

[0012] In some embodiments, after obtaining the product name category library, the method further includes: selecting a target e-commerce product name from the e-commerce product name library; matching the target e-commerce product name with the product name categories in the product name category library; if a product name category corresponding to the target e-commerce product name exists in the product name category library, then adding the target e-commerce product name to the product name set of the corresponding product name category in the product name category library; if a product name category corresponding to the target e-commerce product name does not exist in the product name category library, then creating a corresponding product name category in the product name category library and adding the target e-commerce product name to the product name set of the corresponding product name category in the product name category library; wherein, the target e-commerce product name includes standardized product names and non-standardized product names.

[0013] To achieve the above objectives, a second aspect of this application proposes an e-commerce product name verification device, comprising: an acquisition module and an verification module; the acquisition module is used to acquire an e-commerce product name to be verified; the verification module is used to perform a target verification step on the e-commerce product name to be verified to obtain a verification result of the e-commerce product name to be verified, wherein the verification result indicates that the e-commerce product name to be verified is compliant, or that the e-commerce product name to be verified is non-compliant and the reason for non-compliance; wherein the target verification step includes: acquiring a first product name category to which the e-commerce product name to be verified belongs; determining a first product name filling specification corresponding to the first product name category based on the first product name category and the correspondence between product name categories and product name filling specifications; comparing the e-commerce product name to be verified with the first product name filling specification to obtain a verification result of the e-commerce product name to be verified.

[0014] To achieve the above objectives, a third aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect.

[0015] To achieve the above objectives, a fourth aspect of the present application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.

[0016] The e-commerce product name verification method, apparatus, electronic device, and storage medium proposed in this application obtain the first product name category to which the e-commerce product name to be verified belongs. Based on the first product name category and the correspondence between the product name category and the product name filling specification, the first product name filling specification corresponding to the first product name category to which the e-commerce product name to be verified belongs can be quickly and accurately determined. Furthermore, by comparing the e-commerce product name to be verified with the first product name filling specification, the verification result of whether the e-commerce product name to be verified is compliant or non-compliant and the reason for non-compliance can be quickly and accurately determined, thereby improving the verification efficiency and accuracy of e-commerce product names. Attached Figure Description

[0017] Figure 1 This is a flowchart of the e-commerce product name verification method provided in the embodiments of this application;

[0018] Figure 2 This is a flowchart of the training e-commerce product name verification model provided in the embodiments of this application;

[0019] Figure 3 yes Figure 2 The flowchart of step S202 in the text;

[0020] Figure 4a This is a schematic diagram of the loss function curve of the e-commerce product name verification model provided in the embodiments of this application. Figure 1 ;

[0021] Figure 4b This is a schematic diagram of the loss function curve of the e-commerce product name verification model provided in the embodiments of this application. Figure 2 ;

[0022] Figure 5 This is a flowchart illustrating the correspondence between the category of an e-commerce product name and the product name filling specifications, provided in an embodiment of this application.

[0023] Figure 6 This is a flowchart of the updated product name category library provided in this application;

[0024] Figure 7 This is a flowchart illustrating the e-commerce product name verification method provided in this application embodiment;

[0025] Figure 8 This is a schematic diagram of the structure of the e-commerce product name verification device provided in the embodiments of this application;

[0026] Figure 9This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0028] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0029] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0030] First, let's analyze some of the terms used in this application:

[0031] Artificial intelligence (AI) is a new branch of computer science that studies, develops, and applies theories, methods, technologies, and systems to simulate, extend, and expand human intelligence. It aims to understand the essence of intelligence and produce intelligent machines that can react in a way similar to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing, and expert systems. AI can simulate the information processes of human consciousness and thought. Furthermore, AI utilizes digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceiving the environment, acquiring knowledge, and using that knowledge to achieve optimal results.

[0032] Based on this, embodiments of this application provide a method, apparatus, electronic device, and storage medium for reviewing e-commerce product names, aiming to improve the efficiency and accuracy of e-commerce product name review.

[0033] The e-commerce product name verification method, apparatus, electronic device, and storage medium provided in this application are specifically described through the following embodiments. First, the e-commerce product name verification method in this application is described.

[0034] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0035] Foundational technologies for artificial intelligence generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0036] The e-commerce product name verification method provided in this application relates to the field of artificial intelligence technology and can be applied to e-commerce, logistics, and other scenarios. In e-commerce scenarios, e-commerce platforms have standardized requirements for product names. Before listing products on e-commerce platforms, the e-commerce product name verification method provided in this application can be used to verify the product names. This method can quickly and accurately determine whether a product name is compliant or non-compliant and the reasons for non-compliance. In the case of non-compliant product names, the product names can be adjusted according to the reasons for non-compliance, which helps to successfully list products on e-commerce platforms. In logistics scenarios, importing countries have standardized requirements for the names of imported e-commerce goods. Non-compliant names can lead to customs clearance difficulties or fines due to goods being detained at customs. Before customs clearance, the e-commerce goods name verification method provided in this application can be used to verify the names of e-commerce goods. This method can quickly and accurately determine whether the e-commerce goods names are compliant or non-compliant, and the reasons for non-compliance. If the names are non-compliant, they can be adjusted based on the reasons for non-compliance, which helps e-commerce goods clear customs smoothly and improves logistics efficiency.

[0037] The e-commerce product name verification method provided in this application can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the e-commerce product name verification method, but is not limited to the above forms.

[0038] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0039] Figure 1 This is an optional flowchart of the e-commerce product name verification method provided in the embodiments of this application. Figure 1 The method may include, but is not limited to, steps S101 to S102.

[0040] Step S101: Obtain the product name of the e-commerce product to be reviewed;

[0041] Step S102: Perform the target audit steps on the product name to be audited to obtain the audit result of the product name to be audited. The audit result includes: the product name to be audited is compliant, or the product name to be audited is non-compliant and the reason for non-compliance.

[0042] The target review process includes:

[0043] Obtain the primary product category to which the e-commerce product name to be reviewed belongs;

[0044] Based on the first product category and the correspondence between product category and product name filling specifications, determine the first product name filling specifications corresponding to the first product category;

[0045] The product name to be reviewed is compared with the first product name filling specification to obtain the review result of the product name to be reviewed.

[0046] Steps S101 to S102 as illustrated in this embodiment of the application obtain the first product name category to which the product name to be reviewed belongs, and based on the first product name category and the correspondence between the product name category and the product name filling specification, the first product name filling specification corresponding to the first product name category to which the product name to be reviewed belongs can be quickly and accurately determined. Furthermore, by comparing the product name to be reviewed with the first product name filling specification, the review result of whether the product name to be reviewed is compliant or non-compliant and the reason for non-compliance can be quickly and accurately determined, thereby improving the review efficiency and accuracy of product names.

[0047] In some embodiments, the correspondence between product name categories and product name filling specifications may include, but is not limited to:

[0048] |Serial Number|Product Category|Product Name Filling Guidelines|;

[0049] |1|Clothing / clothing / appare|Target audience + material + specific product name|;

[0050] |2|Shoes / footwear|Material + Specific Product Name|;

[0051] |3|Bag|Material + Specific Product Name|;

[0052] |4|Toy|Material + Specific Product Name|;

[0053] |5|game|Specific purpose + specific product name|;

[0054] |6| Hats / Caps | Target audience / purpose + specific product name |;

[0055] |7|Shaver|Specific Product Name|;

[0056] |8|Gloves|Specific Product Name|;

[0057] |9|Curtains|Specific Product Name|;

[0058] |10|Musical Instruments|Specific Names|.

[0059] In some embodiments, the audit results may also include the execution results of each sub-step in the target audit step for the product name to be audited.

[0060] In one application scenario, taking the e-commerce product name "handbag" as an example, the product name category of the product name "handbag" is "bag". According to the correspondence between product name category and product name filling specifications, the product name filling specifications corresponding to the product name category "bag" are "material + specific product name". Specifically, for the e-commerce product name "handbag" to be reviewed, the following steps are executed sequentially: Step S1: Obtain the product category to which the product name "handbag" belongs; Step S2: Based on the product category "bag" and the correspondence between product category and product name filling specifications, determine the product name filling specifications corresponding to the product category "bag"; Step S3: Compare the product name "handbag" to be reviewed with the product name filling specifications "material + specific product name" corresponding to the product category "bag" to obtain the review result of the product name "handbag" to be reviewed, and finally output the review result of the product name "handbag" to be reviewed: [S1] Bag; [S2] Material + specific product name; [S3] Non-compliant, lacking material description.

[0061] In another application scenario, taking the e-commerce product name "Children's Gloves" as an example, the product category to which the product name "Children's Gloves" belongs is "Gloves". According to the correspondence between product category and product name filling specifications, the product name filling specification corresponding to the product category "Gloves" is "Specific Product Name". The following steps are executed sequentially for the product name "Children's Gloves": Step S1: Obtain the product category to which the product name "Children's Gloves" belongs; Step S2: Based on the product category "Gloves" and the correspondence between product category and product name filling specifications, determine the product name filling specification corresponding to the product category "Gloves"; Step S3: Compare the product name "Children's Gloves" to the product name filling specification "Specific Product Name" corresponding to the product category "Gloves" to determine the review result of the product name "Children's Gloves". Finally, the review result of the product name "Children's Gloves" is output: [S1] Gloves; [S2] Specific Product Name; [S3] Compliant.

[0062] In this embodiment, by performing step S1 on the product name to be reviewed: obtaining the first product name category to which the product name to be reviewed belongs, the system can quickly and accurately match the input product name and categorize it into one of more than 50 product name types, such as electronic products, clothing, and food. Further, by performing step S2 on the product name to be reviewed: determining the first product name filling specification corresponding to the first product name category based on the first product name category and the correspondence between product name categories and product name filling specifications, the system can automatically extract the first product name filling specification corresponding to the first product name category after determining the first product name category to which the product name to be reviewed belongs, thus further passing the review process. Step S3 of the e-commerce product name review process: Compare the e-commerce product name to be reviewed with the first product name filling specification to obtain the review result of the e-commerce product name to be reviewed. According to the first product name filling specification, the e-commerce product name to be reviewed can be checked for compliance and the review result can be output, including whether the e-commerce product name to be reviewed is compliant or non-compliant, and the reason for non-compliance (such as the e-commerce product name to be reviewed lacking elements such as purpose, function, and target audience). Finally, the execution results of steps S1, S2, and S3 are output respectively. This not only improves the accuracy of e-commerce product name review, but also enhances the transparency of the e-commerce product name review process and improves the user experience.

[0063] In some embodiments, step S102, which involves performing a target audit step on the product name to be audited to obtain the audit result of the product name to be audited, includes: using a preset execution program to perform a target audit step on the product name to be audited to obtain the audit result of the product name to be audited.

[0064] In some other embodiments, step S102, which involves performing a target review step on the product name to be reviewed to obtain the review result of the product name to be reviewed, includes: inputting the product name to be reviewed into the trained product name review model, performing the target review step on the product name to be reviewed through the product name review model, and obtaining the review result of the product name to be reviewed.

[0065] In some embodiments, before performing the target review step on the product name to be reviewed in step S102 to obtain the review result of the product name to be reviewed, the product name review provided in this application embodiment further includes: training the product name review model.

[0066] like Figure 2 As shown, training the e-commerce product name verification model may include, but is not limited to, steps S201 to S202:

[0067] Step S201: Obtain the training dataset, which includes a preset e-commerce product name review process and multiple training samples. Each training sample includes an e-commerce product name sample and its tag review results.

[0068] Step S202: Use the training dataset to train the e-commerce product name verification model until the training stopping condition is met, and obtain the trained e-commerce product name verification model;

[0069] The pre-defined e-commerce product name review process includes: obtaining the second product name category to which the e-commerce product name sample belongs; determining the second product name filling specification corresponding to the second product name category based on the second product name category and the correspondence between the product name category and the product name filling specification; and comparing the e-commerce product name sample with the second product name filling specification to obtain the predicted review result of the e-commerce product name sample.

[0070] Optionally, in some embodiments, the model structure of the e-commerce product name verification model can be a model structure with only a decoder part and no encoder part (decoder-on-ly structure), and its base model is the qwen1.5-7B-base model, but this does not limit the model structure of the e-commerce product name verification model.

[0071] Steps S201 to S202 shown in this embodiment of the application, through training an e-commerce product name verification model, follow this preset e-commerce product name verification process: first, obtaining the second product name category to which the e-commerce product name sample belongs; then, based on the second product name category and the correspondence between the product name category and the product name filling specification, determining the second product name filling specification corresponding to the second product name category; and finally, comparing the e-commerce product name sample with the second product name filling specification to obtain the preset verification result of the e-commerce product name sample. This process can improve the accuracy of the e-commerce product name verification model in verifying e-commerce product names, and at the same time, it can ensure that when users use the trained e-commerce product name verification model to verify e-commerce product names, they can not only obtain high e-commerce product name verification efficiency, but also high e-commerce product name verification accuracy.

[0072] Please see Figure 3 In some embodiments, step S202 may include, but is not limited to, performing steps S301 to S303 for each training sample:

[0073] Step S301: Input the training samples and the preset e-commerce product name review process into the e-commerce product name review model, and use the e-commerce product name review model to review the training samples based on the preset e-commerce product name review process to obtain the predicted review results of the training samples.

[0074] Step S302: Determine the loss function value of the e-commerce product name review model based on the predicted review results and the label review results;

[0075] Step S303: If the loss function value does not meet the training stopping condition, adjust the model parameters of the e-commerce product name verification model to obtain the updated e-commerce product name verification model, and use the next training sample to train the updated e-commerce product name verification model until the training stopping condition is met, and obtain the trained e-commerce product name verification model.

[0076] Steps S301 to S303 in this embodiment involve training an e-commerce product name verification model to verify training samples based on a preset e-commerce product name verification process, obtaining predicted verification results for the training samples, and adjusting the parameters of the e-commerce product name verification model after adjusting the parameters of the e-commerce product name verification model if the loss function value determined based on the predicted verification result and the tag verification result does not meet the training stopping condition. This process continues until the loss function value meets the training stopping condition, resulting in a fully trained e-commerce product name verification model. This ensures the reliability of the e-commerce product name verification model's training, improves the accuracy of the e-commerce product name verification model in verifying e-commerce product names, and enhances the user experience of using the e-commerce product name verification model to verify e-commerce product names.

[0077] Optionally, a preset e-commerce product name verification process is used as the system prompt word, e-commerce product name samples in the training samples are used as user prompt words, and the tag verification results corresponding to the e-commerce product name samples in the training samples are used as auxiliary prompt words. These are input into the e-commerce product name verification model. The e-commerce product name verification model verifies the user prompt words in the training samples based on the system prompt words to obtain the predicted verification results of the user prompt words in the training samples. Based on the predicted verification results and the auxiliary prompt words in the training samples, the loss function value of the e-commerce product name verification model is determined. If the loss function value does not meet the training stopping condition, the model parameters of the e-commerce product name verification model are adjusted to obtain an updated e-commerce product name verification model. The updated e-commerce product name verification model is then trained using the next training sample until the training stopping condition is met, resulting in a trained e-commerce product name verification model.

[0078] For example, the system prompt is as follows:

[0079] As a customs officer, you need to review the product names to be reviewed according to the correspondence between product category and product name filling specifications, following these steps:

[0080] Step S1: Please select the product category that best matches the e-commerce product name sample from the product category and product name filling specification correspondence table. This product category will be the second product category to which the e-commerce product name sample belongs. Please answer the question about the second product category. If the e-commerce product name sample does not belong to any product category in the product category column, please answer "Other".

[0081] Step S2: Based on the second product name category, please look up the correspondence table between product name categories and product name filling specifications, and answer the product name filling specifications corresponding to the second product name category in the correspondence table (i.e., the second product name specifications). If the second product name category is "Other", please answer with [Material + Specific Product Name].

[0082] Step S3: Compare the e-commerce product name sample with the second product name filling specification to determine whether the e-commerce product name sample is compliant. If it is compliant, please answer "compliant". If it is not compliant, please answer "non-compliant" and describe the missing elements in the e-commerce product name sample.

[0083] It's worth noting that the product name samples in the training samples can be a single product name or a combination of product names obtained by concatenating them using a concatenation operator. For example, when the product name sample in the training samples is a single product name, the user prompt can be any one of the following: men's short-sleeved shirts, piano, or plush toys. The auxiliary prompt corresponding to the user prompt "men's short-sleeved shirts" is

S1

S2

S3

S1

S2

S3

S1

S2

S3

S1

S2

S3

S1

S2

S3

S1

S2

S3

[0084] In this embodiment, by training the e-commerce product name verification model using a preset e-commerce product name verification process as system prompts, e-commerce product name samples from the training samples as user prompts, and the tag verification results corresponding to the e-commerce product name samples from the training samples as auxiliary prompts, the training efficiency and reliability of the e-commerce product name verification model, as well as the accuracy of the e-commerce product name verification model, can be improved, and the user experience of using the e-commerce product name verification model to verify e-commerce product names can be enhanced.

[0085] In some embodiments, the training task of the e-commerce product name verification model can be defined as a text generation task. The model is trained three times (epochs) using 17,618 training samples, resulting in the loss function curve of the e-commerce product name verification model, as shown below. Figure 4a As shown, the loss function values ​​of both the original loss function curve (origina l) and the smoothed loss function curve (smoothed) eventually decreased to below 0.2, satisfying the training stopping condition. Furthermore, upon satisfying the training stopping condition, validation samples were used to validate the e-commerce product name verification model, yielding the following results: Figure 4b The loss function curves shown, both the original loss function curve (or igina l) and the smoothed loss function curve (smoothed), eventually drop below 0.16, indicating that the e-commerce product name verification model has been trained and validated.

[0086] In this embodiment of the application, when the loss function value of the e-commerce product name verification model meets the training stopping condition, the verification sample is used to verify the e-commerce product name verification model. This can further ensure the training reliability of the e-commerce product name verification model, improve the verification accuracy of the e-commerce product name verification model, and enhance the user experience of using the e-commerce product name verification model to verify e-commerce product names.

[0087] Furthermore, if the product name sample in the training sample is a single product name sample, the product name to be reviewed is a single product name sample; if the product name sample in the training sample is a combined product name sample obtained by concatenating with a concatenation character, the product name to be reviewed is a combined product name sample.

[0088] For example, when the e-commerce product name sample in the training sample is a single e-commerce product name (e.g., e-commerce product name s), the e-commerce product name review model trained based on the training sample and the preset e-commerce product name review process has the ability to review a single e-commerce product name at a time. An e-commerce product name (e.g., e-commerce product name t) can be input as the e-commerce product name to be reviewed into the trained e-commerce product name review model. The trained e-commerce product name review model performs the target review steps on e-commerce product name t to obtain the review result of e-commerce product name t. For example, if e-commerce product name t is "men's short sleeve", the review result of e-commerce product name review model outputting e-commerce product name t can be: [S1] Clothing / clothing / appare l; [S2] Applicable people + material + specific product name; [S3] Non-compliant, lacking material elements. In the training samples, the product names are product name 1, product name 2, product name 3... product name r (r product names). When combined using concatenation characters (e.g., \n) to obtain combined product names: product name 1\n product name 2\n product name 3\n...\n product name r, the product name review model trained based on the training samples and a preset product name review process has the ability to distinguish product names by concatenation characters and to review r product names at a time. It can combine r product names (e.g., product name 1p, product name 2p, product name 3p,..., product name rp) using concatenation characters to obtain combined product names (e.g., e-commerce product names). Product Name 1p\nProduct Name 2p\nProduct Name 3p\n...\nProduct Name rp), as the product names to be reviewed, are input into the trained product name review model. The trained product name review model simultaneously performs the target review steps on the r product names in the combination, and outputs the review results of the r product names in the combination in r groups. For example, if the product name to be reviewed input into the trained product name review model is the combination product name: Men's Short-sleeved Shirt\nPiano\nPlump Toy, the review results output by the product name review model include the following 3 groups: The first group is the review result corresponding to "Men's Short-sleeved Shirt" in the product name to be reviewed: [S1] Clothing ing / appare l; [S2] Applicable audience + material + specific product name; [S3] Non-compliant, lacking material elements; Group 2 is the review result corresponding to "piano" in the product name to be reviewed: [S1] Musical instrument; [S2] Specific product name; [S3] Compliant; Group 3 is the review result corresponding to "plush toy" in the product name to be reviewed: [S1] Toy; [S2] Material + specific product name; [S3] Compliant.

[0089] Comparison revealed that, for the daily volume of e-commerce product name verification, manual processing alone requires 8 customs officers and 20 customer service representatives. However, a trained e-commerce product name verification model capable of verifying a single product name per transaction can verify 2 product names per second, replacing all manual processing with an accuracy rate of 97%. Furthermore, a trained e-commerce product name verification model capable of verifying 5 product names per transaction can verify 5 product names per second, not only replacing all manual processing but also significantly improving verification efficiency, while maintaining an accuracy rate of 95%. Additionally, compared to the model with the single-product-name-verification capability, the cost of using the trained model with the 5-product-name-verification capability decreased from 30,000 yuan per month to approximately 5,000 yuan per month. In practical use, after the same e-commerce product name verification model has undergone the training process of verifying a single e-commerce product name and multiple e-commerce product names at a time, and has the ability to verify both a single e-commerce product name and multiple e-commerce product names at a time, users can flexibly set the number of e-commerce product names that the e-commerce product name verification model can verify at a time, taking into account both verification efficiency and cost.

[0090] In this embodiment, a training sample is used, consisting of a single e-commerce product name sample and / or a combined e-commerce product name sample obtained by concatenating product names using concatenation characters. A preset e-commerce product name review process is used to train the e-commerce product name review model. This enables the trained model to review a single e-commerce product name at a time and / or review multiple e-commerce product names at a time, meeting diverse user needs and further improving the model's review efficiency and user experience.

[0091] In some embodiments, before determining the first product name filling specification corresponding to the first product name category based on the first product name category and the correspondence between the product name category and the product name filling specification in step S102, the e-commerce product name review method provided in this application embodiment further includes: determining the correspondence between the product name category to which the e-commerce product name belongs and the product name filling specification.

[0092] like Figure 5 As shown, determining the correspondence between the product name category and the product name filling specifications may include, but is not limited to, steps S401 to S402:

[0093] Step S401: Obtain the product name category library; the product name category library includes a collection of product names for at least one product name category;

[0094] Step S402: Based on the e-commerce product names in the product name set, determine the correspondence between the product name category to which the e-commerce product name belongs and the product name filling specifications;

[0095] The product names in e-commerce include both standardized and non-standardized names. Optionally, based on a product name category database, algorithms can automatically analyze and organize the product name filling specifications corresponding to each product name category. For example, e-commerce product names must have at least one of the following elements: product category, purpose, function, target audience, and specific product name.

[0096] For example, the product name category library includes a set of product names for the toy category and a set of product names for the game category. The standardized product names in the toy category include: Wooden Children's Toys, Plastic Children's Toys, and the non-standard product name is simply "Toys". Comparing the standardized and non-standard product names in the toy category reveals that the standardized product names are formatted as "Material + Specific Product Name". The non-standard product names lack material descriptions compared to the standardized ones. Therefore, it can be determined that the standard format for filling in the e-commerce product name for the toy category is "Material + Specific Product Name". The standard names in the game category include: Puzzle Games, Casual Games, and the non-standard name is: Games. Similarly, comparing the standard and non-standard names of the game category, we can see that the standard name format for the game category is "specific purpose + specific name". The non-standard name for the game category lacks a specific purpose description compared to the standard name for the toy category. Therefore, it can be determined that the standard name format for e-commerce games is "specific purpose + specific name".

[0097] Steps S401 to S402 shown in this embodiment determine the product name filling specifications for each product name category by using the standardized and non-standard product names in the product name category library. This obtains the correspondence between the product name category to which the product name belongs and the product name filling specifications, ensuring the accuracy of the correspondence between the product name category to which the product name belongs and the product name filling specifications. It also improves the accuracy of training the product name review model based on the correspondence between the product name category to which the product name belongs and the product name filling specifications, and improves the accuracy of reviewing product names using the trained product name review model.

[0098] Following step S401 in some embodiments, the e-commerce product name verification method provided in this application further includes: updating the product name category library. For example... Figure 6 As shown, updating the product name category library may include, but is not limited to, steps S501 to S504:

[0099] Step S501: Select the target e-commerce product name from the e-commerce product name database; wherein, the target e-commerce product name includes standardized product names and non-standardized product names;

[0100] Step S502: Match the target e-commerce product name with the product name categories in the product name category library to determine whether there is a product name category in the product name category library that corresponds to the target e-commerce product name;

[0101] If a product category corresponding to the target e-commerce product name exists in the product category library, then step S503 is executed: add the target e-commerce product name to the product name set of the corresponding product category in the product category library;

[0102] If there is no product category in the product category library that corresponds to the target e-commerce product name, then proceed to step S504: create a corresponding product category in the product category library and add the target e-commerce product name to the product name set of the corresponding product category in the product category library.

[0103] For example, a target e-commerce product name, such as product name A, is arbitrarily selected from the product name category library. Then, product name A is matched with the product name categories in the product name category library to determine whether there is a product name category in the product name category library that is semantically similar to product name A. If there is a product name category B in the product name category library that is semantically similar to product name A, then product name A is added to the product name set of product name category B in the product name category library. If there is no product name category in the product name category library that is semantically similar to product name A, then a new product name category C that is semantically similar to product name A is created in the product name category library, and product name A is added to the product name set of product name category C in the product name category library.

[0104] Steps S501 to S504 in this embodiment of the application, based on the matching result between the target e-commerce product name in the e-commerce product name database and the product name category in the product name category database, perform the operation of adding the target e-commerce product name to the product name set of the corresponding product name category in the product name category database, or creating a new corresponding product name category in the product name category database and adding the target e-commerce product name to the product name set of the corresponding product name category in the product name category database. This ensures the comprehensiveness and accuracy of the e-commerce product names in the product name categories in the product name category database and the product name sets of the product name categories. In turn, it can improve the accuracy of the correspondence between the product name category to which the e-commerce product name belongs and the product name filling specification determined based on the product name category database, as well as the accuracy of training the e-commerce product name review model based on the correspondence between the product name category to which the e-commerce product name belongs and the product name filling specification, and the accuracy of reviewing e-commerce product names using the trained e-commerce product name review model.

[0105] Figure 7This is a flowchart illustrating the e-commerce product name verification method provided in this application embodiment. The e-commerce product name verification method provided in this application embodiment may include, but is not limited to, steps S601 to S605.

[0106] Step S601: Obtain the product name of the e-commerce item to be reviewed;

[0107] Step S602: Match the product name to be reviewed with the historical review information in the historical review database to determine whether there is target historical review information in the historical review database that matches the product name to be reviewed;

[0108] If there is target historical review information in the historical review database that matches the product name to be reviewed, then proceed to step S603: determine the review result of the product name to be reviewed based on the target historical review information;

[0109] If there is no target historical review information matching the product name to be reviewed in the historical review database, then step S604 is executed: the target review step is executed on the product name to be reviewed to obtain the review result of the product name to be reviewed.

[0110] In the embodiments of this application, steps S601 to S604, when there is target historical review information matching the product name to be reviewed in the historical review database, directly determine the review result of the product name to be reviewed based on the target historical review information. When there is no target historical review information matching the product name to be reviewed in the historical review database, the review result of the product name to be reviewed is determined by the trained product name review model. This can efficiently reuse historical review information, greatly reduce duplicate review work, and further improve the review efficiency of product names while ensuring the accuracy of product name review.

[0111] Continue reading Figure 7 After step S604 in some embodiments, the e-commerce product name verification method provided in this application embodiment may also include, but is not limited to, the following steps:

[0112] Step S605: Add the product name to be reviewed and the review result of the product name to be reviewed as target historical review information to the historical review database.

[0113] Step S605 in this embodiment adds the product name to be reviewed and the review result of the product name to be reviewed obtained by performing the target review step on the product name to be reviewed as target historical review information to the historical review database. This achieves efficient storage of historical review information and dynamic updating of historical review information in the historical review database, which can ensure the accuracy and comprehensiveness of historical review information in the historical review database, and thus ensure the accuracy of reviewing product names based on historical review information in the historical review database.

[0114] In one application scenario, taking the e-commerce product name "leather shoes" as an example, the first category of the product name "leather shoes" is "shoes". According to the correspondence between product category and product name filling specifications, the product name filling specification for the first category "shoes" is "material + specific product name". Specifically, the product name "leather shoes" is matched with historical review information in the historical review database to determine if there is any matching historical review information. If there is, the target historical review information is directly extracted from the historical review information to obtain the review result for the product name "leather shoes". If there is no matching historical review information, the following target review steps are performed: obtain the first category of the product name "leather shoes", and according to the first category of the product name "leather shoes", the first category is "shoes". The process involves identifying the product category and its correspondence with the product name filling specifications, determining the first product name filling specifications corresponding to the first product name category to which the product name "leather shoes" belongs, and comparing the product name "leather shoes" with the first product name filling specifications corresponding to the first product name category to which it belongs. The final review result for the product name "leather shoes" is determined as follows: [S1] Shoes; [S2] Material + Specific Product Name; [S3] Qualified. Then, the product name "leather shoes" and its review result are added to the historical review database as target historical review information, so that the review result can be directly extracted from the historical review database later.

[0115] Please see Figure 8 This application also provides an e-commerce product name verification device, which can implement the above-mentioned e-commerce product name verification method. The device includes an acquisition module and a verification module.

[0116] The acquisition module is used to obtain the names of e-commerce products to be reviewed;

[0117] The review module is used to perform the target review steps on the product names to be reviewed and obtain the review results of the product names to be reviewed. The review results include: whether the product name to be reviewed is compliant, or whether the product name to be reviewed is non-compliant and the reason for non-compliance.

[0118] The target review steps include: obtaining the first product name category to which the product name to be reviewed belongs; determining the first product name filling specification corresponding to the first product name category based on the first product name category and the correspondence between product name categories and product name filling specifications; comparing the product name to be reviewed with the first product name filling specification to obtain the review result of the product name to be reviewed.

[0119] The specific implementation of this e-commerce product name verification device is basically the same as the specific implementation of the above-mentioned e-commerce product name verification method, and will not be described again here.

[0120] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned product name verification method. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0121] Please see Figure 9 , Figure 9 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes:

[0122] The processor 901 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 solutions provided in the embodiments of this application.

[0123] The memory 902 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 902 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 902 and is called and executed by the processor 901 using the e-commerce product name verification method of the embodiments of this application.

[0124] The input / output interface 903 is used to implement information input and output;

[0125] The communication interface 904 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0126] Bus 905 transmits information between various components of the device (e.g., processor 901, memory 902, input / output interface 903, and communication interface 904);

[0127] The processor 901, memory 902, input / output interface 903, and communication interface 904 are connected to each other within the device via bus 905.

[0128] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described e-commerce product name verification method.

[0129] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0130] The e-commerce product name verification method, device, electronic device, and storage medium provided in this application embodiment improve the efficiency and accuracy of e-commerce product name verification. The method, device, and storage medium allow users to input the product name to be verified into a trained e-commerce product name verification model and execute the following verification steps: obtaining the first product name category to which the product name belongs; determining the first product name filling specification corresponding to the first product name category based on the first product name category and the correspondence between product name categories and product name filling specifications; comparing the product name to be verified with the first product name filling specification to obtain the verification result; and finally outputting the verification result indicating whether the product name is compliant or non-compliant, along with the reasons for non-compliance. This improves the efficiency and accuracy of e-commerce product name verification. The trained e-commerce product name verification model has the ability to verify individual product names and / or combinations of product names. Users can flexibly set the number of product names that the model can verify at one time, meeting diverse user needs. When a target historical review information matching the product name to be reviewed exists in the historical review database, the review result of the product name can be directly determined based on the target historical review information. This allows for efficient reuse of historical review information, greatly reducing redundant review work and further improving the efficiency of product name review while ensuring accuracy. Furthermore, after inputting the product name to be reviewed into the trained product name review model and obtaining the review result, the product name and its review result are added to the historical review database as target historical review information. This achieves efficient storage of historical review information, ensuring its accuracy and comprehensiveness, and consequently guaranteeing the accuracy of product name reviews based on historical review information in the database.

[0131] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0132] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0133] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; 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.

[0134] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0135] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0136] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0137] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0138] The units described above 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 units can be selected to achieve the purpose of this embodiment according to actual needs.

[0139] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0140] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0141] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for verifying e-commerce product names, characterized in that, include: Obtain the product name to be reviewed; The target review steps are performed on the e-commerce product name to be reviewed to obtain the review result of the e-commerce product name to be reviewed. The review result includes: the e-commerce product name to be reviewed is compliant, or the e-commerce product name to be reviewed is non-compliant and the reason for non-compliance. The target review steps include: Obtain the first product category to which the product name to be reviewed belongs; Based on the first product name category and the correspondence between product name categories and product name filling specifications, determine the first product name filling specifications corresponding to the first product name category; The product name to be reviewed is compared with the first product name filling specification to obtain the review result of the product name to be reviewed.

2. The e-commerce product name verification method as described in claim 1, characterized in that, Before performing the target review step on the product name to be reviewed and obtaining the review result of the product name to be reviewed, the method further includes: The product names to be reviewed are matched with historical review information in the historical review database; If there is no target historical review information matching the product name to be reviewed in the historical review database, then the step of performing target review on the product name to be reviewed and obtaining the review result of the product name to be reviewed is executed. After performing the target review step on the product name to be reviewed and obtaining the review result of the product name to be reviewed, the method further includes: The product name to be reviewed and the review result of the product name to be reviewed are added to the historical review database as the target historical review information.

3. The e-commerce product name verification method as described in claim 2, characterized in that, After matching the product name to be reviewed with historical review information in the historical review database, the method further includes: If the historical review database contains target historical review information that matches the product name to be reviewed, the review result of the product name to be reviewed is determined based on the target historical review information.

4. The e-commerce product name verification method as described in claim 1, characterized in that, The step of performing the target review process on the product name to be reviewed, and obtaining the review result of the product name to be reviewed, includes: The product name to be reviewed is input into the trained product name review model. The target review steps are then performed on the product name to be reviewed by the product name review model to obtain the review result of the product name to be reviewed. Before inputting the product name to be reviewed into the trained product name review model, and performing the target review step on the product name to be reviewed through the product name review model to obtain the review result of the product name to be reviewed, the method further includes: Obtain a training dataset, wherein the training dataset includes multiple training samples, and each training sample includes an e-commerce product name sample and the tag review result corresponding to the e-commerce product name sample; Using the training dataset and the preset e-commerce product name verification process, the preset e-commerce product name verification model is trained until the training stop condition is met, and the trained e-commerce product name verification model is obtained. The preset e-commerce product name review process includes: obtaining the second product name category to which the e-commerce product name sample belongs; determining the second product name filling specification corresponding to the second product name category based on the second product name category and the correspondence between the product name category and the product name filling specification; comparing the e-commerce product name sample with the second product name filling specification to obtain the predicted review result of the e-commerce product name sample.

5. The e-commerce product name verification method as described in claim 4, characterized in that, The step of training the e-commerce product name verification model using the training dataset until the training stopping condition is met, to obtain the fully trained e-commerce product name verification model, includes: For each training sample, perform the following steps: The training samples and the preset e-commerce product name review process are input into the e-commerce product name review model. The e-commerce product name review model reviews the training samples based on the preset e-commerce product name review process to obtain the predicted review result of the training samples. Based on the predicted review results and the label review results, determine the loss function value of the e-commerce product name review model; If the loss function value does not meet the training stopping condition, the model parameters of the e-commerce product name verification model are adjusted to obtain an updated e-commerce product name verification model. The updated e-commerce product name verification model is then trained using the next training sample until the training stopping condition is met, resulting in a fully trained e-commerce product name verification model.

6. The e-commerce product name verification method as described in claim 5, characterized in that, If the product name sample in the training sample is a single product name, then the product name to be reviewed is a single product name. If the product name sample in the training sample is a combined product name obtained by concatenating characters, then the product name to be reviewed is a combined product name.

7. The e-commerce product name verification method as described in claim 1, characterized in that, Before determining the first product name filling specification corresponding to the first product name category based on the first product name category and the correspondence between product name categories and product name filling specifications, the method further includes: Obtain the product name category library; the product name category library includes a set of product names for at least one product name category; Based on the e-commerce product names in the product name set, determine the correspondence between the product name category to which the e-commerce product name belongs and the product name filling specifications; The product names mentioned include both standardized and non-standardized product names.

8. The e-commerce product name verification method as described in claim 7, characterized in that, After obtaining the product name category library, the method further includes: Select the target e-commerce product name from the e-commerce product name database; Match the target e-commerce product name with the product name categories in the product name category library; If a product name category exists in the product name category library that corresponds to the target e-commerce product name, then the target e-commerce product name is added to the product name set of the corresponding product name category in the product name category library; If there is no product name category corresponding to the target e-commerce product name in the product name category library, then a corresponding product name category is created in the product name category library, and the target e-commerce product name is added to the product name set of the corresponding product name category in the product name category library; The target e-commerce product name includes both standardized and non-standardized product names.

9. An e-commerce product name verification device, characterized in that, include: Acquisition module and approval module; The acquisition module is used to acquire the names of the e-commerce products to be reviewed; The review module is used to perform target review steps on the e-commerce product name to be reviewed, and obtain the review result of the e-commerce product name to be reviewed, wherein the review result indicates: the e-commerce product name to be reviewed is compliant, or the e-commerce product name to be reviewed is non-compliant and the reason for non-compliance; The target review steps include: Obtain the first product category to which the product name to be reviewed belongs; Based on the first product name category and the correspondence between product name categories and product name filling specifications, determine the first product name filling specifications corresponding to the first product name category; The product name to be reviewed is compared with the first product name filling specification to obtain the review result of the product name to be reviewed.

10. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the e-commerce product name verification method as described in any one of claims 1-8.

11. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the e-commerce product name verification method as described in any one of claims 1-8.

Images