Resource recommendation method, device, equipment and computer readable storage medium

Abstract

The application discloses a resource recommendation method and device, equipment and a computer readable storage medium, and belongs to the technical field of computers. The method comprises the following steps: acquiring search information input in an independent station; if a resource keyword does not exist in the search information, determining a reference resource category corresponding to the search information, or if the resource keyword exists in the search information and a resource included in the independent station does not match the resource keyword, determining the reference resource category corresponding to the search information; and recommending a first resource of the reference resource category included in the independent station. In the case that a resource keyword does not exist or a resource cannot be searched based on the resource keyword, the resource recommended to the user is a resource meeting the demand of the user, so that the click rate and conversion rate of the recommended resource can be improved, and the total amount of resource transactions of the independent station can be improved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a resource recommendation method, apparatus, device, and computer-readable storage medium. Background Technology

[0002] With the continuous development of computer technology, more and more businesses are recommending their resources to users through independent websites. An independent website is a website built by a business.

[0003] In related technologies, a user's query request is received, which includes search information. If the resource keywords in the search information do not match the literal meaning of the resources contained in the independent website, or if the search information fails to provide resource keywords, there may be a situation with 0 results pages or random recommendations.

[0004] The resource recommendation methods described above result in a low degree of matching between recommended resources and users, leading to low click-through rates and conversion rates. Summary of the Invention

[0005] This application provides a resource recommendation method, apparatus, device, and computer-readable storage medium, which can be used to solve the problem in related technologies where the matching degree between recommended resources and users is low, resulting in low click-through rates and conversion rates of recommended resources. The technical solution is as follows: On one hand, embodiments of this application provide a resource recommendation method, the method comprising: Retrieve search information entered on an independent website; If the search information does not contain the resource keyword, determine the reference resource category corresponding to the search information; or, if the search information contains the resource keyword, and the resources contained in the independent website do not match the resource keyword, determine the reference resource category corresponding to the search information. The first resource in the reference resource category included in the independent website is recommended.

[0006] In one possible implementation, determining the reference resource category corresponding to the search information includes: The large language model is invoked to perform semantic parsing on the search information to obtain the target scene information corresponding to the search information; Based on the correspondence between scene information and resource categories, the reference resource category corresponding to the target scene information is determined.

[0007] In one possible implementation, determining the reference resource category corresponding to the search information includes: The large language model is invoked to perform semantic parsing on the search information to obtain the target scene information corresponding to the search information; If the search information contains resource keywords, the reference resource category corresponding to the target scene information and the resource keywords is determined based on the correspondence between scene information, keywords and resource categories.

[0008] In one possible implementation, before recommending the first resource of the reference resource category included in the independent website, the method further includes: The resources in the reference resource category included in the independent website are used as candidate resources; Based on the operational metrics information of each candidate resource, a recommendation index is determined for each candidate resource. The operational metrics information includes at least one of the following: resource amount, resource click-through rate, resource conversion rate, resource sales volume, or resource add-to-cart volume. The first resource is determined from among the candidate resources based on the recommendation index of each candidate resource.

[0009] In one possible implementation, the method further includes: Acquire historical behavior data, which includes at least one of historical purchase data, historical browsing data, or historical add-to-cart data; Based on the historical behavior data, user attributes are determined; The step of determining the first resource from among the candidate resources based on the recommendation index of each candidate resource includes: The first resource is determined from among the candidate resources based on the user attributes and the recommendation index of each candidate resource.

[0010] In one possible implementation, determining the first resource from among the candidate resources based on the user attributes and the recommendation index of each candidate resource includes: Among the candidate resources, a first candidate resource with a recommendation index not lower than a recommendation index threshold is determined; among the first candidate resources, a first resource whose resource attributes include the user attributes is determined; or... Among the candidate resources, a second candidate resource is determined whose resource attributes include the user attributes; among the second candidate resources, a first resource with a recommendation index not lower than the recommendation index threshold is determined.

[0011] In one possible implementation, the recommendation index threshold is determined based on the number of resources contained in the independent website, and the recommendation index threshold is proportional to the number of resources contained in the independent website.

[0012] On the other hand, embodiments of this application provide a resource recommendation device, the device comprising: The acquisition module is used to retrieve search information entered on the independent website; The determination module is used to determine the reference resource category corresponding to the search information if the search information does not contain the resource keyword, or to determine the reference resource category corresponding to the search information if the search information contains the resource keyword and the resources contained in the independent website do not match the resource keyword. The recommendation module is used to recommend the first resource of the reference resource category contained in the independent website.

[0013] In one possible implementation, the determining module is used to call a large language model to perform semantic parsing on the search information to obtain the target scene information corresponding to the search information; Based on the correspondence between scene information and resource categories, the reference resource category corresponding to the target scene information is determined.

[0014] In one possible implementation, the determining module is used to call a large language model to perform semantic parsing on the search information to obtain the target scene information corresponding to the search information; If the search information contains resource keywords, the reference resource category corresponding to the target scene information and the resource keywords is determined based on the correspondence between scene information, keywords and resource categories.

[0015] In one possible implementation, the determining module is further configured to use the resources of the reference resource category contained in the independent website as candidate resources; Based on the operational metrics information of each candidate resource, a recommendation index is determined for each candidate resource. The operational metrics information includes at least one of the following: resource amount, resource click-through rate, resource conversion rate, resource sales volume, or resource add-to-cart volume. The first resource is determined from among the candidate resources based on the recommendation index of each candidate resource.

[0016] In one possible implementation, the acquisition module is further configured to acquire historical behavior data, which includes at least one of historical purchase data, historical browsing data, or historical add-to-cart data. The determining module is further configured to determine user attributes based on the historical behavior data; The determining module is used to determine the first resource among the candidate resources based on the user attributes and the recommendation index of each candidate resource.

[0017] In one possible implementation, the determining module is configured to: determine a first candidate resource among the candidate resources whose recommendation index is not lower than a recommendation index threshold; and determine a first resource among the first candidate resources whose resource attributes include the user attributes; or... Among the candidate resources, a second candidate resource is determined whose resource attributes include the user attributes; among the second candidate resources, a first resource with a recommendation index not lower than the recommendation index threshold is determined.

[0018] In one possible implementation, the recommendation index threshold is determined based on the number of resources contained in the independent website, and the recommendation index threshold is proportional to the number of resources contained in the independent website.

[0019] On the other hand, embodiments of this application provide a computer device, the computer device including a processor and a memory, the memory storing at least one piece of program code, the at least one piece of program code being loaded and executed by the processor to enable the computer device to implement any of the resource recommendation methods described above.

[0020] On the other hand, a computer-readable storage medium is also provided, wherein at least one piece of program code is stored therein, the at least one piece of program code being loaded and executed by a processor to enable a computer device to implement any of the resource recommendation methods described above.

[0021] On the other hand, a computer program or computer program product is also provided, wherein the computer program or computer program product stores at least one computer instruction, which is loaded and executed by a processor to enable a computer device to implement any of the above-described resource recommendation methods.

[0022] The technical solution provided in this application has at least the following beneficial effects: The technical solution provided in this application addresses the issue of resource keywords not being extracted from search information entered on an independent website, or resource keywords being extracted but no matching resources on the independent website. It matches the search information with reference resource categories, filters resources under those categories, and recommends them. This provides an orderly and user-centric recommendation scheme when precise retrieval fails, avoiding the problem of randomly recommending resources from the independent website that deviates from user needs. The recommended resources are more aligned with user requirements, resulting in a higher degree of matching between recommended resources and users, thereby increasing the click-through rate and conversion rate of the recommended resources, and ultimately increasing the total transaction volume of resources on the independent website. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a schematic diagram of a computer system provided in an embodiment of this application; Figure 2 This is a flowchart of a resource recommendation method provided in an embodiment of this application; Figure 3 This is a flowchart of a resource recommendation method provided in an embodiment of this application; Figure 4 This is a flowchart of a resource recommendation method provided in an embodiment of this application; Figure 5 This is a schematic diagram of the structure of a resource recommendation device provided in an embodiment of this application; Figure 6 This is a schematic diagram of the structure of a terminal device provided in an embodiment of this application; Figure 7 This is a schematic diagram of the structure of a server provided in an embodiment of this application. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0026] It should be noted that the terms "first," "second," etc., used in this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0027] Figure 1 This is a schematic diagram of a computer system provided in an embodiment of this application, such as... Figure 1 As shown, the computer system includes a terminal device 101 and a server 102. The terminal device 101 runs an independent website capable of resource recommendation. The terminal device 101 is used to execute the resource recommendation method provided in this embodiment, or to implement the resource recommendation method provided in this embodiment through interaction between the terminal device 101 and the server 102.

[0028] Independent websites, as the name suggests, are self-built e-commerce websites, distinct from third-party e-commerce platforms. They possess independent domain names and backends, and are not reliant on third-party trading platforms. The resource data, user data, page display logic, and transaction rules of an independent website are all managed independently by the website operator. Users access the independent website through a browser to complete operations such as searching, browsing, and placing orders, without relying on the client of a third-party trading platform. The resources included in an independent website are those of the merchant.

[0029] Server 102 provides background services for the client installed on terminal device 101. In one possible implementation, server 102 undertakes the main computing work, and terminal device 101 undertakes the secondary computing work. Alternatively, server 102 undertakes the secondary computing work, and terminal device 101 undertakes the main computing work. Or, terminal device 101 and server 102 collaborate on computing using a distributed computing architecture.

[0030] Optionally, terminal device 101 can be any electronic device product capable of human-computer interaction with a user through one or more methods such as a keyboard, touchpad, remote control, voice interaction, or handwriting device. For example, terminal device 101 can be a smartphone, tablet computer, laptop computer, desktop computer, smart speaker, smartwatch, PC (Personal Computer), mobile phone, PDA (Personal Digital Assistant), wearable device, PPC (Pocket PC), smart car system, smart TV, etc.

[0031] Terminal device 101 can refer to one of a plurality of terminal devices. This embodiment uses terminal device 101 as an example. Those skilled in the art will know that the number of terminal devices 101 can be more or less. For example, there may be only one terminal device 101, or there may be dozens or hundreds of terminal devices 101, or more. This application embodiment does not limit the number or type of terminal devices 101.

[0032] Server 102 can be a single server, a server cluster consisting of multiple servers, or any of the following: a cloud computing platform or a virtualization center. This embodiment of the application does not limit this. Server 102 communicates directly or indirectly with terminal device 101 via wired or wireless communication. Server 102 has data receiving, data processing, and data sending functions. Of course, server 102 may also have other functions, which are not limited in this embodiment of the application.

[0033] Those skilled in the art should understand that the terminal device 101 and server 102 described above are merely illustrative examples. Other existing or future terminal devices or servers that are applicable to this application should also be included within the scope of protection of this application, and are hereby incorporated by reference.

[0034] This application provides a resource recommendation method, which can be applied to the above-mentioned... Figure 1 The computer system shown is, Figure 2 The flowchart shown in this embodiment of the present application illustrates a resource recommendation method. This method can be implemented by... Figure 1 The method can be executed by terminal device 101, or it can be executed interactively by terminal device 101 and server 102. Taking the execution of this method by terminal device 101 as an example, such as... Figure 2 As shown, the method includes the following steps 201 to 203.

[0035] In step 201, the search information entered on the independent website is obtained.

[0036] In an exemplary embodiment of this application, a browser runs on the terminal device, and the display interface of the terminal device displays relevant information about the browser. This relevant information can be the browser icon, the browser name, or any other information that can uniquely identify the browser. This embodiment of the application does not limit the relevant information about the browser.

[0037] In response to a triggering operation related to browser information, the browser's interface is displayed, which includes a search box. In response to entering the URL of the independent website in the search box, the independent website's interface is displayed, which includes a search box and search controls. The independent website's interface may also display other information, which will not be elaborated upon in this embodiment.

[0038] In one possible implementation, when a user wants to purchase resources on an independent website, the user can enter search information in the search box displayed on the corresponding interface of the independent website. When the user triggers the search control, the terminal device obtains the search information entered on the independent website.

[0039] The search information can be either search text or search images; this application embodiment does not limit the scope of the search information. This application embodiment only uses search text as an example for illustration. When the search information is search text, the search text can be a declarative sentence or an interrogative sentence; this application embodiment does not limit the sentence structure of the search text.

[0040] For example, the search text entered by the user in the search box displayed on the interface of the independent website is: How to choose a gift for Mother's Day? For example, the search text entered by the user in the search box displayed on the interface of the independent website is: things suitable for taking photos of a picnic at the beach.

[0041] For example, the search text entered by the user in the search box displayed on the corresponding interface of the independent website is: bedroom atmosphere arrangement.

[0042] In step 202, if there are no resource keywords in the search information, determine the reference resource category corresponding to the search information; or, if there are resource keywords in the search information, and the resources contained in the independent website do not match the resource keywords, determine the reference resource category corresponding to the search information.

[0043] Specifically, a mismatch between resources and resource keywords within an independent website includes situations where: the resource attributes of the resources within the independent website do not include the resource keywords; or, the similarity between the resource attributes and resource keywords of the resources within the independent website is lower than a first similarity threshold. Resource attributes include at least one of the following: resource name or resource description information.

[0044] The first similarity threshold is set based on experience, or it can be flexibly adjusted according to the implementation environment; this application embodiment does not limit this. The first similarity threshold is directly proportional to the number of resources included in the independent website. That is, the more resources included in the independent website, the higher the first similarity threshold; conversely, the fewer resources included in the independent website, the lower the first similarity threshold.

[0045] The process of determining the similarity between resource attributes and resource keywords of resources contained in an independent website is described in the following content. Please refer to the following content for details. The embodiments of this application will not be repeated here.

[0046] In one possible implementation, the terminal device stores a resource keyword library, which includes multiple keywords. If any keyword from the resource keyword library is not found in the search results, it is determined that no resource keyword exists in the search results.

[0047] In one possible implementation, if the search results do not contain resource keywords, the reference resource category corresponding to the search results is determined using the first implementation method described below. If the search results contain resource keywords, but the resources contained in the independent website do not match the resource keywords, the reference resource category corresponding to the search results is determined using either the first or second implementation method described below.

[0048] The first implementation method involves calling the Large Language Model (LLM) to perform semantic parsing on the search information to obtain the target scene information corresponding to the search information; based on the correspondence between the scene information and the resource category, the reference resource category corresponding to the target scene information is determined.

[0049] Optionally, the process of calling a large language model to perform semantic parsing on the search information to obtain the target scene information corresponding to the search information includes: inputting the search information into the large language model, having the large language model perform semantic parsing on the search information; and using the output of the large language model as the target scene information corresponding to the search information.

[0050] For example, if the search query is "How to choose a gift for Mother's Day", then the target scenario for the search query is Mother's Day.

[0051] For example, if the search query is for things suitable for taking photos during a beach picnic, then the target scene information corresponding to the search query is "beach picnic" and "taking photos".

[0052] For example, if the search query is "bedroom atmosphere setup", then the target scene information corresponding to the search query is "bedroom, atmosphere setup".

[0053] In one possible implementation, the terminal device stores the correspondence between scene information and resource categories. Table 1 below is an exemplary table of the correspondence between scene information and resource categories provided in an embodiment of this application.

[0054] Table 1

[0055] As shown in Table 1 above, when the scenario information is Mother's Day, a gift for Mom, or Mom's birthday, the corresponding resource categories are household goods, massage equipment, personal care, aromatherapy, food and beverages, and tea gift boxes. When the scenario information is other, the corresponding resource categories are shown in Table 1 above, and will not be repeated here in this embodiment.

[0056] It should be noted that Table 1 above is only a correspondence between individual scenario information and resource categories provided in the embodiments of this application. There may be other scenario information and resource categories corresponding to that scenario information. The embodiments of this application do not limit the number of correspondences between scenario information and resource categories.

[0057] Optionally, the process of determining the reference resource category corresponding to the target scene information based on the correspondence between scene information and resource categories includes: taking the resource category corresponding to the target scene information in the correspondence between scene information and resource categories as the reference resource category.

[0058] For example, if the search information is "How to choose a gift for Mother's Day", and the target scenario information corresponding to the search information is Mother's Day, then the reference resource categories corresponding to the target scenario information are household goods, massage equipment, personal care, aromatherapy, food and tea, and tea gift boxes.

[0059] For example, if the search information is "things suitable for taking photos of a picnic at the beach", and the target scene information corresponding to the search information is "picnic at the beach and taking photos", then the reference resource categories corresponding to the target scene information are picnic mats, picnic baskets, and decorative lights.

[0060] For example, if the search information is "bedroom ambiance decoration", and the corresponding target scene information is "bedroom, ambiance decoration", then the reference resource categories corresponding to the target scene information are bedroom ambient lighting, starry sky lighting, decorative wall stickers, aromatherapy, and artificial flowers.

[0061] In the first implementation, when resource keywords are absent from the search information, or when resource keywords exist but the resources contained in the independent website do not match the resource keywords, a large language model is used to perform semantic parsing on the search information to obtain target scenario information. This is combined with a pre-defined correspondence between scenario information and resource categories to determine reference resource categories. This approach overcomes the limitations of traditional literal keyword matching, understands the user's contextual and descriptive query intent, avoids situations where no search results are found due to a lack of entity resource keywords, effectively improves the accuracy of query intent recognition, ensures that fallback recommendation results match the user's actual needs, and thus improves the shopping guide experience and the relevance of resource recommendations on the independent website.

[0062] The second implementation method involves calling a large language model to perform semantic parsing on the search text to obtain the target scene information corresponding to the search text. If there are resource keywords in the search information, the reference resource category corresponding to the target scene information and resource keywords is determined based on the correspondence between scene information, keywords and resource categories.

[0063] The process of calling a large language model to perform semantic parsing on the search text and obtaining the target scene information corresponding to the search information has been described in the first implementation method above. For details, please refer to the relevant content of the first implementation method above. This application embodiment will not repeat it here.

[0064] Optionally, the terminal device stores the correspondence between scene information, keywords and resource categories. After determining the target scene information corresponding to the search information and the resource keywords existing in the search text, the resource category corresponding to the target scene information and resource keywords in the correspondence between scene information, keywords and resource categories is used as the reference resource category.

[0065] In the second implementation, when resource keywords are extracted from the search information but no matching resources are found on the independent website, no results are returned directly. Instead, a large language model is called to perform semantic parsing to obtain target scenario information. At the same time, the target scenario information and resource keywords are combined, and the reference resource category is determined based on the correspondence between scenario information, keywords, and resource categories. This takes into account both the user's original resource needs and usage scenario intent, solves the problem of lost intent when no keywords are matched, and makes the fallback recommendation results more in line with the user's original query needs, improves the matching degree of the recommendation results, and optimizes the user experience of independent website search and shopping guide.

[0066] In step 203, the first resource in the reference resource category included in the independent website is recommended.

[0067] In one possible implementation, before recommending the first resource of the reference resource category included in the independent website, it is necessary to first determine the first resource among the resources included in the independent website. This application embodiment does not limit the process of determining the first resource among the resources included in the independent website.

[0068] Optionally, the process of determining the first resource among the resources contained in the independent website includes: taking the resources of the reference resource category contained in the independent website as candidate resources; and determining the first resource among the candidate resources.

[0069] An independent website contains resources that correspond to one or more resource categories. The resources contained in an independent website are those sold on the website.

[0070] First, resources belonging to the reference resource category within the independent website are filtered out as candidate resources. Then, the primary resource for recommendation is determined from the candidate resources. By limiting the candidate resources before selecting the primary resource, the scope of resource processing can be narrowed, the computational cost can be reduced, and the resources participating in the filtering can be ensured to match the reference resource category. This avoids irrelevant resources that are not related to the user's query intent, ensures the relevance of the output recommended resources to the user's needs, and improves the processing efficiency and recommendation effect of independent website resource recommendations.

[0071] In one possible implementation, the process of determining the first resource among the candidate resources includes: determining the recommendation index of each candidate resource based on the operational indicator information of each candidate resource, wherein the operational indicator information includes at least one of resource amount, resource click-through rate, resource conversion rate, resource sales or resource add-to-cart volume; and determining the first resource among the candidate resources based on the recommendation index of each candidate resource.

[0072] The operational metrics information of the candidate resources can be the operational metrics information of the candidate resources within a reference time period. The reference time period can be any time period, and this embodiment of the application does not limit it. For example, the reference time period is a period of 7 days with the current time as the end time. Another example is a period of time with the start time being the time the independent website begins operation and the end time being the current time.

[0073] Optionally, the process of determining the recommendation index of each candidate resource based on the operational indicator information of each candidate resource includes: for any candidate resource among the candidate resources, determining the recommendation index of any candidate resource based on the operational indicator information of any candidate resource and the weight parameters corresponding to the operational indicator information.

[0074] Optionally, the operational indicator information and the corresponding weight parameters of any candidate resource are weighted and summed to obtain the recommendation index of any candidate resource.

[0075] There are multiple operational metrics, and the sum of the weight parameters corresponding to each operational metric is 1, or the sum of the weight parameters corresponding to each operational metric is not 1.

[0076] For the selected candidate resources, a recommendation index is calculated by combining operational metrics such as resource value, click-through rate, conversion rate, sales volume, and add-to-cart volume. Based on the recommendation index, the top resource is determined. While ensuring that the resource category matches the user's query intent, the market performance and commercial value of the resources are comprehensively considered. This avoids relying solely on random category recommendations and prioritizes resources with better overall performance for push. It takes into account both user needs and the operational requirements of the independent website, further improving the rationality of the recommendation results. This is conducive to increasing the probability of resources being accepted by users and improving the overall shopping guide conversion effect.

[0077] Optionally, the process of determining the first resource among the candidate resources based on the recommendation index of each candidate resource includes: taking the candidate resources whose recommendation index is not lower than the recommendation index threshold as the first resource.

[0078] The recommendation index threshold is set based on experience, or it can be flexibly adjusted according to the implementation environment; this embodiment does not limit this. Optionally, the recommendation index threshold is directly proportional to the number of resources contained in the independent website. That is, the more resources the independent website contains, the higher the recommendation index threshold; conversely, the fewer resources the independent website contains, the lower the recommendation index threshold.

[0079] In one possible implementation, historical behavior data can also be obtained, including at least one of historical purchase data, historical browsing data, or historical add-to-cart data; user attributes are determined based on the historical behavior data; and a first resource is determined among the candidate resources based on the user attributes and the recommendation index of each candidate resource.

[0080] Historical behavior data refers to the historical behavior data of users who log in to the independent website. User attributes include, but are not limited to, sensitivity to new products, sensitivity to best-selling products, sensitivity to promotions, and sensitivity to price.

[0081] The process of determining user attributes based on historical behavioral data includes: if the number of resources belonging to new products in the historical behavioral data is the largest, then the user attribute is determined to be new product sensitive; if the number of resources belonging to best-selling products in the historical behavioral data is the largest, then the user attribute is determined to be best-selling product sensitive; if the number of resources belonging to promotional items in the historical behavioral data is the largest, then the user attribute is determined to be promotional sensitive; if the number of resources belonging to low-priced items in the historical behavioral data is the largest, then the user attribute is determined to be price sensitive.

[0082] By acquiring historical purchase data, browsing data, and add-to-cart data and generating corresponding user attributes, the system selects the first resource based on the recommendation index of candidate resources. This is combined with a comprehensive assessment of user attributes, taking into account both the resource's operational performance and individual user preferences. Instead of solely relying on resource dimensions for uniform ranking and recommendation, the system achieves differentiated resource selection for different users. This ensures that the final first resource aligns with the query intent of the resource category, the overall quality of the resource, and the user's past consumption preferences, enhancing the personalization of resource recommendations and further improving the suitability of the recommendation results for the current user. This contributes to improved user experience and conversion rates on independent websites.

[0083] In one possible implementation, the process of determining the first resource among the candidate resources based on user attributes and the recommendation index of each candidate resource includes: determining the first candidate resource among the candidate resources whose recommendation index is not lower than the recommendation index threshold; determining the first candidate resource whose resource attributes include user attributes; or, determining the second candidate resource among the candidate resources whose resource attributes include user attributes; and determining the first resource among the second candidate resources whose recommendation index is not lower than the recommendation index threshold.

[0084] After the first resource is determined in the above process, it is recommended. This embodiment of the application does not limit the process of recommending the first resource. Optionally, the process of recommending the first resource includes: displaying the resource information of the first resource to recommend it.

[0085] Resource information includes, but is not limited to, resource name, resource image, resource price, and resource description.

[0086] In one possible implementation, if resource keywords exist in the search information and resources contained in the independent website match the resource keywords, a second resource is determined from the resources contained in the independent website based on the resource keywords, and the resource attributes of the second resource match the resource keywords; the second resource is then recommended.

[0087] When resource keywords exist in the search results and matching resources exist on the independent website, the system directly determines and recommends a second matching resource based on the resource keywords. This prioritizes resource keyword retrieval and matching in scenarios where resources can be accurately matched, fully leveraging the high accuracy of resource keyword retrieval. It directly responds to users' explicit resource search needs, avoids unnecessary scenario category deductions, reduces computational costs associated with model calls, ensures query response efficiency, and implements a layered processing logic that prioritizes accurate intent matching and supplements it with fallback category recommendations. This balances retrieval accuracy with system processing performance, improving the overall query adaptability of the independent website shopping guide system.

[0088] In one possible implementation, the process of determining the second resource from the resources contained in the independent website based on the resource keywords includes: determining the third candidate resource whose resource attributes include the resource keywords from the resources contained in the independent website; determining the fourth candidate resource whose resource attributes and resource keywords have a similarity of not less than a first similarity threshold from the resources contained in the independent website; performing deduplication on the third and fourth candidate resources to obtain the fifth candidate resource; and determining the second resource from the fifth candidate resource.

[0089] Resources whose attributes directly contain the resource keywords are selected as the third candidate resources. Resources whose attributes and resource keywords have a similarity of up to the first similarity threshold are selected as the fourth candidate resources. After deduplication, the fifth candidate resources are obtained, and the second resource is determined from these. This approach retains accurate resources that completely match the resource keywords while also incorporating semantically similar fuzzy matching resources. This avoids missing relevant resources due to relying solely on literal string matching. By eliminating duplicate resources in the candidate set, the search accuracy is maintained while expanding the effective search coverage and enriching the pool of resources available for filtering. This ensures that the returned results not only match the resource keywords but also recall semantically related resources, improving the completeness of recall in keyword search scenarios and enhancing the search and shopping guide effect of independent websites.

[0090] It is also necessary to determine the similarity between the resource attributes and resource keywords of the resources contained in the independent website. This application embodiment does not limit the process of determining the similarity between the resource attributes and resource keywords of the resources contained in the independent website. Optionally, for any resource contained in the independent website, a first feature vector corresponding to the resource attribute of any resource is determined; a second feature vector corresponding to the resource keyword is determined; and the similarity between the resource attribute and resource keyword of any resource is determined based on the first feature vector and the second feature vector.

[0091] The process of determining the first feature vector corresponding to the resource attribute of any resource and the process of determining the second feature vector corresponding to the resource keyword are similar. This embodiment only illustrates the process of determining the first feature information corresponding to the resource attribute of any resource. Optionally, the terminal device includes a feature vector determination model. The resource attribute of any resource is input into the feature vector determination model; the output of the feature vector determination model is used as the first feature vector corresponding to the resource attribute of any resource.

[0092] The process of determining the similarity between resource attributes and resource keywords of any resource based on the first feature vector and the second feature vector includes: using the cosine similarity between the first feature vector and the second feature vector as the similarity between resource attributes and resource keywords of any resource.

[0093] For each resource within an independent website, resource attributes are transformed into a first feature vector, and resource keywords are transformed into a second feature vector. The similarity between resource attributes and resource keywords is calculated based on these two sets of feature vectors. This approach overcomes the limitation of traditional literal string matching, which can only identify completely identical text. It enables semantic-level similarity calculation, recognizing content with different wording but similar meanings. This improves the ability to identify synonyms and near-synonyms, effectively recalls semantically related resources, reduces the problem of missing related resources due to differences in wording, improves the accuracy and robustness of similarity determination during keyword retrieval, and optimizes the recall effect of independent website resource retrieval.

[0094] The process of determining the second resource from the fifth candidate resources is similar to the process of determining the first resource from the candidate resources described above, and will not be repeated here. The process of recommending the second resource is similar to the process of recommending the first resource described above, and will not be repeated here either.

[0095] In one possible implementation, when the search information is an image search, a first resource can be determined from the resources contained in the independent website and recommended as follows: The process of determining the first resource from the resources contained in the independent website is as follows: resources whose similarity to the search image is not lower than a second similarity threshold are selected as candidate resources, and the first resource is determined from the candidate resources.

[0096] The second similarity threshold is set based on experience, or can be flexibly adjusted according to the implementation environment; this embodiment does not limit this. Optionally, the second similarity threshold is proportional to the number of resources contained in the independent website. That is, the more resources the independent website contains, the higher the second similarity threshold; conversely, the fewer resources the independent website contains, the lower the second similarity threshold.

[0097] When the search query is for an image, the system identifies and recommends the first resource from among the available resources whose similarity to the search image reaches the second similarity threshold. It supports multimodal image search, breaking through the limitations of pure text search. This allows it to adapt to usage scenarios where users find it difficult to accurately describe the resources they want in words. By using image similarity to retrieve resources, it quickly finds the first resource with a similar appearance and style, enriching the query input formats for independent website guides, improving the effectiveness of resource retrieval in image search scenarios, further refining the overall resource recommendation system, and optimizing the user's search experience.

[0098] In one possible implementation, when the search information is a search for an image, the process of obtaining the search image entered in the independent website includes: displaying an image upload control in response to a trigger operation on the search box; displaying selectable images in response to a trigger operation on the image upload control, the selectable images being images included in the terminal device; and obtaining the search image entered in the independent website in response to a trigger operation on any of the selectable images, and upon receiving a trigger operation on the search control, the terminal device obtaining the search image, which is the triggered image.

[0099] The process of obtaining the search image entered in the independent website can also be as follows: in response to a trigger operation on the search box, display the camera control; in response to a trigger operation on the camera control, call the camera to capture an image; in response to a trigger operation on the search control, the terminal device obtains the search image entered in the independent website, and the search image is the captured image.

[0100] Before selecting resources from the independent website whose resource images have a similarity of at least a second similarity threshold with the search image as the first resource, it is necessary to determine the similarity between the resource images of the resources contained on the independent website and the search image. This process includes: for any resource contained on the independent website, determining the third feature vector corresponding to the resource image of that resource; determining the fourth feature vector corresponding to the search image; and determining the similarity between the resource image of that resource and the search image based on the third and fourth feature vectors.

[0101] The process of determining the third feature vector corresponding to the resource image of any resource and the process of determining the fourth feature vector corresponding to the search image are similar to the process of determining the first feature vector corresponding to the resource attribute of any resource, as described above, and will not be repeated here. The process of determining the similarity between the resource image of any resource and the search image based on the third and fourth feature vectors is similar to the process of determining the similarity between the resource attribute and resource keyword of any resource based on the first and second feature vectors, as described above, and will not be repeated here either.

[0102] Optionally, the process of determining the first resource from the candidate resources after selecting the resource images in the independent website whose similarity with the search image is not lower than the second similarity threshold has been described above, and will not be repeated here in the embodiments of this application.

[0103] The above method addresses situations where search information entered on an independent website fails to extract resource keywords, or where resource keywords are extracted but no matching resources exist on the independent website. By matching search information with reference resource categories and filtering resources under those categories for recommendation, it provides an orderly and user-centric recommendation solution when precise retrieval methods fail. This avoids randomly recommending resources from the independent website, which can cause the recommended resources to deviate from the user's needs. The recommended resources are more aligned with the user's requirements, resulting in a higher degree of matching and thus increasing the click-through rate and conversion rate of the recommended resources, ultimately boosting the total transaction volume of resources on the independent website.

[0104] This application provides a resource recommendation method, which can be applied to the above-mentioned... Figure 1 The computer system shown is, Figure 3 The flowchart shown in this embodiment of the present application illustrates a resource recommendation method. This method can be implemented by... Figure 1 The method can be executed by terminal device 101, or it can be executed interactively by terminal device 101 and server 102. Taking the interactive execution of this method by terminal device 101 and server 102 as an example, as follows... Figure 3 As shown, the method includes the following steps 301 to 309.

[0105] In step 301, the terminal device obtains the search information entered in the independent website.

[0106] In one possible implementation, the process by which the terminal device obtains the search information entered by the user on the independent website has been described in step 201 above. For details, please refer to the relevant content of step 201 above. This application embodiment will not repeat the description here.

[0107] In step 302, the terminal device sends search information to the server.

[0108] In one possible implementation, the terminal device and the server communicate via a wired or wireless network. After obtaining the search information, the terminal device sends the search information to the server.

[0109] Optionally, a search control is displayed on the terminal device. After receiving a trigger operation on the search control, the terminal device sends a search request to the server, which carries search information.

[0110] In step 303, if there are no resource keywords in the search information, the server determines the reference resource category corresponding to the search information; or, if there are resource keywords in the search information, and the resources contained in the independent website do not match the resource keywords, the server determines the reference resource category corresponding to the search information.

[0111] In one possible implementation, if the search information does not contain resource keywords, the process by which the server determines the reference resource corresponding to the search information is similar to the process by which the terminal device determines the reference resource category corresponding to the search information in step 202 above. See the relevant content of step 202 above for details, and this embodiment will not repeat it here. If the search information contains resource keywords, and the resources contained in the independent website do not match the resource keywords, the process by which the server determines the reference resource category corresponding to the search information is similar to the process by which the terminal device determines the reference resource category corresponding to the search information in step 202 above. See the relevant content of step 202 above for details, and this embodiment will not repeat it here.

[0112] In step 304, the server determines the resource category as the first resource of the reference resource category from the resources contained in the independent site.

[0113] In one possible implementation, the process by which the server determines the first resource with the reference resource category among the resources contained in the independent site is similar to the process by which the terminal device determines the first resource with the reference resource category among the resources contained in the independent site in step 203 above. For details, please refer to the relevant content of step 203 above. This application embodiment will not repeat the details here.

[0114] In step 305, the server sends the resource information of the first resource to the terminal device.

[0115] In one possible implementation, after the server identifies the first resource, it sends a search result to the terminal device, the search result carrying the resource information of the first resource.

[0116] In step 306, the terminal device receives the resource information of the first resource and displays the resource information of the first resource.

[0117] In one possible implementation, the process of the terminal device displaying the resource information of the first resource has been described in step 203 above. For details, please refer to the relevant content of step 203 above. This application embodiment will not repeat the description here.

[0118] In step 307, if the search information contains resource keywords and the independent website contains resources that match the resource keywords, the server determines the second resource that matches the resource keywords from the resources contained in the independent website.

[0119] In one possible implementation, if the search information contains resource keywords and the independent website contains resources that match the resource keywords, the process by which the server determines the second resource that matches the resource keywords from the resources contained in the independent website is similar to the process by which the terminal device determines the second resource from the resources contained in the independent website based on the resource keywords in step 203 above. For details, please refer to the relevant content of step 203 above. This application embodiment will not be described again here.

[0120] In step 308, the server sends the resource information of the second resource to the terminal device.

[0121] In one possible implementation, after identifying the second resource, the server sends a search result to the terminal device, the search result carrying the resource information of the second resource.

[0122] In step 309, the terminal device receives the resource information of the second resource and displays the resource information of the second resource.

[0123] In one possible implementation, the process of the terminal device displaying the resource information of the second resource has been described in step 203 above. For details, please refer to the relevant content of step 203 above. This application embodiment will not repeat the description here.

[0124] Figure 4 This is a flowchart of a resource recommendation method provided in an embodiment of this application, such as... Figure 4 As shown, the method includes the following.

[0125] Step 401: The user enters content on the independent website.

[0126] In one possible implementation, the process of a user entering content on an independent website has been described in step 201 above. Please refer to the relevant content of step 201 for details. This application embodiment will not repeat the description here.

[0127] Step 402: Determine whether the user entered text or an image search.

[0128] In one possible implementation, after obtaining the user's input, it is determined whether the user input is search text or search image. If the user input is search image, step 403 is executed; if the user input is search text, step 405 is executed.

[0129] This application does not limit the method for determining whether the user input is search text or search image. Optionally, if the user input includes an image, then the user input is determined to be a search image; if the user input does not include an image, then the user input is determined to be a search text.

[0130] Step 403: If the user enters a search image, determine the first resource from the resources contained in the independent website based on the search image.

[0131] In one possible implementation, the process of determining the first resource from the resources contained in the independent website based on the searched image has been described in step 203 above. For details, please refer to the relevant content of step 203. This application embodiment will not repeat the description here.

[0132] Step 404: Recommend the first resource to the user.

[0133] In one possible implementation, the process of recommending the first resource to the user has been described in step 203 above. Please refer to the relevant content of step 203 for details. This application embodiment will not repeat the description here.

[0134] Step 405: If the user enters search text, determine whether resource keywords exist in the search text.

[0135] In one possible implementation, the process of determining whether resource keywords exist in the search text has been described in step 202 above. Please refer to the relevant content of step 202 for details. This application embodiment will not repeat the description here.

[0136] Step 406: If the search text contains resource keywords, determine whether the independent website contains resources that match the resource keywords.

[0137] In one possible implementation, the process of determining whether there are resources in the independent website that match the resource keywords has been described in step 203 above. Please refer to the relevant content of step 203 for details. This application embodiment will not repeat the description here.

[0138] Step 407: If there are resources in the independent website that match the resource keywords, determine the second resource that matches the resource keywords from the resources in the independent website.

[0139] In one possible implementation, the process of determining the second resource that matches the resource keyword among the resources contained in the independent website has been described in step 203 above. For details, please refer to the relevant content of step 203. This application embodiment will not repeat the description here.

[0140] Step 408: Recommend a second resource to the user.

[0141] In one possible implementation, the process of recommending a second resource to the user has been described in step 203 above. Please refer to the relevant content of step 203 for details. This application embodiment will not repeat the description here.

[0142] Step 409: If there are no resource keywords in the search text, determine the target scenario information corresponding to the search text; or, if there are resource keywords in the search text, and the resources contained in the independent website do not match the resource keywords, determine the target scenario information corresponding to the search text.

[0143] In one possible implementation, the process of determining the target scene information corresponding to the search text has been described in step 202 above. For details, please refer to the relevant content of step 202. This application embodiment will not repeat the description here.

[0144] Step 410: Based on the correspondence between scene information and resource categories, determine the reference resource category corresponding to the target scene information.

[0145] In one possible implementation, the process of determining the reference resource category corresponding to the target scene information based on the correspondence between scene information and resource category has been described in step 202 above. For details, please refer to the relevant content of step 202. This application embodiment will not repeat the description here.

[0146] Step 411: Determine the resource category as the first resource of the reference resource category from the resources contained in the independent website.

[0147] In one possible implementation, the process of determining the first resource of the reference resource category among the resources contained in the independent site has been described in step 203 above. For details, please refer to the relevant content of step 203. This application embodiment will not repeat the description here.

[0148] Step 412: Recommend the first resource to the user.

[0149] In one possible implementation, the process of recommending the first resource to the user has been described in step 203 above. Please refer to the relevant content of step 203 for details. This application embodiment will not repeat the description here.

[0150] Figure 5 The diagram shown is a structural schematic of a resource recommendation device provided in an embodiment of this application. Figure 5 As shown, the device includes: The acquisition module 501 is used to acquire search information entered in the independent website; The determination module 502 is used to determine the reference resource category corresponding to the search information if there is no resource keyword in the search information, or to determine the reference resource category corresponding to the search information if there is a resource keyword in the search information and the resources contained in the independent website do not match the resource keyword. The recommendation module 503 is used to recommend the first resource in the reference resource category contained in the independent website.

[0151] In one possible implementation, module 502 is used to call a large language model to perform semantic parsing on the search information to obtain the target scene information corresponding to the search information; based on the correspondence between scene information and resource categories, the reference resource category corresponding to the target scene information is determined.

[0152] In one possible implementation, module 502 is used to call a large language model to perform semantic parsing on the search information to obtain the target scene information corresponding to the search information; if there are resource keywords in the search information, the reference resource category corresponding to the target scene information and resource keywords is determined based on the correspondence between scene information, keywords and resource categories.

[0153] In one possible implementation, the determining module 502 is further configured to select resources from the reference resource categories included in the independent website as candidate resources; determine the recommendation index of each candidate resource based on the operational indicator information of each candidate resource, wherein the operational indicator information includes at least one of resource amount, resource click-through rate, resource conversion rate, resource sales or resource add-to-cart volume; and determine the first resource among the candidate resources based on the recommendation index of each candidate resource.

[0154] In one possible implementation, the acquisition module 501 is further configured to acquire historical behavior data, which includes at least one of historical purchase data, historical browsing data, or historical add-to-cart data. The determination module 502 is also used to determine user attributes based on historical behavior data; The determination module 502 is used to determine the first resource among the candidate resources based on user attributes and the recommendation index of each candidate resource.

[0155] In one possible implementation, the determining module 502 is used to determine a first candidate resource whose recommendation index is not lower than the recommendation index threshold among the candidate resources; determine a first resource whose resource attributes include user attributes among the first candidate resources; or, determine a second candidate resource whose resource attributes include user attributes among the candidate resources; and determine a first resource whose recommendation index is not lower than the recommendation index threshold among the second candidate resources.

[0156] In one possible implementation, the recommendation index threshold is determined based on the number of resources contained in the independent website, and the recommendation index threshold is directly proportional to the number of resources contained in the independent website.

[0157] The aforementioned device addresses situations where search information entered on an independent website fails to extract resource keywords, or where resource keywords are extracted but no matching resources exist on the independent website. It matches the search information with reference resource categories, filters resources within those categories, and recommends them. This provides an orderly and user-centric recommendation solution when precise retrieval fails, avoiding the problem of randomly recommending resources from the independent website that deviates from user needs. The device ensures that recommended resources align with user requirements, resulting in a high degree of matching between recommended resources and users. This, in turn, increases the click-through rate and conversion rate of recommended resources, thereby increasing the total transaction volume of resources on the independent website.

[0158] It should be understood that the above-described apparatus is only illustrated by the division of the functional modules described above when implementing its functions. In practical applications, the functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and their specific implementation process can be found in the method embodiments, which will not be repeated here.

[0159] Figure 6 This illustration shows a structural block diagram of a terminal device 600 provided in an exemplary embodiment of this application. The terminal device 600 can be any electronic device product capable of human-computer interaction with a user through one or more methods such as a keyboard, touchpad, remote control, voice interaction, or handwriting device. Examples include PCs (Personal Computers), mobile phones, smartphones, PDAs (Personal Digital Assistants), wearable devices, PPCs (Pocket PCs), tablet computers, smart car systems, smart TVs, smart speakers, and smartwatches.

[0160] Typically, terminal device 600 includes a processor 601 and a memory 602.

[0161] Processor 601 may include one or more processing cores, such as a quad-core processor, an octa-core processor, etc. Processor 601 may be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). Processor 601 may also include a main processor and a coprocessor. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, processor 601 may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content required to be displayed on the screen. In some embodiments, processor 601 may also include an AI (Artificial Intelligence) processor, which is used to handle computational operations related to machine learning.

[0162] Memory 602 may include one or more computer-readable storage media, which may be non-transitory. Memory 602 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in memory 602 is used to store at least one instruction, which is executed by processor 601 to implement the resource recommendation method provided in the method embodiments of this application.

[0163] In some embodiments, the terminal device 600 may also optionally include a peripheral device interface 603 and at least one peripheral device. The processor 601, memory 602, and peripheral device interface 603 can be connected via a bus or signal line. Each peripheral device can be connected to the peripheral device interface 603 via a bus, signal line, or circuit board. Specifically, the peripheral device includes at least one of the following: a radio frequency circuit 604, a display screen 605, a camera assembly 606, an audio circuit 607, and a power supply 608.

[0164] Peripheral interface 603 can be used to connect at least one I / O (Input / Output) related peripheral device to processor 601 and memory 602. In some embodiments, processor 601, memory 602 and peripheral interface 603 are integrated on the same chip or circuit board; in some other embodiments, any one or two of processor 601, memory 602 and peripheral interface 603 can be implemented on separate chips or circuit boards, which is not limited in this embodiment.

[0165] The radio frequency (RF) circuit 604 is used to receive and transmit RF (Radio Frequency) signals, also known as electromagnetic signals. The RF circuit 604 communicates with communication networks and other communication devices via electromagnetic signals. The RF circuit 604 converts electrical signals into electromagnetic signals for transmission, or converts received electromagnetic signals back into electrical signals. Optionally, the RF circuit 604 includes: an antenna system, an RF transceiver, one or more amplifiers, a tuner, an oscillator, a digital signal processor, a codec chipset, a user identity module card, etc. The RF circuit 604 can communicate with other terminal devices through at least one wireless communication protocol. This wireless communication protocol includes, but is not limited to: the World Wide Web, metropolitan area networks, intranets, various generations of mobile communication networks (2G, 3G, 4G, and 5G), wireless local area networks, and / or WiFi (Wireless Fidelity) networks. In some embodiments, the RF circuit 604 may also include circuitry related to NFC (Near Field Communication), which is not limited in this application.

[0166] Display screen 605 is used to display a UI (User Interface). This UI may include graphics, text, icons, videos, and any combination thereof. When display screen 605 is a touch display screen, it also has the ability to collect touch signals on or above its surface. These touch signals can be input as control signals to processor 601 for processing. In this case, display screen 605 can also be used to provide virtual buttons and / or a virtual keyboard, also known as soft buttons and / or a soft keyboard. In some embodiments, there may be one display screen 605, disposed on the front panel of terminal device 600; in other embodiments, there may be at least two display screens, disposed on different surfaces of terminal device 600 or in a folded design; in still other embodiments, display screen 605 may be a flexible display screen, disposed on a curved or folded surface of terminal device 600. Furthermore, display screen 605 may be configured as a non-rectangular irregular shape, i.e., a non-rectangular screen. Display screen 605 may be made of materials such as LCD (Liquid Crystal Display) or OLED (Organic Light-Emitting Diode).

[0167] The camera assembly 606 is used to acquire images or videos. Optionally, the camera assembly 606 includes a front-facing camera and a rear-facing camera. Typically, the front-facing camera is located on the front panel of the terminal device 600, and the rear-facing camera is located on the back of the terminal device 600. In some embodiments, there are at least two rear-facing cameras, which are any one of a main camera, a depth-sensing camera, a wide-angle camera, and a telephoto camera, to achieve background blurring by fusion of the main camera and the depth-sensing camera, panoramic shooting by fusion of the main camera and the wide-angle camera, VR (Virtual Reality) shooting, or other fusion shooting functions. In some embodiments, the camera assembly 606 may also include a flash. The flash can be a single-color temperature flash or a dual-color temperature flash. A dual-color temperature flash is a combination of a warm-light flash and a cool-light flash, which can be used for light compensation at different color temperatures.

[0168] The audio circuit 607 may include a microphone and a speaker. The microphone is used to collect sound waves from the user and the environment, converting the sound waves into electrical signals that are input to the processor 601 for processing, or input to the radio frequency circuit 604 to achieve voice communication. For stereo sound acquisition or noise reduction purposes, multiple microphones may be used, each located at a different part of the terminal device 600. The microphone may also be an array microphone or an omnidirectional microphone. The speaker is used to convert electrical signals from the processor 601 or the radio frequency circuit 604 into sound waves. The speaker may be a conventional diaphragm speaker or a piezoelectric ceramic speaker. When the speaker is a piezoelectric ceramic speaker, it can convert electrical signals not only into audible sound waves but also into inaudible sound waves for purposes such as distance measurement. In some embodiments, the audio circuit 607 may also include a headphone jack.

[0169] Power supply 608 is used to supply power to the various components in terminal device 600. Power supply 608 can be AC ​​power, DC power, a disposable battery, or a rechargeable battery. When power supply 608 includes a rechargeable battery, the rechargeable battery can be a wired rechargeable battery or a wireless rechargeable battery. A wired rechargeable battery is a battery that is charged via a wired line, and a wireless rechargeable battery is a battery that is charged via a wireless coil. The rechargeable battery can also be used to support fast charging technology.

[0170] In some embodiments, the terminal device 600 further includes one or more sensors 609. The one or more sensors 609 include, but are not limited to, an accelerometer 610, a gyroscope 611, a pressure sensor 612, an optical sensor 613, and a proximity sensor 614.

[0171] Accelerometer 610 can detect the magnitude of acceleration along the three coordinate axes of a coordinate system established by terminal device 600. For example, accelerometer 610 can be used to detect the components of gravitational acceleration along the three coordinate axes. Processor 601 can control display screen 605 to display the user interface in either a landscape or portrait view based on the gravitational acceleration signal acquired by accelerometer 610. Accelerometer 610 can also be used for games or for acquiring user motion data.

[0172] The gyroscope sensor 611 can detect the orientation and rotation angle of the terminal device 600. The gyroscope sensor 611 can work in conjunction with the accelerometer sensor 610 to collect the user's 3D movements on the terminal device 600. Based on the data collected by the gyroscope sensor 611, the processor 601 can perform the following functions: motion sensing (e.g., changing the UI based on the user's tilt), image stabilization during shooting, game control, and inertial navigation.

[0173] The pressure sensor 612 can be disposed on the side bezel of the terminal device 600 and / or on the lower layer of the display screen 605. When the pressure sensor 612 is disposed on the side bezel of the terminal device 600, it can detect the user's grip signal on the terminal device 600, and the processor 601 can perform left / right hand recognition or quick operation based on the grip signal collected by the pressure sensor 612. When the pressure sensor 612 is disposed on the lower layer of the display screen 605, the processor 601 can control the operable controls on the UI interface based on the user's pressure operation on the display screen 605. The operable controls include at least one of button controls, scroll bar controls, icon controls, and menu controls.

[0174] An optical sensor 613 is used to collect ambient light intensity. In one embodiment, the processor 601 can control the display brightness of the display screen 605 based on the ambient light intensity collected by the optical sensor 613. Specifically, when the ambient light intensity is high, the display brightness of the display screen 605 is increased; when the ambient light intensity is low, the display brightness of the display screen 605 is decreased. In another embodiment, the processor 601 can also dynamically adjust the shooting parameters of the camera assembly 606 based on the ambient light intensity collected by the optical sensor 613.

[0175] The proximity sensor 614, also known as a distance sensor, is typically mounted on the front panel of the terminal device 600. The proximity sensor 614 is used to detect the distance between the user and the front of the terminal device 600. In one embodiment, when the proximity sensor 614 detects that the distance between the user and the front of the terminal device 600 is gradually decreasing, the processor 601 controls the display screen 605 to switch from a screen-on state to a screen-off state; when the proximity sensor 614 detects that the distance between the user and the front of the terminal device 600 is gradually increasing, the processor 601 controls the display screen 605 to switch from a screen-off state to a screen-on state.

[0176] Those skilled in the art will understand that Figure 6 The structure shown does not constitute a limitation on the terminal device 600, and may include more or fewer components than shown, or combine certain components, or use different component arrangements.

[0177] Figure 7This is a schematic diagram of the server structure provided in the embodiments of this application. The server 700 can vary considerably due to different configurations or performance. It may include one or more Central Processing Units (CPUs) 701 and one or more memories 702. The one or more memories 702 store at least one line of program code, which is loaded and executed by the one or more processors 701 to implement the resource recommendation methods provided in the various method embodiments described above. Of course, the server 700 may also have wired or wireless network interfaces, a keyboard, and input / output interfaces for input and output. The server 700 may also include other components for implementing device functions, which will not be elaborated here.

[0178] In an exemplary embodiment, a computer-readable storage medium is also provided, which stores at least one piece of program code that is loaded and executed by a processor to enable a computer device to implement any of the resource recommendation methods described above.

[0179] Optionally, the aforementioned computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), magnetic tape, floppy disk, and optical data storage device, etc.

[0180] In an exemplary embodiment, a computer program or computer program product is also provided, which stores at least one computer instruction that is loaded and executed by a processor to enable a computer device to implement any of the resource recommendation methods described above.

[0181] It should be noted that all information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, stored data, displayed data, etc.), and signals involved in this application have been authorized by the user or fully authorized by all parties, and the collection, use, and processing of related data must comply with the relevant laws, regulations, and standards of the relevant countries and regions. For example, the historical behavioral data and resource attributes involved in this application were obtained with full authorization.

[0182] It should be understood that "multiple" as used in this article refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0183] The above description is merely an exemplary embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the principles of this application should be included within the protection scope of this application.

Claims

1. A resource recommendation method, characterized in that, The method includes: Retrieve search information entered on an independent website; If the search information does not contain the resource keyword, determine the reference resource category corresponding to the search information; or, if the search information contains the resource keyword, and the resources contained in the independent website do not match the resource keyword, determine the reference resource category corresponding to the search information. The first resource in the reference resource category included in the independent website is recommended.

2. The method according to claim 1, characterized in that, Determining the reference resource category corresponding to the search information includes: The large language model is invoked to perform semantic parsing on the search information to obtain the target scene information corresponding to the search information; Based on the correspondence between scene information and resource categories, the reference resource category corresponding to the target scene information is determined.

3. The method according to claim 1, characterized in that, Determining the reference resource category corresponding to the search information includes: The large language model is invoked to perform semantic parsing on the search information to obtain the target scene information corresponding to the search information; If the search information contains resource keywords, the reference resource category corresponding to the target scene information and the resource keywords is determined based on the correspondence between scene information, keywords and resource categories.

4. The method according to any one of claims 1 to 3, characterized in that, Before recommending the first resource of the reference resource category included in the independent website, the method further includes: The resources in the reference resource category included in the independent website are used as candidate resources; Based on the operational metrics information of each candidate resource, a recommendation index is determined for each candidate resource. The operational metrics information includes at least one of the following: resource amount, resource click-through rate, resource conversion rate, resource sales volume, or resource add-to-cart volume. The first resource is determined from among the candidate resources based on the recommendation index of each candidate resource.

5. The method according to claim 4, characterized in that, The method further includes: Acquire historical behavior data, which includes at least one of historical purchase data, historical browsing data, or historical add-to-cart data; Based on the historical behavior data, user attributes are determined; The step of determining the first resource from among the candidate resources based on the recommendation index of each candidate resource includes: The first resource is determined from among the candidate resources based on the user attributes and the recommendation index of each candidate resource.

6. The method according to claim 5, characterized in that, The step of determining the first resource from among the candidate resources based on the user attributes and the recommendation index of each candidate resource includes: Among the candidate resources, a first candidate resource with a recommendation index not lower than a recommendation index threshold is determined; among the first candidate resources, a first resource whose resource attributes include the user attributes is determined; or... Among the candidate resources, a second candidate resource is determined whose resource attributes include the user attributes; among the second candidate resources, a first resource with a recommendation index not lower than the recommendation index threshold is determined.

7. The method according to claim 6, characterized in that, The recommendation index threshold is determined based on the number of resources contained in the independent website, and the recommendation index threshold is directly proportional to the number of resources contained in the independent website.

8. A resource recommendation device, characterized in that, The device includes: The acquisition module is used to retrieve search information entered on the independent website; The determination module is used to determine the reference resource category corresponding to the search information if the search information does not contain the resource keyword, or to determine the reference resource category corresponding to the search information if the search information contains the resource keyword and the resources contained in the independent website do not match the resource keyword. The recommendation module is used to recommend the first resource of the reference resource category contained in the independent website.

9. A computer device, characterized in that, The computer device includes a processor and a memory, the memory storing at least one piece of program code, the at least one piece of program code being loaded and executed by the processor to enable the computer device to implement the resource recommendation method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one piece of program code, which is loaded and executed by a processor to enable the computer to implement the resource recommendation method as described in any one of claims 1 to 7.

Images