Goods and commodity matching method, equipment, medium and product

Abstract

The invention discloses a goods and commodity matching method and device, a medium and a product, and relates to the field of data processing, and the method comprises the steps: obtaining the first commodity data of a first commodity, the first commodity being any goods of a brand party sold by an e-commerce platform; determining goods data having the highest similarity with the first goods data from the goods data; each piece of goods data is data information of each piece of goods stored by a brand party; and determining the goods corresponding to the goods data with the highest similarity, recording the goods as the goods with the highest similarity, and determining the first goods as the goods matched with the goods with the highest similarity. According to the invention, the timeliness and accuracy of matching between the commodity and the goods are improved.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method, apparatus, medium and product for matching goods and commodities. Background Technology

[0002] A single brand / company may sell its products on multiple e-commerce platforms (JD.com, Taobao, Douyin, etc.), requiring the brand to analyze the sales performance of its products across all channels from a holistic perspective. However, for the same product, its product ID, product name, and product price differ on each platform, making it difficult for the brand to effectively match products with their customers during data integration.

[0003] The existing product and commodity matching mechanism is as follows: For each e-commerce platform, a mapping relationship table between the e-commerce platform's product ID and the brand's product ID is manually created. Then, this relationship table is used to match the e-commerce platform's product ID with the brand's product ID, thereby achieving product matching between different e-commerce platforms.

[0004] However, in reality, a brand usually has a dozen or even dozens of stores on major e-commerce platforms. Each store requires operators to manually maintain the correspondence between platform product IDs and item IDs, which consumes a lot of manpower and resources. Moreover, the skill levels of different operators vary, and there are often inaccurate, missing, or incorrect information, which leads to a low accuracy rate of matching products with goods. Summary of the Invention

[0005] The purpose of this application is to provide a method, device, medium, and product for matching goods and commodities, in order to solve the problems of low accuracy and high manpower and material resources required for matching goods and commodities as described in the prior art.

[0006] To achieve the above objectives, this application provides the following solution: Firstly, this application provides a method for matching goods and commodities, including: Obtain the first product data of the first product, where the first product is any item from a brand sold on the e-commerce platform; The product data with the highest similarity to the first product data is determined from the product data; the product data includes the data information of each product stored by the brand owner; The product corresponding to the product data with the highest similarity is identified and denoted as the product with the highest similarity. The first product is then identified as the product matched by the product with the highest similarity.

[0007] Optionally, the step of "determining the product data with the highest similarity to the first product data from the product data" includes... Obtain data for each item; Determine the similarity between the first product data and the data of each item to obtain multiple similarity scores; The maximum similarity is determined from multiple similarity values, and the product data corresponding to the maximum similarity is determined as the product data with the highest similarity.

[0008] Optionally, the product data includes product name and product image, and the goods data includes goods name and goods image; The phrase "obtaining the first product data of the first product" includes: obtaining the first product name and the first product image of the first product; The phrase "obtaining product data for each product" includes: obtaining the first product data for the first product, which includes the name and image of the first product, and the first product can be any product owned by the brand owner; Methods for determining the similarity between the first commodity data and the first product data include: Determine the name similarity between the first product name and the first item name; Determine the image similarity between the first product image and the first goods image; Based on the name similarity and the image similarity, the similarity between the first product data and the first goods data is determined.

[0009] Optionally, the product data further includes product price and product specifications, and the goods data further includes goods price and goods specifications; after "determining the image similarity between the first product image and the first goods image", the method further includes: Determine the price similarity between the price of the first commodity and the price of the first item; Determine the similarity between the specifications of the first commodity and the specifications of the first goods; The similarity between the first product data and the first goods data is determined based on the name similarity, image similarity, price similarity, and specification similarity.

[0010] Optionally, "determining the name similarity between the first product name and the first goods name" includes: Extract the semantic vector of the first product name; Extract the semantic vector of the first product name; Determine the semantic vector similarity between the semantic vector of the product and the semantic vector of the goods, and record the semantic vector similarity as the name similarity.

[0011] Optionally, "determining the image similarity between the first product image and the first goods image" includes: Extract the feature vector from the first product image, and denote it as the product feature vector; Extract the feature vector from the first product image, and denote it as the product feature vector; Determine the feature vector similarity between the product feature vector and the goods feature vector, and record the feature vector similarity as the image similarity.

[0012] Optionally, the semantic vector similarity is cosine similarity, and the feature vector similarity is cosine similarity; The step of “determining the semantic vector similarity between the product semantic vector and the goods semantic vector, and recording the semantic vector similarity as the name similarity” includes: determining the cosine similarity between the product semantic vector and the goods semantic vector, and recording the cosine similarity as the name similarity; The step of “determining the feature vector similarity between the product feature vector and the goods feature vector, and recording the feature vector similarity as the image similarity” includes: determining the cosine similarity between the product feature vector and the goods feature vector, and recording the cosine similarity as the image similarity.

[0013] In a second aspect, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method described in any one of the first aspects above.

[0014] Thirdly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any one of the first aspects above.

[0015] Fourthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the method described in any one of the first aspects above.

[0016] The product-product matching method provided in this application identifies the product data with the highest similarity to each product data; determines the product with the highest similarity corresponding to the product data with the highest similarity; and identifies the product as the product corresponding to the product with the highest similarity, thereby establishing a product-product correspondence. In contrast, existing technologies rely on manual maintenance of the product-product correspondence. Therefore, compared to existing technologies, this application significantly reduces personnel maintenance costs while greatly improving the timeliness and accuracy of product-product matching, providing precise data support for brands to analyze multi-product sales from an omnichannel perspective. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a diagram illustrating the application environment of the goods-commodity matching method in one embodiment of this application; Figure 2 A flowchart illustrating a method for matching goods and commodities according to an embodiment of this application; Figure 3 A flowchart illustrating a method for determining the most similar product data according to an embodiment of this application; Figure 4 and Figure 5 A flowchart illustrating a method for determining the similarity between commodity data and goods data, provided in an embodiment of this application; Figure 6 This is a flowchart illustrating a method for determining the similarity between commodity data and goods data, provided in an embodiment of this application. Figure 7 and Figure 8 A flowchart illustrating a method for determining name similarity according to an embodiment of this application; Figure 9 A flowchart illustrating a name preprocessing method for product names and goods names provided in this application embodiment; Figure 10 and Figure 11 A flowchart illustrating an image similarity determination method provided in an embodiment of this application; Figure 12 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0019] 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 a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0020] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the contents of this application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0021] This application provides an application environment for a goods-commodity matching method. See [link to relevant documentation]. Figure 1 The application environment includes an e-commerce platform database, a brand platform database, a terminal, and a brand server. The terminal communicates with the brand server via a network. A data storage system stores the data that the brand server needs to process. This data storage system can be set up independently, integrated into the brand server, or located in the cloud or on another brand server. The terminal can send data to be processed to the brand server. Upon receiving the data, the brand server can store it and retrieve it from the storage location when processing is needed, or it can perform processing tasks while storing the data. The brand server can provide the terminal with the obtained data tags for the data. Furthermore, in some embodiments, the product-to-goods matching method can be implemented independently by the brand server or the terminal. For example, the terminal can directly perform data tagging on the data to be processed, or the brand server can retrieve the data to be processed from the data storage system and perform data tagging on it.

[0022] The e-commerce platform database stores data related to the products sold on the platform, such as product identification codes (product IDs), product brands, product prices, and product specifications. The brand platform database stores data related to its own products, such as product IDs, product brands, product prices, and product specifications.

[0023] The terminals can be, but are not limited to, various desktop computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, and smart in-vehicle devices. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted devices. The brand server can be implemented using a dedicated brand server, a brand server cluster consisting of multiple brand servers, or a cloud brand server.

[0024] In one exemplary embodiment, see Figure 2 As shown, a method for matching goods and products is provided. This method is executed by a computer device, specifically by a terminal or a brand server, or by both a terminal and a brand server. In this embodiment, the method is applied to... Figure 1 Taking the brand's server as an example, the explanation includes the following steps 101 to 103: Step 101: Obtain the first product data of the first product, where the first product is any item from a brand sold on the e-commerce platform. In this context, "e-commerce platform" refers to any e-commerce platform that sells the brand's goods. Therefore, the first product is any product from any brand that is sold on any e-commerce platform.

[0025] The first product data is stored by the e-commerce platform, and the brand owner has the right and ability to access it. Of course, the brand owner can also obtain the data from the e-commerce platform in advance, store it locally, and update it regularly so that it can be retrieved locally when needed.

[0026] The first product data includes data information of the first product stored on the e-commerce platform, used to reflect the characteristics of the first product. The first product data includes the first product name and the first product image.

[0027] Furthermore, both the e-commerce platform's data storage system and the brand's platform's data storage system are equipped with API interfaces, enabling the brand's server to exchange data with these two storage systems.

[0028] The brand owns multiple products and authorizes different e-commerce platforms to sell them according to demand.

[0029] In this application, for ease of description, the goods owned by the brand owner are referred to as "goods" on the brand owner's platform and as "products" on the e-commerce platform. Each product and each item is assigned an ID number for identification.

[0030] Step 102: Identify the product data with the highest similarity to the first product data from the product data; the product data includes the data information of each product stored by the brand owner; The data for each product is stored in the brand's database and can be retrieved promptly when needed.

[0031] The product data includes product names and product images.

[0032] Step 103: Determine the product corresponding to the product data with the highest similarity, and denot it as the product with the highest similarity. Then, determine the first product as the product matched by the product with the highest similarity.

[0033] The above method can be used to establish a correspondence between the most similar product and the first product.

[0034] By using the method described above to identify all the corresponding products for each item, the pairing relationships between each item can be established. Combined with sales data, the sales performance of each item on different platforms can be analyzed. Based on this sales data, corresponding decisions can be made; for example, for platforms with high profit margins or large sales volumes, the supply can be increased to obtain more profits.

[0035] The brand in this application is used as an example, but this method can be used for any brand.

[0036] A brand can allow multiple e-commerce platforms to sell its products. For example, these platforms could be JD.com, Taobao, Douyin, etc.

[0037] The product-product matching method provided in this application identifies the product data with the highest similarity to each product data; determines the product with the highest similarity corresponding to the product data with the highest similarity; and identifies the product as the product corresponding to the product with the highest similarity, thereby establishing a product-product correspondence. In contrast, existing technologies rely on manual maintenance of the product-product correspondence. Therefore, compared to existing technologies, this application significantly reduces personnel maintenance costs while greatly improving the timeliness and accuracy of product-product matching, providing precise data support for brands to analyze multi-product sales from an omnichannel perspective.

[0038] For example, this matching can determine which e-commerce platform sells the most products and profits the most. It can also identify whether an e-commerce platform is selling unauthorized products. For instance, if a product not authorized by the brand is matched with other products, it indicates that the product is not authorized for sale by the brand.

[0039] In addition, this method is used to match goods and products across different e-commerce platforms, so that the matching of products across different platforms will not have a large deviation, which will also affect business decisions.

[0040] Furthermore, it does not require a large amount of human resources or have high maintenance costs. The above method can be executed once at predetermined intervals, such as every hour. Compared with manual maintenance, the data in this application is updated more promptly, reducing data latency.

[0041] Optionally, see Figure 3 In another exemplary embodiment of this application, step 102 includes steps 201-203: Step 201: Obtain the data for each item; Step 202: Determine the similarity between the first product data and each product data to obtain multiple similarity scores; Further, step 202 can specifically be: for each product data, calculate the similarity between the first product data and the product data, thereby obtaining multiple similarity scores.

[0042] Further, step 202 can be: determining the similarity between the first product name and the first product image and the product names and product images of each item, and obtaining multiple similarity scores.

[0043] Step 203: Determine the maximum similarity from multiple similarities, and determine the product data corresponding to the maximum similarity as the product data with the highest similarity.

[0044] Optionally, see Figure 4 and Figure 5 In another exemplary embodiment of this application, the product data includes product name and product image, and the goods data includes goods name and goods image; Step 101 includes: obtaining the first product name and the first product image of the first product; Step 201 includes: obtaining first product data of the first product, the first product data including the name of the first product and the image of the first product, the first product being any product owned by the brand owner; The similarity between the first commodity data and the first goods data is determined through the following steps 301 to 303: Step 301: Determine the name similarity between the first product name and the first goods name; Step 302: Determine the image similarity between the first product image and the first goods image; Step 303: Based on the name similarity and the image similarity, determine the similarity between the first product data and the first goods data.

[0045] Further, step 303 can be: determining the name weight for name similarity, determining the image weight for image similarity, and the sum of the name weight and the image weight is 1; calculating the first product of the name similarity and the name weight, and calculating the second product of the image similarity and the image weight; calculating the sum of the first product and the second product to obtain the similarity between the first product data and the first commodity data.

[0046] in addition, Figure 5 The open platform in this context refers to one way of acquiring data, which also includes other methods. The platform will expose external interfaces for data distribution.

[0047] Optionally, see Figure 6 In another exemplary embodiment of this application, the commodity data further includes commodity price and commodity specifications, and the product data further includes product price and product specifications; after step 302, the method further includes the following steps 401 to 403: Step 401: Determine the price similarity between the price of the first commodity and the price of the first product; Step 402: Determine the similarity between the specifications of the first commodity and the specifications of the first goods; Step 403: Based on the name similarity, image similarity, price similarity, and specification similarity, determine the similarity between the first product data and the first goods data.

[0048] Determine the name weight for name similarity, the image weight for image similarity, the price weight for price similarity, and the specification weight for specification similarity. The sum of the name weight, image weight, price weight, and specification weight is 1. Calculate the first product of the name similarity and name weight, the second product of the image similarity and image weight, the third product of the price similarity and price weight, and the fourth product of the specification similarity and specification weight. Calculate the sum of the first, second, third, and fourth products to obtain the similarity between the first product data and the first goods data.

[0049] Among them, the weight of name, image, price, and specifications can be dynamically adjusted according to the actual situation, and can be empirical values.

[0050] Optionally, see Figure 7 and Figure 8 In another exemplary embodiment of this application, step 301 includes steps 501 to 503: Step 501: Extract the product semantic vector of the first product name; Step 502: Extract the semantic vector of the first product name; Optionally, see Figure 9 Before extracting the semantic vectors of goods and products, the names of goods and products can be preprocessed.

[0051] The preprocessing includes removing marketing terms, standardizing unit specifications, and replacing brand synonyms.

[0052] Remove marketing keywords: Eliminate distracting words such as "free shipping" and "special offer"; Standardize the unit of measurement: convert "500g" to "0.5kg" to resolve numerical discrepancies; Brand synonym replacement: Build a dictionary (e.g., "NIKE→Nike", "Apple→Apple") to achieve synonym replacement.

[0053] The above preprocessing rules can be dynamically configured according to the specific circumstances of the brand / industry, and the system will process them automatically after configuration.

[0054] Alternatively, the paraphrase-multilingual-MiniLM-L12-v2 model can be used to extract semantic vectors from product names and item names: Furthermore, the model can be fine-tuned based on brand industry data to strengthen the vector alignment of category keywords (such as "Snapdragon 8 Gen3" and "Snapdragon 8.3").

[0055] Step 503: Determine the semantic vector similarity between the semantic vector of the product and the semantic vector of the goods, and record the semantic vector similarity as the name similarity.

[0056] Optionally, see Figure 10 and Figure 11 In another exemplary embodiment of this application, step 302 includes steps 601 to 603: Step 601: Extract the feature vector of the first product image, denoted as the product feature vector; Step 602: Extract the feature vector of the first product image, denoted as the product feature vector; Optionally, before extracting the feature vectors of the image, the image can be preprocessed, including: I. Size Normalization Target size: uniformly scaled to the CNN input requirement (224×224), using anisotropic scaling (direct stretching) or isotropic scaling (filling the background).

[0057] Filling strategy: Fill the extended area with the average pixel value of the image edge to reduce distortion (highest accuracy when padding=16).

[0058] II. Color and Enhancement Channel conversion: RGB three channels are separated and mean normalization is performed (ImageNet mean [0.485, 0.486, 0.406], standard deviation [0.229, 0.224, 0.225]).

[0059] Anti-interference processing: Remove non-subject elements such as watermarks and logos, and use OpenCV's inpaint algorithm to repair the background. Optionally, a ResNet50 model can be used for image feature extraction. This model can be fine-tuned to fit the brand's product domain using ImageNet pre-trained weights, depending on the specific brand.

[0060] Step 603: Determine the feature vector similarity between the product feature vector and the goods feature vector, and record the feature vector similarity as the image similarity.

[0061] Optionally, before calculating the similarity of feature vectors, PCA (Principal Component Analysis) can be used to achieve unsupervised linear dimensionality reduction of feature vectors, aiming to transform high-dimensional data into a low-dimensional representation while retaining the most critical variation information in the data. Its core principle is as follows: First, the maximum variance theory (maximum separability) aims to find the direction of maximum variance after data projection, ensuring that the data is as dispersed as possible after dimensionality reduction.

[0062] First principal component: The direction with the largest variance in the data projection.

[0063] Second principal component: the direction orthogonal to the first principal component and with the second largest variance, and so on. Mathematical expression: Solving the problem through optimization:

[0064] in, w Projection direction; X i : is the first i A sample data vector (usually a column vector); m : Represents the number of samples (i.e., the number of data points); ∑: Summation symbol, indicating summation over all samples; st The dot (.) is an abbreviation for "subject to," meaning "constrained by" or "subject to conditions." It is not an independent variable, but rather a standard notation for introducing constraints in optimization problems. Here, it indicates that the optimization problem needs to satisfy the constraint condition || w ||=1; Constraints || w ||=1: This requires that the projection direction w is a unit vector (i.e., the Euclidean norm is 1). This ensures that the optimization problem has a unique solution and avoids scaling issues.

[0065] Second, diagonalization of the covariance matrix. step: 1. Data decentralization (subtract the mean from each column).

[0066] 2. Calculate the covariance matrix C = (X... T X) / m, (X is a decentralized data matrix).

[0067] 3. Eigenvalue decomposition of C: C = V⋀V T eigenvalue λ i Sort from largest to smallest, corresponding to the feature vector V i": That is, the direction of the principal component.

[0068] Key properties: The covariance among principal components is 0 (orthogonality), and the variance of the principal components is equal to the eigenvalue λ. i .

[0069] Furthermore, before calculating image similarity, L2 normalization aims to adjust the vectors to unit length (L2 norm of 1), thereby eliminating the impact of feature scale differences on the model. The core principles and formulas are as follows: L2 norm definition: For a vector x = ( x 1, x 2, ..., x n), whose L2 norm is the square root of the sum of squares of its elements:

[0070] For example, the L2 norm of the vector [3, 4] is

[0071] Normalization operation: Divide the vector by its L2 norm to obtain a new vector.

[0072] After normalization, the vector length is always 1 (e.g., [3, 4 → [0.6, 0.8]), and the direction remains unchanged.

[0073] Optionally, in another exemplary embodiment of this application, the semantic vector similarity is cosine similarity, and the feature vector similarity is cosine similarity; Step 503 includes: determining the cosine similarity between the product semantic vector and the goods semantic vector, and recording the cosine similarity as the name similarity.

[0074] Step 603 includes: determining the cosine similarity between the product feature vector and the goods feature vector, and recording the cosine similarity as the image similarity.

[0075] The formula for calculating cosine similarity is: cos(A,B) = A·B / (‖A‖‖B‖) Here, A and B are two vectors whose similarity is to be calculated, and the result is a value between 0 and 1.

[0076] When the image is a name, A and B are the semantic vector of the product and the semantic vector of the goods, respectively; when the image is an image, A and B are the feature vector of the product and the feature vector of the goods, respectively.

[0077] This application designs and implements a method for matching products with branded goods across multiple e-commerce platforms. This method allows for the automatic matching of branded goods information based on product name, product image, product price, and product specifications when product information is imported from various e-commerce platforms.

[0078] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram can be found in [reference needed]. Figure 12 As shown, this computer device includes a processor, memory, input / output interfaces (I / O), and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores data related to goods-to-product matching methods. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it can implement a goods-to-product matching method.

[0079] Those skilled in the art will understand, see Figure 12 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0080] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0081] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0082] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0083] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data that have been agreed to by the user or have been fully agreed to by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0084] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. In the embodiments provided in this application, any reference to memory, database, or other media can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0085] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic units, data processing logic units, etc., and are not limited to these.

[0086] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0087] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for matching goods with commodities, characterized in that, include: Obtain the first product data of the first product, which includes the first product name and the first product image. The first product is any item from a brand sold on the e-commerce platform. The product data with the highest similarity to the first product name and the first product image is determined from the product data; the product data includes the product name and product image of each product stored by the brand owner; The product corresponding to the product data with the highest similarity is identified and denoted as the product with the highest similarity. The first product is then identified as the product matched by the product with the highest similarity.

2. The goods and commodities matching method according to claim 1, characterized in that, The phrase "determining the product data that has the highest similarity to the first product data from the product data" includes: Obtain data for each item; Determine the similarity between the first product name and the first product image and the product names and images of each item, and obtain multiple similarity scores; The maximum similarity is determined from multiple similarity values, and the product data corresponding to the maximum similarity is determined as the product data with the highest similarity.

3. The goods and commodities matching method according to claim 2, characterized in that, The phrase "obtaining the first product data of the first product" includes: obtaining the first product name and the first product image of the first product; The phrase "obtaining product data for each product" includes: obtaining the first product data for the first product, which includes the name and image of the first product, and the first product can be any product owned by the brand owner; Methods for determining the similarity between the first commodity data and the first product data include: Determine the name similarity between the first product name and the first item name; Determine the image similarity between the first product image and the first goods image; Based on the name similarity and the image similarity, the similarity between the first product data and the first goods data is determined.

4. The goods and commodities matching method according to claim 3, characterized in that, The product data also includes product price and product specifications, and the goods data also includes goods price and goods specifications; after "determining the image similarity between the first product image and the first goods image", the method further includes: Determine the price similarity between the price of the first commodity and the price of the first item; Determine the similarity between the specifications of the first commodity and the specifications of the first goods; The similarity between the first product data and the first goods data is determined based on the name similarity, image similarity, price similarity, and specification similarity.

5. The goods and commodities matching method according to claim 3, characterized in that, The "determining the name similarity between the first product name and the first goods name" includes: Extract the semantic vector of the first product name; Extract the semantic vector of the first product name; Determine the semantic vector similarity between the semantic vector of the product and the semantic vector of the goods, and record the semantic vector similarity as the name similarity.

6. The goods and commodities matching method according to claim 5, characterized in that, The "determining the image similarity between the first product image and the first goods image" includes: Extract the feature vector from the first product image, and denote it as the product feature vector; Extract the feature vector from the first product image, and denote it as the product feature vector; Determine the feature vector similarity between the product feature vector and the goods feature vector, and record the feature vector similarity as the image similarity.

7. The goods and commodities matching method according to claim 6, characterized in that, The semantic vector similarity is cosine similarity, and the feature vector similarity is cosine similarity; The step of "determining the semantic vector similarity between the product semantic vector and the goods semantic vector, and recording the semantic vector similarity as the name similarity" includes: determining the cosine similarity between the product semantic vector and the goods semantic vector, and recording the cosine similarity as the name similarity; The step of "determining the feature vector similarity between the product feature vector and the goods feature vector, and recording the feature vector similarity as the image similarity" includes: determining the cosine similarity between the product feature vector and the goods feature vector, and recording the cosine similarity as the image similarity.

8. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the goods-commodity matching method according to any one of claims 1-7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the goods and commodities matching method according to any one of claims 1-7.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the goods and commodities matching method according to any one of claims 1-7.

Images