Article recall method and device, electronic equipment and computer readable storage medium

Abstract

The invention provides an article recall method and device, electronic equipment and a computer readable storage medium, belongs to the technical field of data processing, and particularly relates to the technical fields of big data, search recall and the like. According to the specific implementation scheme, the method comprises the steps of obtaining article information of a to-be-replaced article according to an article detail page of the to-be-replaced article; retrieving according to the item information to obtain a plurality of items recalled for the first time; and according to the article information, determining an article with the highest similarity with the to-be-replaced article from the plurality of initially recalled articles as a final recalled article.

Description

Technical Field

[0001] The present disclosure relates to the field of data processing technology, and in particular to the technical fields of big data, search and recall, etc. Specifically, the present disclosure relates to an item recall method and device, an electronic device, and a computer-readable storage medium. Background Art

[0002] With the rapid development of the Internet, e-commerce has gradually matured and is closely related to people's lives.

[0003] In order to improve the experience and competitiveness of the e-commerce platform, it may be necessary to replace the products found on other e-commerce platforms, and the replaced products can also meet the user needs in the same way. Summary of the invention

[0004] The present disclosure provides an item recall method and device, an electronic device, and a computer-readable storage medium.

[0005] According to a first aspect of the present disclosure, there is provided an item recall method, the method comprising:

[0006] According to the item details page of the item to be replaced, obtain the item information of the item to be replaced;

[0007] Retrieve a plurality of first-time recalled items according to the item information;

[0008] An item having the highest similarity to the item to be replaced is determined from the plurality of initially recalled items according to the item information as a final recalled item.

[0009] According to a second aspect of the present disclosure, there is provided an item recall device, the device comprising:

[0010] A parsing module, used for acquiring item information of the item to be replaced according to the item details page of the item to be replaced;

[0011] A retrieval module, used for searching and acquiring a plurality of initially recalled items according to the item information;

[0012] A screening module is used to determine, from the multiple initially recalled items, an item with the highest similarity to the item to be replaced as a final recalled item according to the item information.

[0013] According to a third aspect of the present disclosure, an electronic device is provided, the electronic device comprising:

[0014] at least one processor; and

[0015] A memory in communication with the at least one processor; wherein,

[0016] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the item recall method.

[0017] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable a computer to execute the above-mentioned item recall method.

[0018] According to a fifth aspect of the present disclosure, a computer program product is provided, comprising a computer program, which implements the above-mentioned item recall method when executed by a processor.

[0019] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0021] Figure 1 It is a flowchart of a method for face liveness detection provided by an embodiment of the present disclosure;

[0022] Figure 2 It is a schematic diagram of a process of obtaining item information of an item to be replaced according to information of a core area in a specific implementation of an item recall method provided by an embodiment of the present disclosure;

[0023] Figure 3 is a flowchart of some steps of another item recall method provided by an embodiment of the present disclosure;

[0024] Figure 4 is a flowchart of some steps of another item recall method provided by an embodiment of the present disclosure;

[0025] Figure 5 is a flowchart of some steps of another item recall method provided by an embodiment of the present disclosure;

[0026] Figure 6 is a flowchart of some steps of another item recall method provided by an embodiment of the present disclosure;

[0027] Figure 7 is a flowchart of some steps of another item recall method provided by an embodiment of the present disclosure;

[0028] Figure 8 is a structural schematic diagram of an item recall device provided by an embodiment of the present disclosure;

[0029] Fig. 9 It is a block diagram of an electronic device used to implement the item recall method of the embodiment of the present disclosure. DETAILED DESCRIPTION

[0030] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0031] The B2B (Business-to-Business) industry has a wide variety of products, including standard products such as electronic chips and hardware, as well as non-standard products such as large-scale machinery and equipment, instruments and meters, and recycled materials.

[0032] In some possible implementations, the category of a standard product may be defined by fixed parameters, such as the category of an electronic chip may be confirmed by its model and brand information.

[0033] However, the parameters and features of non-standard products cannot be matched one-to-one, so it is impossible to determine similar products that meet user needs through product parameters.

[0034] The item recall method and device, electronic device, and computer-readable storage medium provided in the embodiments of the present disclosure are intended to solve at least one of the above technical problems in the prior art.

[0035] The item recall method provided in the embodiment of the present disclosure can be executed by an electronic device such as a terminal device or a server, and the terminal device can be a vehicle-mounted device, a user equipment (User Equipment, UE), a mobile device, a user terminal, a terminal, a cellular phone, a cordless phone, a personal digital assistant (Personal Digital Assistant, PDA), a handheld device, a computing device, a vehicle-mounted device, a wearable device, etc. The method can be implemented by a processor calling a computer-readable program instruction stored in a memory. Alternatively, the method can be executed by a server.

[0036] Figure 1 FIG. 1 is a flow chart of an item recall method provided by an embodiment of the present disclosure. Figure 1 As shown in , the item recall method provided by the embodiment of the present disclosure may include step S110, step S120, and step S130.

[0037] In step S110, the item information of the item to be replaced is obtained according to the item details page of the item to be replaced;

[0038] In step S120, a plurality of first recalled items are retrieved based on the item information;

[0039] In step S130, an item with the highest similarity to the item to be replaced is determined from the multiple initially recalled items according to the item information as the final recalled item.

[0040] For example, in step S110, the item details page of the item to be replaced may be a page provided by the e-commerce platform that specifically introduces the item to be replaced.

[0041] In some possible implementations, the item details page may include multiple areas, such as tabs, item titles, core areas, other areas, and the like.

[0042] Among them, the core area is the area where the core information representing the item (such as the main picture of the product, product price, logistics, evaluation, product parameters, manufacturer information, etc.) is located, and other areas are usually page subsidiary areas, such as the recommendation module, which includes product recommendations, manufacturer recommendations, recommendation words, and content recommendations.

[0043] In some possible implementations, the core area of ​​the item details page may be determined by performing area detection on the item details page of the item to be replaced, and the item information of the item to be replaced may be obtained through the information of the core area.

[0044] In some possible implementations, the item information of the item to be replaced may include at least one core word and at least one parameter item, wherein the core word is a keyword characterizing the characteristics of the item to be replaced, and the parameter item is a parameter of the item to be replaced, such as general item parameters, regional parameters, brand parameters, etc.

[0045] For example, in “Shandong steamed bun machine, fully automatic Xu Zhong steamed bread machine, available in school breakfast shops, inexpensive and of high quality”, “steamed bun machine” and “steamed bread machine” are keywords that represent the types of items to be replaced, and are core words; “fully automatic”, “school”, and “breakfast shop” are parameters of “steamed bun machine” and “steamed bread machine”, and are parameter items.

[0046] In some possible implementations, the item title of the item to be replaced may be obtained through information in the core area of ​​the item details page, and the item title may be parsed into core words and parameter items.

[0047] In some possible implementations, the item title can be parsed by a pre-trained model, which can parse the item title into core words and parameter items, and calculate the word distance between the parameter items and the core words, so as to obtain the initial recalled items according to the item information.

[0048] In some possible implementations, the item information of the item to be replaced may also include manufacturer information of the item to be replaced and a picture of the item to be replaced, which may be obtained from the information in the core area.

[0049] The manufacturer information of the items to be replaced may include the manufacturer's credit code, the region where the manufacturer is located, the manufacturer's main product categories, etc.

[0050] In some possible implementations, in step S120, searching based on item information may be performed by combining core words and parameter items into a query, searching based on the query, and using the retrieved multiple items as initial recall items.

[0051] In some possible implementations, a core word and a parameter item that is closest to the core word may be used to form a query word.

[0052] In some possible implementations, when the number of initially recalled items obtained based on a query term does not meet a preset requirement (such as failing to reach a preset number), a core term is used as a query term and a search is performed again to obtain more initially recalled items so that the number of initially recalled items obtained meets the preset requirement.

[0053] In some possible implementations, searching according to the query term may be searching at pre-set candidate sites according to the query term.

[0054] Among them, the pre-set candidate site may be a specific e-commerce platform, that is, the initial recalled items obtained in the item recall method provided in the embodiment of the present disclosure may be items sold on the specific e-commerce platform.

[0055] In some possible implementations, in step S130, the similarity between the to-be-replaced object and the initially recalled object may be calculated to determine the object with the highest similarity to the to-be-replaced object from the initially recalled objects as the final recalled object.

[0056] In some possible implementations, the similarity between the item to be replaced and the initially recalled item may be determined by calculating the similarity between the item detail page of the initially recalled item and the item detail page of the item to be replaced.

[0057] In some possible implementations, the similarity between the to-be-replaced item and the initially-recalled item may also be determined by the similarity between the manufacturer information of the to-be-replaced item and the manufacturer information of the initially-recalled item.

[0058] In some possible implementations, the similarity between the to-be-replaced object and the initially recalled object may also be determined by the similarity between the object image of the to-be-replaced object and the object image of the initially recalled object.

[0059] In some possible implementations, before calculating the similarity between the items to be replaced and the initially recalled items, the initially recalled items may be filtered by item information, and some items among multiple initially recalled items that are obviously different from the items to be replaced may be deleted without performing similarity calculations, so as to reduce the amount of calculations and improve efficiency.

[0060] In some possible implementations, the item details page of the item to be replaced may be replaced with the item details page of the item to be replaced, so that the user can browse the item details of the item to be recalled. Since the first recalled item may be retrieved and obtained on a specific e-commerce platform, the finally recalled item is also retrieved and obtained on a specific e-commerce platform.

[0061] In the item recall method provided in the embodiment of the present disclosure, item information obtained from the item details page of the item to be replaced is retrieved and screened to obtain the final recalled item with the highest similarity to the item to be replaced. Since the final recalled item is retrieved and filtered based on the item information, the final recalled item and the item to be replaced can meet the same user needs, which solves the problem that when the item to be replaced is a non-standard product, it is impossible to determine an item similar to it that meets the user's needs through item parameters.

[0062] In some possible implementations, the item recall method provided in the embodiments of the present disclosure can be used to replace items sold on e-commerce platforms that may have fraud risks with items sold on more credible and secure e-commerce platforms, thereby providing protection for users' safe browsing and online shopping.

[0063] In some possible implementations, the item recall method provided in the embodiments of the present disclosure can be used to replace items sold on e-commerce platforms that occupy more memory and network resources with items sold on e-commerce platforms that occupy less memory and network resources, so as to reduce the memory and network resources occupied by online shopping browsed by users and provide users with smoother services.

[0064] The following is a detailed introduction to the item recall method provided by the embodiment of the present disclosure.

[0065] As described above, the core area of ​​the item details page can be determined by performing area detection on the item details page of the item to be replaced, and the item information of the item to be replaced can be obtained through the information of the core area.

[0066] Among them, the core area is the area where the core information representing the item (such as the main picture of the product, product price, logistics, evaluation, product parameters, manufacturer information, etc.) is located, and other areas are usually page subsidiary areas, such as the recommendation module, which includes product recommendations, manufacturer recommendations, recommendation words, and content recommendations.

[0067] Since the layouts of item details pages of different e-commerce platforms are basically the same, the relative position of the core area of ​​the item details page is predictable.

[0068] In some possible implementations, the core area of ​​the item details page may be acquired by utilizing the relative position of the core area of ​​the item details page on the item details page.

[0069] In some possible implementations, an image segmentation algorithm may be used to perform region detection on the item details page to obtain a core region of the item details page.

[0070] Through area detection, other content on the item details page can be quickly filtered out, so that the core information of the item to be replaced can be quickly and easily located, and the item information of the item to be replaced can be obtained efficiently and quickly.

[0071] The above are merely examples of methods for implementing region detection, and other methods for implementing region detection are within the protection scope of the embodiments of the present disclosure.

[0072] Figure 2 FIG. 1 shows a schematic diagram of obtaining the item information of the item to be replaced based on the information of the core area in a specific implementation manner. Figure 2 As shown in , the core area can obtain the item title of the item on the e-commerce platform, the manufacturer information of the manufacturer that produces the item to be replaced, the picture of the item to be replaced in the item details page (i.e., the main picture of the item to be replaced), etc.

[0073] In some possible implementations, the manufacturer information may include the manufacturer's credit code, the region where the manufacturer is located, the manufacturer's main product categories, etc.

[0074] In some possible implementations, the item information of the item to be replaced may be obtained by parsing the content of the core area.

[0075] In some possible implementations, the item information of the item to be replaced may include at least one core word and at least one parameter item. Figure 2As shown in , the core words and parameter items can be obtained by parsing the item title.

[0076] In some possible implementations, the item title may be input into a pre-trained parsing model, and the model may be used to parse the item title into core words, parameter items, invalid content, and the like.

[0077] Among them, the core words are keywords that characterize the characteristics of the items to be replaced, and the parameter items are the parameters of the items to be replaced, such as general parameters of the items, regional parameters, brand parameters, etc.

[0078] The model can also be used to calculate the word distance between the parameter item and the core word. When multiple core words are parsed, the word distance between the parameter item and the core word can specifically be the word distance between the parameter item and the core word closest to it.

[0079] The word distance between the parameter item and the core word closest to it may specifically be the number of words between the parameter item and the core word closest to it. In some specific implementations, a space is counted as one word during the calculation process, and multiple consecutive spaces are also counted as one word.

[0080] For example, the title of the item to be replaced is "Shandong steamed bun machine, fully automatic Xu Zhong steamed bread machine, available in school breakfast shops, high quality and low price", and the core words determined through analysis are "steamed bun machine" and "steamed bread machine". The obtained parameter items include the common parameters of the item "fully automatic", "school", and "breakfast shop", the brand parameter "Xu Zhong", and the regional parameter "Shandong". The invalid content obtained is "high quality and low price" and "available".

[0081] Among them, the word distance between "fully automatic" and the nearest core word "steamed bun machine" is 2, the word distance between "school" and the nearest core word "steamed bun machine" is 2, the word distance between "breakfast shop" and the nearest core word "steamed bun machine" is 3, the word distance between "Xu Zhong" and the nearest core word "steamed bun machine" is 1, and the word distance between "Shandong" and the nearest core word "steamed bun machine" is 1.

[0082] As described above, a search may be performed based on core words and parameter items to obtain multiple recalled items.

[0083] Figure 3 FIG. 1 shows a flow chart of a specific implementation method for searching and obtaining the first recalled items according to the core words and parameter items. Figure 3 As shown in , searching according to core words and parameter items to obtain multiple recalled items may include step S310.

[0084] In step S310, a query word is determined according to at least one core word and at least one parameter item, and a search is performed according to the query word to obtain a plurality of first-time recalled items.

[0085] In some possible implementations, invalid content obtained through parsing may be filtered, the obtained core words and parameter items may be combined into query words, and retrieval may be performed based on the query words to obtain multiple initial recall results.

[0086] In some possible implementations, for each core word, a query word may be determined based on each core word and a parameter item that is closest to the core word.

[0087] Specifically, for each core word, a parameter item with the smallest distance to the core word may be obtained to form a query word with the core word.

[0088] Compared with other parameter items, the parameter item closest to the core word has a greater probability of being more related to the core word. Therefore, the query word composed of the core word and the parameter item closest to the core word is more likely to more accurately describe the characteristics of the item to be replaced, and the initial recalled items retrieved based on the query word are more similar to the item to be replaced.

[0089] The final recalled items are determined from the initial recalled items, that is, the initial recalled items are candidates for the final recalled items, and the initial recalled items list is the candidate list for the final recalled items.

[0090] The more candidate items included in the candidate list, the greater the possibility that the quality of the final recalled items (i.e., the similarity with the items to be replaced) is higher. In order to ensure that the final recalled items obtained can meet the requirements, the number of initial recalled items included in the corresponding candidate list should meet certain requirements.

[0091] In some possible implementations, when the number of initially recalled items obtained based on a query term consisting of a core word and a parameter item with the smallest distance from the core word does not meet a preset requirement (such as failing to reach a preset number), the core word and other parameter items (i.e., parameter items other than the parameter item with the smallest distance from the core word) may be used to form a query term to increase the number of initially recalled items obtained.

[0092] In some possible implementations, the core words may be used as query words to perform a search again to obtain more first-time recalled items, so that the number of first-time recalled items obtained meets a preset requirement.

[0093] Figure 4 FIG. 1 shows a schematic diagram of a process of using a core word as a query word and performing a search again in a specific implementation manner, such as Figure 4 As shown in , using the core word as the query word, searching again may include step S410.

[0094] In step S410, when the number of initially recalled items does not meet a preset requirement, at least one core word information is used as a query word to perform a search to obtain a plurality of initially recalled items.

[0095] In some possible implementations, each core word is used as a query word to perform a search and obtain multiple search results.

[0096] The search results obtained by searching with "times" as the query word may overlap with the search results obtained by searching with the query word and the query word composed of the query word and the parameter items. Therefore, these results need to be merged to ensure that the merged results meet the preset number.

[0097] In the case where the number of combined results exceeds the preset number, the combined results may be screened according to the ranking of the results generated by the search platform for the search. Since the results generated by the search platform are ranked according to the relevance to the query term, screening the combined results according to the ranking of the results generated by the search platform may retain the results with a higher relevance to the query term as the first recalled items.

[0098] For example, the title of the item to be replaced is "Shandong steamed bun machine, fully automatic Xu Zhong steamed bread machine, available in school breakfast shops, high quality and low price", and the core words determined through analysis are "steamed bun machine" and "steamed bread machine". The obtained parameter items include the common parameters of the item "fully automatic", "school", and "breakfast shop", the brand parameter "Xu Zhong", and the regional parameter "Shandong". The invalid content obtained is "high quality and low price" and "available".

[0099] Among them, the word distance between "fully automatic" and the nearest core word "steamed bun machine" is 2, the word distance between "school" and the nearest core word "steamed bun machine" is 2, the word distance between "breakfast shop" and the nearest core word "steamed bun machine" is 3, the word distance between "Xu Zhong" and the nearest core word "steamed bun machine" is 1, and the word distance between "Shandong" and the nearest core word "steamed bun machine" is 1.

[0100] That is to say, the parameter item with the shortest word distance to the core word "steamed bun machine" is "Xu Zhong", and the parameter item with the shortest word distance to the core word "baozi machine" is "Shandong". Therefore, "Xu Zhong steamed bun machine" and "Shandong baozi machine" are used as query words for retrieval to obtain multiple first-time recalled items.

[0101] Assuming that the number of initially recalled items obtained is 50 and 30 respectively, the number of initially recalled items after merger is 70, and the number of initially recalled items desired is 100, then "fully automatic steamed bun machine" can be used as the search term to obtain multiple initially recalled items.

[0102] Assuming that the number of the merged first-recall items is 90, which still does not meet the desired number of first-recall items, you can use "steamed bun machine" and "dumpling machine" as search terms to search again and obtain multiple first-recall items.

[0103] Assuming that the number of merged first-recall items is 105, the five first-recall items with the weakest relevance are deleted according to the relevance ranking given by the retrieval platform to obtain 100 first-recall items.

[0104] In some possible implementations, searching according to the query term may be searching at pre-set candidate sites according to the query term.

[0105] Among them, the pre-set candidate site may be a specific e-commerce platform, that is, the initial recalled items obtained in the item recall method provided in the embodiment of the present disclosure may be items sold on the specific e-commerce platform.

[0106] In some possible implementations, a specific e-commerce platform may be a more credible and secure e-commerce platform. Therefore, the initially recalled items obtained are also items sold on a more credible and secure e-commerce platform, and are safer than the items to be replaced.

[0107] In some possible implementations, the specific e-commerce platform may be an e-commerce platform that occupies less memory and network resources. Therefore, the first recalled items obtained are also items sold on the e-commerce platform that occupies less memory and network resources. Compared with purchasing items to be replaced, purchasing first-time recalled items occupies less memory and network resources, and can provide users with smoother services.

[0108] As described above, the similarity between the to-be-replaced object and the initially recalled object can be calculated to determine the object with the highest similarity to the to-be-replaced object from the initially recalled objects as the final recalled object.

[0109] In some possible implementations, the similarity between the item to be replaced and the initially recalled item may be determined by calculating the similarity between the item detail page of the initially recalled item and the item detail page of the item to be replaced.

[0110] Figure 5 A schematic diagram of a process for calculating the similarity between the item details page of the initially recalled item and the item details page of the item to be replaced in a specific implementation manner is shown. Figure 5 As shown in , calculating the similarity between the item details page of the initially recalled item and the item details page of the to-be-replaced item may include step S510 and step S520.

[0111] In step S510, a plurality of first-time recalled items are used as a search candidate queue, and the item titles of the items to be replaced are used as query terms for search;

[0112] In step S520, the similarity between the item detail pages of the plurality of initially recalled items and the item detail pages of the item to be replaced is determined according to the search results.

[0113] In some possible implementations, the obtained multiple first-recalled items are used as a retrieval candidate queue, and the item titles of the items to be replaced are used as query terms for retrieval. The similarity between the item details page of the first-recalled item and the item details page of the item to be replaced is determined based on the ranking of the first-recalled items in the retrieval results.

[0114] Since the search results generated by the search platform are sorted according to the relevance to the query terms, and this search uses the item title of the item to be replaced as the query term, the order of the search results obtained is based on the relevance to the item title of the item to be replaced. The higher the ranking of the first-recall item, the stronger the relevance to the item title of the item to be replaced, and the higher the similarity between its item details page and the item details page of the item to be replaced. Conversely, the lower the ranking of the first-recall item, the lower the relevance to the item title of the item to be replaced, and the lower the similarity between its item details page and the item details page of the item to be replaced.

[0115] This method can utilize the retrieval algorithm of the retrieval platform to simply and efficiently calculate the similarity between the item details page of the initially recalled item and the item details page of the item to be replaced.

[0116] Since the manufacturers of the same items must have similarities, in some possible implementations, the similarity between the item to be replaced and the initially recalled item can also be determined based on the similarity between the manufacturer information of the item to be replaced and the manufacturer information of the initially recalled item.

[0117] Figure 6 A schematic diagram of a process for calculating the similarity between the manufacturer information of the to-be-replaced item and the manufacturer information of the initially recalled item in a specific implementation manner is shown.

[0118] like Figure 6 As shown in , it is determined whether the manufacturer credit code of the item to be replaced is consistent with the manufacturer credit code of the initially recalled item. If the credit codes are consistent, it means that the manufacturer that produced the item to be replaced is consistent with the manufacturer that produced the initially recalled item, and the manufacturer information of the item to be replaced has the highest similarity with the manufacturer information of the initially recalled item.

[0119] If the credit codes are inconsistent, it means that the manufacturer of the item to be replaced is inconsistent with the manufacturer of the initially recalled item. In this case, the region and main product category of the manufacturer of the item to be replaced and the manufacturer of the initially recalled item are judged. If the region of the manufacturer of the item to be replaced and the manufacturer of the initially recalled item are the same, it means that the manufacturer of the item to be replaced and the manufacturer of the initially recalled item may be related manufacturers. Similarly, if the main product category of the manufacturer of the item to be replaced and the manufacturer of the initially recalled item are the same, it means that the manufacturer of the item to be replaced and the manufacturer of the initially recalled item may be related manufacturers. In other words, the manufacturer information of the item to be replaced and the manufacturer information of the initially recalled item have a certain degree of similarity.

[0120] Among them, whether the region where the manufacturer of the item to be replaced is located is consistent with the region where the manufacturer of the item to be replaced is located can be determined based on the distance between the location of the manufacturer of the item to be replaced and the location of the manufacturer of the item to be recalled.

[0121] If the region and main product category of the manufacturer of the item to be replaced are inconsistent with those of the manufacturer of the initially recalled item, it means that the manufacturer information of the item to be replaced is very low in similarity with the manufacturer information of the initially recalled item.

[0122] In some specific implementations, the similarity between the manufacturer information of the item to be replaced and the manufacturer information of the initially recalled item may be represented by a score.

[0123] If the manufacturer's credit code of the item to be replaced is consistent with the manufacturer's credit code of the initially recalled item, the production score of the initially recalled item shall be determined as the first score; if the manufacturer's credit code of the item to be replaced is inconsistent with the manufacturer's credit code of the initially recalled item, the region of the manufacturer of the item to be replaced is consistent with the region of the manufacturer of the initially recalled item, and the main product category of the manufacturer of the item to be replaced is consistent with the main product category of the manufacturer of the initially recalled item, the production score of the initially recalled item shall be determined as the second score; if the manufacturer's credit code of the item to be replaced is inconsistent with the manufacturer's credit code of the initially recalled item , if the region of the manufacturer of the to-be-replaced item is inconsistent with the region of the manufacturer of the initially recalled item, and the main product category of the manufacturer of the to-be-replaced item is consistent with the main product category of the manufacturer of the initially recalled item, the production score of the initially recalled item is determined to be the third score; if the manufacturer credit code of the to-be-replaced item is inconsistent with the manufacturer credit code of the initially recalled item, the region of the manufacturer of the to-be-replaced item is consistent with the region of the manufacturer of the initially recalled item, and the main product category of the manufacturer of the to-be-replaced item is inconsistent with the main product category of the manufacturer of the initially recalled item, the production score of the initially recalled item is determined to be the fourth score;

[0124] Among them, the first score is greater than the second score, the second score is greater than the third score, and the second score is greater than the fourth score.

[0125] In some possible implementations, the similarity between the to-be-replaced object and the initially recalled object may also be determined by the similarity between the object image of the to-be-replaced object and the object image of the initially recalled object.

[0126] In some possible implementations, the item image of the item to be replaced may be the item main image of the item to be replaced, and the item image of the item being initially recalled may be the item main image of the item being initially recalled.

[0127] Since the main image is the image that best reflects the characteristics of the product, the similarity of the main image also represents the similarity between the item to be replaced and the initially recalled item to a certain extent.

[0128] Specifically, a method for calculating image similarity may be used to determine the similarity between the image of the object to be replaced and the image of the object initially recalled.

[0129] Figure 7 A schematic diagram of a process for determining the item with the highest similarity to the item to be replaced from multiple initially recalled items as the final recalled item based on the similarity between the item details page of the initially recalled item and the item details page of the item to be replaced, the similarity between the manufacturer information of the item to be replaced and the manufacturer information of the initially recalled item, and the similarity between the picture of the item to be replaced and the picture of the initially recalled item in a specific implementation manner is shown, as shown in FIG. Figure 7 As shown in , it may include step S710, step S720, step S730, and step S740.

[0130] In step S710, a similarity score of the initially recalled item is determined based on the similarity between the item details page of the initially recalled item and the item details page of the item to be replaced;

[0131] In step S720, the production score of the initially recalled item is determined based on the manufacturer information of the item to be replaced and the manufacturer information of the initially recalled item;

[0132] In step S730, the picture similarity score of the initially recalled item is determined according to the picture similarity between the picture of the item to be replaced and the picture of the initially recalled item;

[0133] In step S740, an item with the highest similarity to the item to be replaced is determined from among the multiple initially recalled items according to the similarity scores, production scores, and image similarity scores of the multiple initially recalled items as the final recalled item.

[0134] In some possible implementations, before step S710, the first recalled items may be filtered by item information, and some items among the multiple first recalled items that are obviously different from the items to be replaced may be deleted without performing similarity calculations, so as to reduce the amount of calculations and improve efficiency.

[0135] In some possible implementations, for each initially recalled item, the item information of the initially recalled item may be obtained. The specific method for obtaining the item information of the initially recalled item is consistent with the specific method for obtaining the item information of the item to be replaced, and will not be described in detail here.

[0136] As described above, the item information obtained from the initial recall of items may also include at least one core word, at least one parameter item, manufacturer information, pictures of the initial recall of items, and the like.

[0137] In some possible implementations, multiple first-recalled items are grouped into a first-recalled item queue. For each first-recalled item in the first-recalled item queue, if a core word of the first-recalled item does not have the same core word as a core word of an item to be replaced, the first-recalled item is deleted from the first-recalled item queue.

[0138] According to the type of the item to be replaced, the expert system is used to determine the key parameter items in the parameter items of the item to be replaced. If the item to be replaced is an electronic chip, the key parameter items are the brand and model; if the item to be replaced is rubber, the key parameter items are the CAS number and material.

[0139] For each first-recalled item in the first-recalled item queue, if the key parameter items of the first-recalled item are inconsistent with the key parameter items of the item to be replaced, the first-recalled item will be deleted from the first-recalled item queue; if the key parameter items of the first-recalled item are not obtained, the deduction value of the first-recalled item is determined, and the deduction value is used to determine the item with the highest similarity to the item to be replaced from multiple first-recalled items as the final recalled item.

[0140] In some possible implementations, the deduction value d may be 0.1.

[0141] In some possible implementations, in step S710, Figure 5 The illustrated method calculates the similarity between the item detail page of each initially recalled item in the initially recalled item queue and the item detail page of the item to be replaced.

[0142] In some possible implementations, while calculating the similarity between the item details page of the first recalled item and the item details page of the item to be replaced, the store score and quality score of the store selling the first recalled item are also considered to obtain a similarity score a. <a≤1。

[0143] In some possible implementations, in step S720, the Figure 6 The method shown calculates a production score b for each initially recalled item in the initially recalled item queue.

[0144] In some specific implementations, if the manufacturer credit code of the item to be replaced is consistent with the manufacturer credit code of the item initially recalled, the production score b of the item initially recalled is determined to be 1; if the manufacturer credit code of the item to be replaced is inconsistent with the manufacturer credit code of the item initially recalled, the production score b of the item initially recalled is determined to be b1+b2;

[0145] Among them, if the straight-line distance between the manufacturer of the item to be replaced and the manufacturer of the first recalled item is within 5km, b1 = 0.1; if the manufacturer of the item to be replaced and the manufacturer of the first recalled item are in the same city, b1 = 0; if the manufacturer of the item to be replaced and the manufacturer of the first recalled item are not in the same city, b1 = -0.05. b2 can be determined by the ratio of the number of repeated types in the main categories of the manufacturer of the item to be replaced and the manufacturer of the first recalled item to the sum of the main categories of the manufacturer of the item to be replaced and the manufacturer of the first recalled item.

[0146] In some possible implementations, in step S730, for each first-recalled item in the first-recalled item queue, the image similarity score c of the first-recalled item can be obtained by inputting the item main image of the item to be replaced and the item main image of the first-recalled item into a pre-trained image similarity calculation model, where 0 <c≤1。

[0147] In some possible implementations, in step S740, the final score k of each initially recalled item in the initially recalled item queue is k=a+b+cd, and the initially recalled item with the largest k value in the initially recalled item queue is taken as the final recalled item.

[0148] Based on Figure 1 The same principle as shown in the method, Figure 8 A schematic diagram of the structure of an item recall device provided by an embodiment of the present disclosure is shown. Figure 8 As shown, the item recall device 80 may include:

[0149] The parsing module 810 is used to obtain the item information of the item to be replaced according to the item details page of the item to be replaced;

[0150] A retrieval module 820, configured to retrieve a plurality of initially recalled items according to item information;

[0151] The screening module 830 is used to determine the item with the highest similarity to the item to be replaced from the multiple initially recalled items according to the item information as the final recalled item.

[0152] In the item recall device provided in the embodiment of the present disclosure, the item information obtained from the item details page of the item to be replaced is retrieved and screened to obtain the final recalled item with the highest similarity to the item to be replaced. Since the final recalled item is retrieved and filtered based on the item information, the final recalled item and the item to be replaced can meet the same user needs, which solves the problem that when the item to be replaced is a non-standard product, it is impossible to determine an item similar to it that meets the user's needs through item parameters.

[0153] In some possible implementations, the item information of the item to be replaced includes at least one core word and at least one parameter item; the retrieval module 820 is also used to determine a query word based on the at least one core word and the at least one parameter item, and perform a search based on the query word to obtain multiple first-recalled items; wherein the core word is a keyword that characterizes the characteristics of the item to be replaced; and the parameter item is a parameter of the item to be replaced.

[0154] In some possible implementations, the retrieval module 820 includes: a query word unit, used to determine the query word according to each core word and the parameter item closest to the core word; and a primary retrieval unit, used to perform retrieval according to the query word to obtain multiple primary recalled items.

[0155] In some possible implementations, the retrieval module 820 further includes: a second retrieval unit configured to retrieve at least one core word information as a query word to obtain a plurality of initially recalled items when the number of initially recalled items does not meet a preset requirement.

[0156] In some possible implementations, the screening module 830 includes: a similarity calculation unit, used to calculate the similarity between the item details page of the initially recalled item and the item details page of the item to be replaced; and a final calculation unit, used to determine, from among multiple initially recalled items, the item with the highest similarity to the item to be replaced as the final recalled item based at least on the similarity.

[0157] In some possible implementations, the similarity calculation unit includes: a retrieval subunit, used to use multiple first-recalled items as a retrieval candidate queue and the item title of the item to be replaced as a query term for retrieval; a relevance acquisition unit, used to determine the similarity between the item details page of the multiple first-recalled items and the item details page of the item to be replaced based on the retrieval results.

[0158] In some possible implementations, the item information of the item to be replaced also includes the manufacturer information of the item to be replaced and the picture of the item to be replaced; the final calculation unit includes: a first scoring sub-unit, used to determine the similarity score of the initially recalled item according to the similarity between the item details page of the initially recalled item and the item details page of the item to be replaced; a second scoring sub-unit, used to determine the production score of the initially recalled item according to the manufacturer information of the item to be replaced and the manufacturer information of the initially recalled item; a third scoring sub-unit, used to determine the picture similarity score of the initially recalled item according to the picture similarity between the picture of the item to be replaced and the picture of the initially recalled item; a score calculation sub-unit, used to determine the item with the highest similarity to the item to be replaced from multiple initially recalled items as the final recalled item according to the similarity scores, production scores, and picture similarity scores of multiple initially recalled items.

[0159] In some possible implementations, the manufacturer information includes the manufacturer's credit code, the manufacturer's region, and the manufacturer's main product category; the second score sub-unit is used to: when the manufacturer credit code of the to-be-replaced item is consistent with the manufacturer credit code of the initially recalled item, determine the production score of the initially recalled item as the first score; when the manufacturer credit code of the to-be-replaced item is inconsistent with the manufacturer credit code of the initially recalled item, the region of the manufacturer of the to-be-replaced item is consistent with the region of the manufacturer of the initially recalled item, and the main product category of the manufacturer of the to-be-replaced item is consistent with the main product category of the manufacturer of the initially recalled item, determine the production score of the initially recalled item as the second score; when the manufacturer credit code of the to-be-replaced item is inconsistent with the manufacturer credit code of the initially recalled item, the region of the manufacturer of the to-be-replaced item is consistent with the region of the manufacturer of the initially recalled item, and the main product category of the manufacturer of the to-be-replaced item is consistent with the main product category of the manufacturer of the initially recalled item If the manufacturer credit code of the to-be-replaced item is inconsistent, the region of the manufacturer of the to-be-replaced item is inconsistent with the region of the manufacturer of the initially recalled item, and the main product category of the manufacturer of the to-be-replaced item is consistent with the main product category of the manufacturer of the initially recalled item, the production score of the initially recalled item is determined to be the third score; if the manufacturer credit code of the to-be-replaced item is inconsistent with the manufacturer credit code of the initially recalled item, the region of the manufacturer of the to-be-replaced item is consistent with the region of the manufacturer of the initially recalled item, and the main product category of the manufacturer of the to-be-replaced item is inconsistent with the main product category of the manufacturer of the initially recalled item, the production score of the initially recalled item is determined to be the fourth score; the first score is greater than the second score, the second score is greater than the third score, and the second score is greater than the fourth score.

[0160] In some possible implementations, the screening module 830 also includes: a screening unit, used to: determine at least one core word and at least one parameter item of the first recalled item; when the core word of the item to be replaced does not have the same word as the core word of the first recalled item, delete the first recalled item from the first recalled item queue; when the key parameter item of the item to be replaced is inconsistent with the key parameter item of the first recalled item, delete the first recalled item from the first recalled item queue; when the first recalled item does not have a key item parameter, determine the deduction value of the first recalled item; the deduction value is used to determine the item with the highest similarity to the item to be replaced from multiple first recalled items as the final recalled item; wherein the key parameter item is determined from the parameter item according to the expert system; the similarity calculation unit is used to: calculate the similarity between the item details page of the first recalled item in the first recalled item queue and the item details page of the item to be replaced.

[0161] In some possible implementations, the parsing module 810 includes: a region unit, configured to perform region detection on the item details page of the item to be replaced to determine a core region of the item details page; and a region analysis unit, configured to obtain item information of the item to be replaced based on information of the core region.

[0162] In some possible implementations, the item information of the item to be replaced includes at least one core word and at least one parameter item; the regional analysis unit is further used to: obtain the item title of the item to be replaced based on the information of the core area; input the item title of the item to be replaced into a pre-trained parsing model to obtain at least one core word and at least one parameter item; wherein the core word is a keyword that characterizes the characteristics of the item to be replaced; and the parameter item is a parameter of the item to be replaced.

[0163] In some possible implementations, the retrieval module 820 is further configured to: perform a search at a preset candidate site according to the item information to obtain a plurality of initially recalled items.

[0164] It can be understood that the above modules of the item recall device in the embodiment of the present disclosure have the function of realizing Figure 1 The functions of the corresponding steps of the item recall method in the embodiment shown in . The function can be implemented by hardware, or by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above functions. The above modules can be software and / or hardware, and the above modules can be implemented separately or integrated with multiple modules. For the functional description of each module of the above item recall device, please refer to Figure 1 The corresponding description of the item recall method in the embodiment shown in will not be repeated here.

[0165] In the technical solution disclosed herein, the acquisition, storage and application of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0166] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0167] The electronic device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the item recall method provided in the embodiment of the present disclosure.

[0168] Compared with the prior art, the electronic device retrieves and filters the item information obtained from the item details page of the item to be replaced to obtain the final recalled item with the highest similarity to the item to be replaced. Since the final recalled item is retrieved and filtered based on the item information, the final recalled item and the item to be replaced can meet the same user needs, which solves the problem that when the item to be replaced is a non-standard product, it is impossible to determine an item similar to it that meets the user's needs through item parameters.

[0169] The readable storage medium is a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable a computer to execute the item recall method provided in the embodiment of the present disclosure.

[0170] Compared with the prior art, the readable storage medium retrieves and filters the item information obtained from the item details page of the item to be replaced to obtain the final recalled item with the highest similarity to the item to be replaced. Since the final recalled item is retrieved and filtered based on the item information, the final recalled item and the item to be replaced can meet the same user needs, which solves the problem that when the item to be replaced is a non-standard product, it is impossible to determine an item similar to it that meets the user's needs through item parameters.

[0171] The computer program product includes a computer program, and when the computer program is executed by a processor, the method for recalling items provided in the embodiment of the present disclosure is implemented.

[0172] Compared with the prior art, the computer program product retrieves and filters the item information obtained from the item details page of the item to be replaced to obtain the final recalled item with the highest similarity to the item to be replaced. Since the final recalled item is retrieved and filtered based on the item information, the final recalled item and the item to be replaced can meet the same user needs, which solves the problem that when the item to be replaced is a non-standard product, it is impossible to determine an item similar to it that meets the user's needs through item parameters.

[0173] Fig. 9 A schematic block diagram of an example electronic device 900 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0174] like Fig. 9 As shown, the device 900 includes a computing unit 901, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 902 or a computer program loaded from a storage unit 908 into a random access memory (RAM) 903. In the RAM 903, various programs and data required for the operation of the device 900 can also be stored. The computing unit 901, the ROM 902, and the RAM 903 are connected to each other via a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0175] A number of components in the device 900 are connected to the I / O interface 905, including: an input unit 906, such as a keyboard, a mouse, etc.; an output unit 907, such as various types of displays, speakers, etc.; a storage unit 908, such as a disk, an optical disk, etc.; and a communication unit 909, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 909 allows the device 900 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0176] The computing unit 901 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 901 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 901 performs the various methods and processes described above, such as the item recall method. For example, in some embodiments, the item recall method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 908. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 900 via the ROM 902 and / or the communication unit 909. When the computer program is loaded into the RAM 903 and executed by the computing unit 901, one or more steps of the item recall method described above may be performed. Alternatively, in other embodiments, the computing unit 901 may be configured to perform the item recall method in any other appropriate manner (e.g., by means of firmware).

[0177] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0178] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0179] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0180] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0181] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0182] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0183] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0184] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for recalling an item, comprising: According to the item details page of the item to be replaced, obtain the item information of the item to be replaced; Retrieve a plurality of first-time recalled items according to the item information; An item having the highest similarity to the item to be replaced is determined from the plurality of initially recalled items according to the item information as a final recalled item.

2. The method according to claim 1, wherein: The item information of the item to be replaced includes at least one core word and at least one parameter item; The step of retrieving and acquiring a plurality of initially recalled items according to the item information includes: Determine a query word according to at least one of the core words and at least one of the parameter items, and perform a search according to the query word to obtain a plurality of first-time recalled items; The core words are keywords that characterize the features of the item to be replaced; and the parameter items are parameters of the item to be replaced.

3. The method according to claim 2, wherein: The step of determining a query word according to at least one of the core words and at least one of the parameter items, and performing a search according to the query word to obtain a plurality of first-time recalled items includes: Determine a query word according to each of the core words and the parameter item closest to the core word; A search is performed according to the query term to obtain a plurality of first-time recalled items.

4. The method according to claim 2, wherein: After searching according to the query word and obtaining a plurality of initially recalled items, the method further includes: When the number of the initially recalled items does not meet a preset requirement, at least one of the core word information is used as a query word to perform a search to obtain a plurality of initially recalled items.

5. The method according to claim 2, wherein: The step of determining, from the plurality of initially recalled items, an item having the highest similarity to the item to be replaced as a final recalled item according to the item information includes: Calculating the similarity between the item details page of the initially recalled item and the item details page of the to-be-replaced item; At least according to the similarity, an item with the highest similarity to the item to be replaced is determined from the multiple initially recalled items as the final recalled item.

6. The method according to claim 5, wherein: The calculating the similarity between the item details page of the initially recalled item and the item details page of the to-be-replaced item includes: Using the plurality of initially recalled items as a search candidate queue and using the item title of the item to be replaced as a query word for searching; Determine the similarity between the item detail pages of the plurality of initially recalled items and the item detail page of the item to be replaced according to the search result.

7. The method according to claim 5, wherein: The item information of the item to be replaced also includes the manufacturer information of the item to be replaced and a picture of the item to be replaced; The step of determining, at least according to the similarity, from the plurality of initially recalled items, an item having the highest similarity to the item to be replaced as the final recalled item comprises: Determining a similarity score of the initially recalled item according to a similarity between an item details page of the initially recalled item and an item details page of the item to be replaced; Determining the production score of the initially recalled item according to the manufacturer information of the item to be replaced and the manufacturer information of the initially recalled item; Determining a picture similarity score of the initially recalled item according to a picture similarity between the picture of the to-be-replaced item and the picture of the initially recalled item; According to the similarity scores, the production scores, and the image similarity scores of the multiple initially recalled items, an item with the highest similarity to the item to be replaced is determined from the multiple initially recalled items as a final recalled item.

8. The method according to claim 7, wherein: The manufacturer information includes the manufacturer's credit code, the manufacturer's region, and the manufacturer's main product category; The determining the production score of the initially recalled item according to the manufacturer information of the item to be replaced and the manufacturer information of the initially recalled item includes: In the case where the manufacturer credit code of the to-be-replaced item is consistent with the manufacturer credit code of the initially recalled item, determining the production score of the initially recalled item as a first score; In the case where the manufacturer credit code of the to-be-replaced item is inconsistent with the manufacturer credit code of the initially recalled item, the region of the manufacturer of the to-be-replaced item is consistent with the region of the manufacturer of the initially recalled item, and the main product category of the manufacturer of the to-be-replaced item is consistent with the main product category of the manufacturer of the initially recalled item, the production score of the initially recalled item is determined to be the second score; If the manufacturer credit code of the to-be-replaced item is inconsistent with the manufacturer credit code of the initially recalled item, the region of the manufacturer of the to-be-replaced item is inconsistent with the region of the manufacturer of the initially recalled item, and the main product category of the manufacturer of the to-be-replaced item is consistent with the main product category of the manufacturer of the initially recalled item, the production score of the initially recalled item is determined to be the third score; In the case where the manufacturer credit code of the to-be-replaced item is inconsistent with the manufacturer credit code of the initially recalled item, the region of the manufacturer of the to-be-replaced item is consistent with the region of the manufacturer of the initially recalled item, and the main product category of the manufacturer of the to-be-replaced item is inconsistent with the main product category of the manufacturer of the initially recalled item, the production score of the initially recalled item is determined to be the fourth score; The first score is greater than the second score, the second score is greater than the third score, and the second score is greater than the fourth score.

9. The method according to claim 5, wherein: Before calculating the similarity between the item details page of the initially recalled item and the item details page of the to-be-replaced item, the method further includes: Determining at least one core word and at least one parameter item of the initially recalled item; In the case where the core word of the to-be-replaced item does not have the same core word as the core word of the initially recalled item, deleting the initially recalled item from the initially recalled item queue; In the case where the key parameter items of the to-be-replaced item are inconsistent with the key parameter items of the initially recalled item, deleting the initially recalled item from the initially recalled item queue; In the case that the first recalled item does not have key item parameters, determining a deduction value of the first recalled item; the deduction value is used to determine the item with the highest similarity to the item to be replaced from the multiple first recalled items as the final recalled item; wherein the key parameter items are determined from the parameter items according to an expert system; The calculating the similarity between the item details page of the initially recalled item and the item details page of the to-be-replaced item includes: The similarity between the item details page of the initially recalled item in the initially recalled item queue and the item details page of the to-be-replaced item is calculated.

10. The method according to claim 1, wherein: The acquiring the item information of the item to be replaced according to the item details page of the item to be replaced includes: Performing area detection on the item details page of the item to be replaced to determine the core area of ​​the item details page; The item information of the item to be replaced is obtained according to the information of the core area.

11. The method according to claim 10, wherein: The item information of the item to be replaced includes at least one core word and at least one parameter item; The acquiring the item information of the item to be replaced according to the information of the core area includes: Acquire the item title of the item to be replaced according to the information of the core area; Inputting the title of the item to be replaced into a pre-trained parsing model to obtain at least one of the core words and at least one parameter item; The core words are keywords that characterize the features of the item to be replaced; and the parameter items are parameters of the item to be replaced.

12. The method according to claim 1, wherein: The step of retrieving and acquiring a plurality of initially recalled items according to the item information comprises: A plurality of first-time recalled items are retrieved at pre-set candidate sites according to the item information.

13. An item recall device, comprising: A parsing module, used for acquiring item information of the item to be replaced according to the item details page of the item to be replaced; A retrieval module, used for searching and acquiring a plurality of initially recalled items according to the item information; A screening module is used to determine, from the multiple initially recalled items, an item with the highest similarity to the item to be replaced as a final recalled item according to the item information.

14. The device according to claim 13, wherein: The item information of the item to be replaced includes at least one core word and at least one parameter item; The retrieval module is further used to determine a query word according to at least one of the core words and at least one of the parameter items, and perform a search according to the query word to obtain a plurality of first-time recalled items; The core words are keywords that characterize the features of the item to be replaced; and the parameter items are parameters of the item to be replaced.

15. The device according to claim 14, wherein: The retrieval module comprises: A query word unit, used to determine a query word according to each of the core words and a parameter item with a word distance closest to the core word; The initial retrieval unit is used to perform retrieval according to the query word to obtain multiple initial recalled items.

16. The device according to claim 14, wherein: The retrieval module also includes: The re-retrieval unit is used to retrieve at least one of the core word information as a query word to obtain multiple initially recalled items when the number of the initially recalled items does not meet the preset requirement.

17. The device according to claim 14, wherein: The screening module comprises: a similarity calculation unit, configured to calculate the similarity between the item details page of the initially recalled item and the item details page of the to-be-replaced item; The final calculation unit is used to determine, at least according to the similarity, from the multiple initially recalled items, the item with the highest similarity to the item to be replaced as the final recalled item.

18. An electronic device, comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 12.

19. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-12.

20. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 12.

Images