Method and device for identifying same commodity, server and storage medium

By performing word segmentation processing and feature matching calculation on product information, the difficulty of identifying the same product caused by inconsistent product classification on e-commerce platforms is solved, and the rapid and accurate identification of the same product is achieved, and the procurement management efficiency is improved.

CN119067749APending Publication Date: 2024-12-03SICHUAN JIUHE DIGITAL TECHNOLOGY CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411120852.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-15
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

On multiple e-commerce platforms, the same product may be classified under different categories due to different classification methods, making it difficult for users to quickly determine whether the product is the same product, causing difficulties in purchasing management.

Method used

By obtaining the product information of the product to be identified, the word segmentation process is determined, the phrase segmentation group to be matched, the feature information of the product feature library has been matched, the matching degree and group similarity are calculated, and if the matching degree and similarity reach the preset threshold, the product is determined For the same product.

Benefits of technology

It realizes the rapid and accurate determination of the same product to be identified based on the matching product feature library, and improves the efficiency and accuracy of product procurement management.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119067749A_ABST
    Figure CN119067749A_ABST
Patent Text Reader

Abstract

The invention is suitable for the technical field of commodity classification, and provides a same commodity identification method and device, a server and a storage medium, and the method comprises the steps: obtaining the commodity information of a to-be-identified commodity, carrying out the word segmentation of the commodity information, and determining a to-be-matched word segmentation group corresponding to the commodity information after the word segmentation; matching the to-be-matched keyword group with the feature phrases of the feature information of the matched commodity feature library, and determining the matching degree of the to-be-identified commodity and the feature information of the matched commodity feature library; the feature information with the corresponding matching degree larger than or equal to a preset matching degree threshold value serves as target feature information, and the group similarity between the feature phrases of the target feature information and the to-be-matched segmented phrases of the to-be-recognized commodity is determined; and sorting the plurality of group similarities, determining the highest group similarity, and if the highest group similarity is greater than or equal to a preset group similarity threshold, determining that the matched commodity corresponding to the highest group similarity is the same commodity as the to-be-identified commodity.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application belongs to the technical field of commodity classification, and in particular, relates to a method, device, server and storage medium for identifying a commodity. Background Art

[0002] With the development of Internet information technology, e-commerce business has been growing day by day. At the same time, a number of online shopping mall platforms with different characteristics have also emerged. Manufacturers, agents, channel suppliers and other institutions often publish their products on multiple platforms for sale.

[0003] Since each e-commerce platform has a different method of product classification, and many products have diverse attributes that can correspond to different categories, the same product may be classified under different categories on different platforms. After each platform pushes the products to the digital procurement network, it is difficult to ensure unified classification, which makes it difficult for users to quickly determine whether two or more products are the same product, causing great difficulties for procurement management. Summary of the invention

[0004] The embodiments of the present application provide a method, device, server and storage medium for identifying the same commodity, which can solve the technical problem in the related art that it is difficult for users to quickly determine whether two or more commodities are the same commodity.

[0005] A first aspect of an embodiment of the present application provides a method for identifying a same commodity, comprising:

[0006] Acquire product information of the product to be identified, perform word segmentation processing on the product information, and determine a word segmentation group to be matched corresponding to the product information after the word segmentation processing, wherein the word segmentation group to be matched includes a keyword group to be matched and an auxiliary word group to be matched;

[0007] Matching the keyword group to be matched with the feature phrase of the feature information of the matched commodity feature library, and determining the matching degree between the commodity to be identified and the feature information of the matched commodity feature library, wherein the feature phrase corresponds to the matched commodity;

[0008] If the matching degree between the commodity to be identified and multiple feature information in the matched commodity feature library is greater than or equal to a preset matching degree threshold, the feature information whose corresponding matching degree is greater than or equal to the preset matching degree threshold is used as the target feature information, and the group similarity between the feature phrases of each target feature information and the to-be-matched sub-phrase groups of the commodity to be identified is determined, wherein the feature phrases of the target feature information include key phrases and auxiliary phrases;

[0009] The multiple group similarities are sorted to determine the highest group similarity. If the highest group similarity is greater than or equal to a preset group similarity threshold, the corresponding matched product and the product to be identified are the same product.

[0010] A second aspect of an embodiment of the present application provides a same-goods identification device, including:

[0011] An information acquisition unit is used to acquire commodity information of the commodity to be identified, perform word segmentation processing on the commodity information, and determine a to-be-matched word segmentation group corresponding to the commodity information after the word segmentation processing, wherein the to-be-matched word segmentation group includes a to-be-matched keyword group and a to-be-matched auxiliary word group;

[0012] A word segmentation matching unit matches the keyword group to be matched with the feature phrase of the feature information of the matched commodity feature library, and determines the matching degree between the commodity to be identified and the feature information of the matched commodity feature library, wherein the feature phrase corresponds to the matched commodity;

[0013] A group similarity determination unit, if the matching degree between the commodity to be identified and the plurality of feature information in the matched commodity feature library is greater than or equal to a preset matching degree threshold, then taking the feature information whose corresponding matching degree is greater than or equal to the preset matching degree threshold as the target feature information, and determining the group similarity between the feature phrases of each target feature information and the to-be-matched sub-phrase groups of the commodity to be identified, wherein the feature phrases of the target feature information include keyword phrases and auxiliary phrases;

[0014] The same product determination unit sorts the similarities of multiple target groups and determines the highest group similarity. If the highest group similarity is greater than or equal to a preset group similarity threshold, the corresponding matched product and the product to be identified are the same product.

[0015] A third aspect of an embodiment of the present application provides a server, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, each step of the same commodity identification method provided in the first aspect is implemented.

[0016] A fourth aspect of an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, each step of the same commodity identification method provided in the first aspect is implemented.

[0017] The same commodity identification method, device, server and storage medium provided by the embodiments of the present application have the following beneficial effects: first, the commodity information of the commodity to be identified is obtained, the commodity information is segmented, and the segmented word group to be matched corresponding to the commodity information after the segmentation processing is determined, wherein the segmented word group to be matched includes a keyword group to be matched and an auxiliary word group to be matched; then, the keyword group to be matched is matched with the feature word group of the feature information of the matched commodity feature library, and the matching degree of the commodity to be identified and the feature information of the matched commodity feature library is determined, wherein the feature word group corresponds to the matched commodity; then, if the commodity to be identified If the matching degree between the commodity and multiple feature information in the matched commodity feature library is greater than or equal to the preset matching degree threshold, the feature information with the corresponding matching degree greater than or equal to the preset matching degree threshold is used as the target feature information, and the group similarity between the feature phrases of each target feature information and the to-be-identified commodity segmented phrases is determined, wherein the target feature information feature phrases include keyword phrases and auxiliary phrases; finally, the multiple group similarities are sorted to determine the highest group similarity, and if the highest group similarity is greater than or equal to the preset group similarity threshold, the corresponding matched commodity and the to-be-identified commodity are the same commodity. Through the matching degree between the to-be-identified keyword phrase after the segmentation of the commodity to be identified and the feature information in the matched commodity feature library and the group similarity between the commodity to be identified and the commodity corresponding to the feature information in the matched commodity feature library, it is possible to quickly and accurately determine the same commodity of the commodity to be identified based on the matched commodity feature library, which helps to improve the efficiency of commodity procurement management. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 is a flowchart of an implementation method of a same commodity identification method provided in an embodiment of the present application;

[0020] Figure 2 is a schematic diagram of a method for identifying the same commodity provided in one embodiment of the present application;

[0021] Figure 3 is a schematic diagram of the same commodity recognition result provided by an embodiment of the present application;

[0022] Figure 4 is a flowchart for implementing the determination of group similarity provided by an embodiment of the present application;

[0023] Figure 5 is a flowchart for determining a matching degree according to an embodiment of the present application;

[0024] Figure 6 is a structural block diagram of a commodity identification device provided in one embodiment of the present application;

[0025] Figure 7 It is a structural block diagram of a server provided in an embodiment of the present application. DETAILED DESCRIPTION

[0026] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0027] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0028] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0029] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0030] In order to illustrate the technical solution of the present application, the following embodiments are used for illustration.

[0031] See also Figure 1 , Figure 1This is a flowchart for implementing the same commodity identification method provided in one embodiment of the present application. The flowchart may include the following steps 101 to 104.

[0032] Step 101, obtain product information of a product to be identified, perform word segmentation processing on the product information, and determine a word group to be matched corresponding to the product information after the word segmentation processing.

[0033] The to-be-matched word groups include to-be-matched keyword groups and to-be-matched auxiliary word groups. The keyword groups in the to-be-matched keyword groups are words that meet pre-set verification rules or appear frequently in product information, and other word groups outside the to-be-matched keyword groups are auxiliary word groups.

[0034] The keyword group to be matched includes a number of keywords to be matched, and the auxiliary phrase group to be matched includes a number of auxiliary phrases to be matched.

[0035] The above-mentioned product information is usually information used to describe the product to be identified, such as product brand, product name, specification model, etc.

[0036] The above-mentioned word segmentation processing is an operation of dividing a continuous text sequence into independent terms or words.

[0037] In this embodiment, the execution subject of the above-mentioned same commodity identification method is usually a server, and specifically a server of a reference node. It should be noted that the server can be hardware or software. When the server is hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or it can be implemented as a single server. When the server is software, it can be implemented as multiple software or software modules, or it can be implemented as a single software or software module, which is not specifically limited here.

[0038] In practice, the above-mentioned execution entity may use word segmentation tools, such as jieba, HanLP, etc., to segment the product information, and use the keywords obtained after the segmentation as the keywords to be matched.

[0039] For product information in different fields or industries, you can create a custom dictionary, including industry terms, brand names, specific product names, etc., to improve the accuracy of word segmentation and avoid some proper nouns being mis-segmented or missed.

[0040] In actual applications, the keywords corresponding to products of different brands may be different. The same word may be a keyword under one brand but not under another brand. The rules for becoming a keyword can be set according to actual usage needs.

[0041] Here, keywords may be determined based on predetermined rules. In practice, in order to improve the accuracy and efficiency of identifying the same product of a specific brand, exclusive predetermined rules may be set for the brand. At the same time, the rules may be adjusted and maintained during use.

[0042] In practice, before performing word segmentation processing on product information, it is also possible to preliminarily classify the product to be identified based on its brand to improve the accuracy of identifying the same product and the search efficiency.

[0043] Step 102, matching the keyword group to be matched with the feature phrases of the feature information of the matched commodity feature library, and determining the matching degree between the commodity to be identified and the feature information of the matched commodity feature library.

[0044] Among them, the characteristic phrases correspond to the matched products.

[0045] The above-mentioned matched product feature library is a feature library formed by extracting features from products that have been identified by the same product, and the feature library includes multiple feature information. Each feature information corresponds to a feature phrase, and the feature phrase includes multiple feature words used to describe the product. It should be pointed out that in the initial state, the feature library does not include any feature information, and the feature information of the feature library is continuously updated and accumulated through subsequent identification of the same product.

[0046] In practice, for each feature information in the feature library, the execution entity may use the keyword to be matched to perform similarity matching with the feature phrase of the information, and summarize the results of the similarity matching as the matching degree between the product to be identified and the feature information. The similarity matching method may include cosine similarity, Jaccard similarity, BERT similarity, etc.

[0047] Step 103, if the matching degree between the to-be-identified product and multiple feature information in the matched product feature library is greater than or equal to a preset matching degree threshold, the feature information whose corresponding matching degree is greater than or equal to the preset matching degree threshold is used as the target feature information, and the group similarity between the feature phrases of each target feature information and the to-be-matched sub-phrase groups of the to-be-identified product is determined.

[0048] The characteristic phrases of the target characteristic information include a keyword phrase and an auxiliary phrase. The keyword phrase includes a number of keywords, and the auxiliary phrase includes a number of auxiliary words.

[0049] The preset matching degree threshold is usually a preset matching degree threshold, for example, 0.55. When the actual matching degree is greater than or equal to the preset matching degree threshold, it indicates that the commodity to be identified and the target commodity may be the same commodity.

[0050] Specifically, the keyword group to be matched is matched with the keyword group of the characteristic phrase of the target characteristic information, and the auxiliary phrase to be matched is matched with the auxiliary phrase of the characteristic phrase of the target characteristic information.

[0051] In actual applications, when the matching degree between the product to be identified and multiple feature information in the matched product feature library is greater than or equal to the preset matching degree threshold, it means that there are multiple groups of target products that may be the same product to be identified, and further screening of the multiple groups of target products is required.

[0052] As an example, for the to-be-matched segmented words of different brands of products to be identified, verification rules and corresponding weight coefficients can be pre-set. Then, the above-mentioned execution entity can determine whether to perform weighting based on whether the to-be-matched segmented words meet the verification rules. The weighted character string length is used as the feature length of the to-be-matched segmented words. The sum of the feature lengths of each to-be-matched segmented word becomes the segmented feature length after the product to be identified is segmented. This length is used to calculate the matching degree and group similarity.

[0053] As another example, for each target product information, the above-mentioned execution entity can determine which words in the segmented words are keywords and which words are auxiliary words according to the rules, where the keywords are used to quickly match similar feature information in the feature library, and determine which product groups corresponding to which feature information the product information to be matched should be matched with to calculate the group similarity based on the obtained matching degree.

[0054] Step 104 , sorting multiple group similarities to determine the highest group similarity. If the highest group similarity is greater than or equal to a preset group similarity threshold, the corresponding matched product and the product to be identified are the same product.

[0055] The highest group similarity is a group similarity with the largest corresponding value among multiple group similarities.

[0056] In practice, the execution entity may sort the multiple group similarities by size to determine the highest group similarity, and then determine whether the product group corresponding to the feature information corresponding to the highest group similarity is the same product as the product to be identified.

[0057] The same product identification method provided in the present embodiment first obtains product information of the product to be identified, performs word segmentation processing on the product information, and determines the to-be-matched word segmentation group corresponding to the product information after the word segmentation processing, wherein the to-be-matched word segmentation group includes the to-be-matched keyword group and the to-be-matched auxiliary word group; then, the to-be-matched keyword group is matched with the feature word group of the feature information of the matched product feature library, and the matching degree of the product to be identified and the feature information of the matched product feature library is determined, wherein the feature word group corresponds to the matched product; then, if the matching degree of the product to be identified and multiple feature information in the matched product feature library is greater than or equal to a preset matching degree threshold, the feature information with the corresponding matching degree greater than or equal to the preset matching degree threshold is used as the target feature information, and the group similarity between the feature word group of each target feature information and the to-be-matched word segmentation group of the product to be identified is determined, wherein the feature word group of the target feature information includes the keyword group and the auxiliary word group; finally, multiple group similarities are sorted to determine the highest group similarity, and if the highest group similarity is greater than or equal to the preset group similarity threshold, the corresponding matched product and the product to be identified are the same product. By determining the degree of matching between the keyword group to be matched after the segmentation of the commodity to be identified and the characteristic information in the feature library of the matched commodities, and the group similarity between the commodity groups corresponding to the characteristic information of the commodity to be identified and the matched commodities, it is possible to quickly and accurately determine the same commodity of the commodity to be identified based on the feature library of the matched commodities, which helps to improve the efficiency of commodity procurement management; by determining the degree of matching between the keyword group to be matched and the characteristic phrase, the efficiency of identifying the same commodity is improved, and by determining the group similarity between the keyword group to be matched, the auxiliary phrase to be matched, the keyword group, and the auxiliary phrase, it helps to improve the accuracy of identifying the same commodity.

[0058] In some embodiments, if the highest group similarity is greater than or equal to a preset group similarity threshold, the target product corresponding to the highest group similarity is determined to be the same product as the product to be identified. If the highest group similarity is less than the preset group similarity threshold, it is determined that there is no same product as the product to be identified in the current product library, and the system automatically extracts the features of the product to be identified and creates a new feature in the feature library for subsequent determination of other products. The preset group similarity threshold is a pre-set value, which is usually greater than the preset matching threshold, for example, 0.65.

[0059] like Figure 2 As shown, Figure 2This is a schematic diagram of the same product identification method provided by an embodiment of the present application. First, the product information of the product to be identified is segmented to obtain the keyword group to be matched and the auxiliary phrase to be matched; then the matching degree of the keyword group to be matched is calculated with each feature phrase in the matched product feature library. The matching degree of multiple products of brand A with the keyword group to be matched is greater than or equal to the preset matching degree threshold. The keyword group to be matched and the keyword group of product A1, the auxiliary phrase to be matched and the auxiliary phrase of product A1 are used to calculate the group similarity between the product to be identified and product A1; the keyword group to be matched and the keyword group of product A2, the auxiliary phrase to be matched and the auxiliary phrase of product A1 are used to calculate the group similarity between the product to be identified and product A1; The group similarity between the commodity to be identified and commodity A2 is calculated by using the keyword group to be matched and the keyword group of commodity A3, and the auxiliary phrase to be matched and the auxiliary phrase of commodity A3. Finally, the three group similarities are sorted to determine the maximum group similarity, and it is determined whether the maximum group similarity is greater than or equal to the preset group similarity threshold. If the maximum group similarity is greater than or equal to the preset group similarity threshold, the matched commodity corresponding to the maximum group similarity is determined as the same commodity as the commodity to be identified.

[0060] It should be noted that when determining the matching degree and group similarity, it is necessary to verify the multiple groups of input words based on the preset verification rules, and match each word in each group with the preset verification rules. When any group has words matching the verification rules, other groups must also have words matching the verification rules, otherwise the similarity of the multiple groups of input words is judged to be zero.

[0061] like Figure 3 As shown, Figure 3 This is a schematic diagram of the same product recognition result provided by an embodiment of the present application. The product information of the product to be recognized is segmented to obtain a keyword group to be matched and an auxiliary phrase to be matched; the keyword group to be matched is matched with the feature phrase of the feature information of the matched product feature library to obtain the matching degree of the feature information of the product to be recognized and the matched product feature library.

[0062] See also Figure 4 , Figure 4 This is a flowchart for implementing the determination of group similarity provided by an embodiment of the present application. The flowchart may include the following steps 401 to 403.

[0063] Step 401 , determining intersection features and union features according to the word groups to be matched of the commodity to be identified and the word groups of commodity group information corresponding to the feature information of the matched commodity feature library.

[0064] The intersection feature is the words or features that exist in common in several to-be-matched segmented words and the segmented word groups of the product group information corresponding to the feature information, and the union feature is the sum of the words or features of several to-be-matched keywords and the feature phrase groups.

[0065] In practice, the execution entity may match each keyword to be matched with each feature word in the feature phrase, obtain words or features that exist in both the keyword phrase to be matched and the feature phrase, and determine them as intersection features.

[0066] In practice, the execution entity may determine the sum of words or features of several keywords and feature phrases to be matched as a union feature.

[0067] Step 402, determining the intersection feature length according to the intersection feature, and determining the union feature length according to the union feature.

[0068] Among them, the above-mentioned intersection feature length is the feature length corresponding to the intersection feature, and the above-mentioned union feature length is the feature length corresponding to the union feature.

[0069] In practice, the above-mentioned execution entity can adopt a pre-set feature length algorithm, such as the term frequency-inverse document frequency weighted algorithm (Term Frequency-Inverse Document Frequency, TF-IDF), the best match algorithm (Best Match 25, BM25), the part-of-speech weighted algorithm, the entropy weighted algorithm, etc., to determine the intersection feature length and the union feature length.

[0070] Step 403, determining the group similarity according to the intersection feature length and the union feature length.

[0071] As an example, the intersection feature length and the union feature length are calculated according to preset rules. The above-mentioned execution entity can match each word in the intersection feature and union feature group with the preset verification rules. When matching the preset verification rules, the feature length of the word is the sum of the word string length and the weighted length corresponding to the verification rule. Otherwise, the feature length of the word is the string length of the word. The feature lengths of each word are summarized to form the intersection feature length and the union feature length, and their ratio is determined as the group similarity.

[0072] The method for determining group similarity provided in the embodiment of the present application, the intersection feature can accurately reflect the words or features commonly contained in the to-be-identified product and the target product, the union feature can reflect the range of the total words or features of the to-be-identified product and the target product, and the group similarity determined by the intersection feature length and the union feature length can accurately reflect the similarity between the to-be-identified product and the target product, which helps to improve the accuracy of identifying the same product.

[0073] In some embodiments, the above-mentioned determining the group similarity based on the intersection feature length and the union feature length may include: determining the ratio of the intersection feature length to the union feature length as the group similarity.

[0074] In practice, the ratio of the intersection feature length to the union feature length can reflect the proportion of words or features commonly contained in the to-be-identified product and the target product to the total words or features of the to-be-identified product and the target product, which can further accurately reflect the similarity between the to-be-identified product and the target product, and help to further improve the accuracy of identifying the same product.

[0075] In practice, when determining the matching degree and group similarity, it is necessary to verify the input multiple groups of words based on the preset verification rules, and match each word in each group with the preset verification rules. When any group has a word matching the verification rule, the other groups must also have words matching the verification rule, otherwise the similarity of the input multiple groups of words is determined to be zero. This algorithm is of decisive significance for distinguishing the same series of products with mostly the same product description information but different key parameters.

[0076] See also Figure 5 , Figure 5 This is a flowchart for determining a matching degree provided by an embodiment of the present application. The flowchart may include the following steps 501 to 502.

[0077] Step 501, a plurality of keywords to be matched after word segmentation of the product to be identified are matched with feature phrases of feature information in a matched product feature library respectively, and all matching results are summarized to obtain the matching length of all feature information of the product to be identified and the matched product feature library.

[0078] The above-mentioned word segmentation matching length is the sum of the characteristic lengths of all matching words between the characteristic phrase and the keyword phrase to be matched.

[0079] Each keyword to be matched after the segmentation of the product to be identified is matched with the feature phrase of each feature information in the matched product feature library. If a match can be made, the matching length of the feature information is added to the feature length of the keyword to be matched, and the segmentation matching length of the product to be identified and each feature information is summarized.

[0080] In practice, the execution entity may use the matching algorithm to obtain the matching length between the feature information of the product to be identified and all the matched product feature libraries.

[0081] Step 502, determining the matching degree according to the feature length of the keyword group to be matched of the product to be identified, the feature length of the feature phrase of the feature information of the matched product feature library, and the matching length of the feature information of the product to be identified and all the matched product feature libraries.

[0082] In practice, the matching length between the product to be identified and the feature information, the feature length of the keyword group to be matched and the feature length of the feature phrase of the feature information should all be calculated through the feature length. The matching degree is the result of dividing the matching length between the product to be identified and the feature information by the minimum value of the feature length of the keyword group to be matched and the feature length of the feature phrase of the feature information.

[0083] In practice, the feature length calculation matches the input word with the preset verification rule. When the match is successful, the feature length of the word is the sum of the word string length and the weighted length corresponding to the verification rule. Otherwise, the feature length of the word is the string length of the word. When a group of words is input, the feature length of each word will be calculated separately, and the feature length of all words can be summed up to get the feature length of the phrase.

[0084] In some embodiments, before performing word segmentation processing on the product information, it also includes: performing a pre-processing operation on the product information.

[0085] The preprocessing operation includes at least one of data format unification, interference information removal, and information completion.

[0086] Data format unification usually means standardizing product information according to a pre-set format. For product information from different sources, its format is standardized into a consistent format to facilitate subsequent processing and comparison. For example, price, weight, etc. are unified into a consistent unit and format.

[0087] Removing interference information usually involves deleting invalid information, redundant information, erroneous information, sales information irrelevant to product information, etc. For example, special characters, spaces, "Contact customer service", "non-standard customization", "random color", etc. in the name or price are deleted.

[0088] Information completion usually involves filling in missing parts of the data, such as filling "S110935-15g" with "S110935, 15g".

[0089] In this embodiment, the chaotic commodity information can be sorted out through preprocessing, the required information can be extracted and processed and optimized through algorithms to facilitate subsequent processing.

[0090] In some embodiments, after determining a number of keywords to be matched corresponding to the product information after word segmentation processing, the method may further include: determining the characteristic length of the keyword group to be matched according to the characteristic length of each word in the keyword group to be matched.

[0091] By weighting the length of keywords, different weights can be assigned according to the different importance of keywords or the influence of feature length, which can more accurately measure the importance of keywords and better utilize these keyword information in subsequent processing.

[0092] In practice, the execution entity may use a preset feature length algorithm to weight the length of each keyword group to be matched.

[0093] As an example, the execution subject may simply use the length of the keyword as a weight factor, and the longer the keyword, the higher the weight.

[0094] As another example, the execution subject may use a nonlinear function to weight the keyword length to more finely adjust the weights of keywords of different lengths.

[0095] In some embodiments, the same commodity identification method may further include: forming a feature phrase according to a preset format with a number of keywords to be matched corresponding to the commodity to be identified, storing the formed feature phrase in a feature library, and updating the feature library in real time.

[0096] See also Figure 6 , Figure 6 6 is a structural block diagram of a commodity identification device 600 provided in one embodiment of the present application, including:

[0097] The information acquisition unit 601 is used to acquire the commodity information of the commodity to be identified, perform word segmentation processing on the commodity information, and determine the word segmentation group to be matched corresponding to the commodity information after the word segmentation processing, wherein the word segmentation group to be matched includes the keyword group to be matched and the auxiliary word group to be matched;

[0098] The word segmentation matching unit 602 is used to match the keyword group to be matched with the feature word group of the feature information of the matched commodity feature library, and determine the matching degree between the commodity to be identified and the feature information of the matched commodity feature library, wherein the feature word group corresponds to the matched commodity;

[0099] The group similarity determination unit 603 is used to determine the group similarity between the feature phrases of each target feature information and the to-be-identified product to-be-identified product to-be-matched sub-phrase groups, if the matching degree between the to-be-identified product and the plurality of feature information in the matched product feature library is greater than or equal to a preset matching degree threshold, and then use the feature information with a corresponding matching degree greater than or equal to the preset matching degree threshold as the target feature information, and determine the group similarity between the feature phrases of each target feature information and the to-be-matched sub-phrase groups of the to-be-identified product, wherein the feature phrases of the target feature information include keyword phrases and auxiliary phrases;

[0100] The same product determination unit 604 is used to sort the similarities of multiple target groups and determine the highest group similarity. If the highest group similarity is greater than or equal to a preset group similarity threshold, the corresponding matched product and the product to be identified are the same product.

[0101] The same commodity identification device provided in the present embodiment first obtains commodity information of the commodity to be identified, performs word segmentation processing on the commodity information, and determines the to-be-matched word segmentation group corresponding to the commodity information after the word segmentation processing, wherein the to-be-matched word segmentation group includes the to-be-matched keyword group and the to-be-matched auxiliary word group; then, the to-be-matched keyword group is matched with the feature word group of the feature information of the matched commodity feature library, and the matching degree of the commodity to be identified and the feature information of the matched commodity feature library is determined, wherein the feature word group corresponds to the matched commodity; then, if the matching degree of the commodity to be identified and multiple feature information in the matched commodity feature library is greater than or equal to a preset matching degree threshold, the feature information with the corresponding matching degree greater than or equal to the preset matching degree threshold is used as the target feature information, and the group similarity between the feature word group of each target feature information and the to-be-matched word segmentation group of the commodity to be identified is determined, wherein the feature word group of the target feature information includes the keyword group and the auxiliary word group; finally, the multiple group similarities are sorted to determine the highest group similarity, and if the highest group similarity is greater than or equal to the preset group similarity threshold, the corresponding matched commodity and the commodity to be identified are the same commodity. By determining the degree of matching between the keyword group to be matched after the segmentation of the commodity to be identified and the characteristic information in the feature library of the matched commodities, and the group similarity between the commodity groups corresponding to the characteristic information of the commodity to be identified and the matched commodities, it is possible to quickly and accurately determine the same commodity of the commodity to be identified based on the feature library of the matched commodities, which helps to improve the efficiency of commodity procurement management; by determining the degree of matching between the keyword group to be matched and the characteristic phrase, the efficiency of identifying the same commodity is improved, and by determining the group similarity between the keyword group to be matched, the auxiliary phrase to be matched, the keyword group, and the auxiliary phrase, it helps to improve the accuracy of identifying the same commodity.

[0102] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / units are based on the same concept as the embodiment of the same product identification method in this application. Their specific functions and technical effects can be found in the embodiment section of the same product identification method and will not be repeated here.

[0103] See also Figure 7 , Figure 7 700 is a structural block diagram of a server 700 provided in an embodiment of the present application. The server 700 in this embodiment includes: at least one processor 701 ( Figure 7Only one processor is shown in the figure), memory 702, and a computer program 703 stored in the memory 702 and executable on at least one processor 701, such as a time synchronization program for a wireless network. When the processor 701 executes the computer program 703, the steps in the above-mentioned embodiments of the same commodity identification method are implemented. When the processor 701 executes the computer program 703, the functions of the modules / units in the above-mentioned embodiments of the apparatus are implemented, for example, Figure 6 The functions of the information acquisition unit 601 to the same product determination unit 604 are shown.

[0104] Exemplarily, the computer program 703 can be divided into one or more units, one or more units are stored in the memory 702, and are executed by the processor 701 to complete the present application. One or more units can be a series of computer program instruction segments that can perform specific functions, and the instruction segments are used to describe the execution process of the computer program 703 in the server 700. For example, the computer program 703 can be divided into an information acquisition unit, a word segmentation matching unit, a group similarity determination unit, and a same product determination unit. The specific functions of each unit have been described in the above embodiments and will not be repeated here.

[0105] The server 700 may be a computing device such as a server, a desktop computer, a tablet computer, a cloud server, and a mobile terminal. The server 700 may include, but is not limited to, a processor 701 and a memory 702. Those skilled in the art will appreciate that Figure 7 This is only an example of the server 700 and does not constitute a limitation of the server 700. The server 700 may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the server may also include input and output devices, network access devices, buses, etc.

[0106] The processor 701 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0107] The memory 702 may be an internal storage unit of the server 700, such as a hard disk or memory of the server 700. The memory 702 may also be an external storage device of the server 700, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the server 700. Optionally, the memory 702 may also include both an internal storage unit of the server 700 and an external storage device. The memory 702 is used to store computer programs and other programs and data required by the server 700. The memory 702 may also be used to temporarily store data that has been output or is to be output.

[0108] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0109] If the integrated module is implemented in the form of 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 present application implements all or part of the processes in the above-mentioned embodiment method, and the computer program that can be completed by instructing the relevant hardware through a computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device that can carry computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electric carrier signals and telecommunication signals.

[0110] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0111] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for identifying identical commodities, characterized in that: The method comprises: Acquire commodity information of the commodity to be identified, perform word segmentation processing on the commodity information, and determine a word segmentation group to be matched corresponding to the commodity information after the word segmentation processing, wherein the word segmentation group to be matched includes a keyword group to be matched and an auxiliary word group to be matched; Matching the keyword group to be matched with the feature phrases of the feature information of the matched commodity feature library, and determining the matching degree between the commodity to be identified and the feature information of the matched commodity feature library, wherein the feature phrases correspond to the matched commodity; If the matching degree between the commodity to be identified and the plurality of feature information in the matched commodity feature library is greater than or equal to a preset matching degree threshold, the feature information whose corresponding matching degree is greater than or equal to the preset matching degree threshold is used as the target feature information, and the group similarity between the feature phrases of each target feature information and the to-be-matched sub-phrase groups of the commodity to be identified is determined, wherein the feature phrases of the target feature information include keyword phrases and auxiliary phrases; The plurality of group similarities are sorted to determine a highest group similarity. If the highest group similarity is greater than or equal to a preset group similarity threshold, the corresponding matched commodity and the commodity to be identified are the same commodity.

2. The same commodity identification method according to claim 1, characterized in that: If the matching degree between the commodity to be identified and multiple feature information in the matched commodity feature library is greater than or equal to a preset matching degree threshold, the feature information with a corresponding matching degree greater than or equal to the preset matching degree threshold is used as the target feature information, and the group similarity between the feature phrases of each target feature information and the to-be-matched sub-phrase groups of the commodity to be identified is determined, wherein the feature phrases of the target feature information include keyword phrases and auxiliary phrases, including: Determine intersection features and union features according to the to-be-matched word groups of the to-be-identified commodity and the feature word groups of the target feature information; Determine the intersection feature length according to the intersection feature, and determine the union feature length according to the union feature; The group similarity is determined according to the intersection feature length and the union feature length.

3. The same commodity identification method according to claim 2, characterized in that: The determining the group similarity according to the intersection feature length and the union feature length comprises: The ratio of the intersection feature length to the union feature length is determined as the group similarity.

4. The same commodity identification method according to claim 1, characterized in that: The method further comprises: The keyword group to be matched is matched with the feature phrase of the feature information of the matched product feature library. When the keyword group to be matched contains a preset verification rule and the feature phrase does not contain the preset verification rule, or the feature phrase contains the preset verification rule and the keyword group to be matched does not contain the preset verification rule, the group similarity is determined to be zero.

5. The same commodity identification method according to claim 1, characterized in that: The method of matching the keyword group to be matched with the feature phrase of the feature information of the matched commodity feature library to determine the matching degree between the commodity to be identified and the feature information of the matched commodity feature library includes: Matching the multiple keyword groups to be matched after the word segmentation of the product to be identified with the feature word groups of the feature information of the matched product feature library respectively, and summarizing all the matching results to obtain the matching length of the product to be identified and all the feature information of the matched product feature library; The matching degree is determined according to the characteristic length of the keyword group to be matched of the product to be identified, the characteristic length of the characteristic phrase of the characteristic information of the matched product characteristic library, and the matching length of the characteristic information of the product to be identified and all the matched product characteristic libraries.

6. The same commodity identification method according to claim 5, characterized in that: Determining the degree of matching according to the feature length of the keyword group to be matched of the commodity to be identified, the feature length of the feature phrase of the feature information of the matched commodity feature library, and the matching length of the feature information of the commodity to be identified and all the matched commodity feature libraries, including: Determine the value of the matching length between the keyword group to be matched and the feature information of the matched commodity feature library; The matching degree is determined by the ratio of the matching length of the feature information of the product to be identified and the matched product feature library to the minimum value of the feature length of the key word group to be matched of the product to be identified and the feature length of the feature phrase of the feature information of the matched product feature library.

7. The same commodity identification method according to claim 1, characterized in that: Before the word segmentation processing is performed on the commodity information of the commodity to be identified, the method further includes: The commodity information of the commodity to be identified is preprocessed, wherein the preprocessing operation includes at least one of data format unification, interference information removal, and information completion.

8. The same commodity identification method according to any one of claims 1 to 7, characterized in that: After determining a number of keywords to be matched corresponding to the product information after the word segmentation processing, the method further includes: When the keyword to be matched meets the preset verification rule, the length of the keyword to be matched is weighted according to the length set by the preset verification rule.

9. A device for identifying identical goods, characterized in that: include: An information acquisition unit, used to acquire commodity information of a commodity to be identified, perform word segmentation processing on the commodity information, and determine a keyword group to be matched corresponding to the commodity information after the word segmentation processing, wherein the keyword group to be matched includes a plurality of keywords to be matched; A word segmentation matching unit matches the keyword group to be matched with a feature phrase of feature information of a matched commodity feature library, and determines a matching degree between the commodity to be identified and the feature information of the matched commodity feature library, wherein the feature phrase corresponds to the matched commodity; A group similarity determination unit, if the matching degree between the commodity to be identified and the plurality of feature information in the matched commodity feature library is greater than or equal to a preset matching degree threshold, then taking the feature information whose corresponding matching degree is greater than or equal to the preset matching degree threshold as the target feature information, and determining the group similarity between the feature phrases of each target feature information and the to-be-matched sub-phrase groups of the commodity to be identified, wherein the feature phrases of the target feature information include keyword phrases and auxiliary phrases; The same product determination unit sorts the similarities of the plurality of target groups and determines the highest group similarity. If the highest group similarity is greater than or equal to a preset group similarity threshold, the corresponding matched product and the product to be identified are the same product.

10. A server comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the same commodity identification method as described in any one of claims 1 to 8 is implemented.

11. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the same commodity identification method as described in any one of claims 1 to 8 is implemented.

Citation Information

Cited By

  • Goods and commodity matching method, equipment, medium and product

    CN120996911A