Resource recommendation method and device, electronic equipment and medium

Abstract

The invention provides a resource recommendation method and device, electronic equipment, a computer readable storage medium and a computer program product, and relates to the field of artificial intelligence, in particular to the technical field of information recommendation and agents. According to the implementation scheme, first interest point information corresponding to a target user is obtained, wherein the first interest point information is determined based on at least one historical resource browsed by the target user; reference point-of-interest information is obtained, the reference point-of-interest information is determined based on at least one reference resource, and the reference resource and the historical resource are from different platforms; generating at least one piece of recommendation description information through the large language model, wherein the at least one piece of recommendation description information is information which is generated by the large language model according to the reference point-of-interest information and is associated with the first point-of-interest information; and based on the at least one piece of recommendation description information, determining at least one to-be-recommended resource for recommending to the target user, the at least one to-be-recommended resource being from the same platform as the historical resource.

Description

Technical Field

[0001] This disclosure relates to the field of artificial intelligence, and more particularly to the fields of information recommendation and intelligent agent technology, specifically to a resource recommendation method, apparatus, electronic device, computer-readable storage medium, and computer program product. Background Technology

[0002] Artificial intelligence (AI) is the study of enabling computers to simulate certain human thought processes and intelligent behaviors (such as learning, reasoning, thinking, and planning). It encompasses both hardware and software technologies. AI hardware technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, and big data processing. AI software technologies primarily include computer vision, speech recognition, natural language processing, machine learning / deep learning, big data processing, and knowledge graph technologies.

[0003] Resource recommendation is a service that relies on data mining and analysis techniques, combined with users' historical behavior and interests, to filter suitable content from massive resources and push it to target users. This solves the problem of information overload and improves the efficiency of users acquiring resources. With the continuous development of internet and artificial intelligence technologies, user needs are becoming increasingly personalized and diverse. Resource recommendation has become an indispensable part of modern society, and its application scenarios are becoming increasingly widespread. By reasonably controlling the content presented to users, user needs can be met. Summary of the Invention

[0004] This disclosure provides a resource recommendation method, apparatus, electronic device, computer-readable storage medium, and computer program product.

[0005] According to one aspect of this disclosure, a resource recommendation method is provided, comprising: acquiring first point of interest information corresponding to a target user, wherein the first point of interest information is determined based on at least one historical resource browsed by the target user; acquiring reference point of interest information, wherein the reference point of interest information is determined based on at least one reference resource, the at least one reference resource and the at least one historical resource are from different platforms; generating at least one recommendation description based on the first point of interest information and the reference point of interest information using a large language model, wherein the at least one recommendation description is information generated by the large language model based on the reference point of interest information and associated with the first point of interest information; and determining at least one resource to be recommended to the target user based on the at least one recommendation description, wherein the at least one resource to be recommended and the at least one historical resource are from the same platform.

[0006] According to another aspect of this disclosure, a resource recommendation apparatus is provided, comprising: a first acquisition module configured to acquire first point of interest information corresponding to a target user, wherein the first point of interest information is determined based on at least one historical resource browsed by the target user; a second acquisition module configured to acquire reference point of interest information, wherein the reference point of interest information is determined based on at least one reference resource, and the at least one reference resource and the at least one historical resource are from different platforms; a generation module configured to generate at least one recommendation description based on the first point of interest information and the reference point of interest information using a large language model, wherein the at least one recommendation description is information generated by the large language model based on the reference point of interest information and associated with the first point of interest information; and a determination module configured to determine at least one resource to be recommended to the target user based on the at least one recommendation description, wherein the at least one resource to be recommended and the at least one historical resource are from the same platform.

[0007] According to another aspect of this disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; the memory storing instructions executable by the at least one processor to enable the at least one processor to perform the methods described in this disclosure.

[0008] According to another aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions for causing a computer to perform the methods described in this disclosure.

[0009] According to another aspect of this disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the methods described in this disclosure.

[0010] According to one or more embodiments of this disclosure, by leveraging the generation capabilities of cross-platform interest data and large language models, it is possible to accurately capture users' dynamically changing interest needs, improve the matching degree between recommended resources and user preferences, and enrich the diversity and timeliness of recommended content.

[0011] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0012] The accompanying drawings exemplify embodiments and form part of the specification, serving together with the textual description to explain exemplary implementations of the embodiments. The illustrated embodiments are for illustrative purposes only and do not limit the scope of the claims. Throughout the drawings, the same reference numerals refer to similar but not necessarily identical elements.

[0013] Figure 1 A schematic diagram of an exemplary system in which the various methods described herein may be implemented according to embodiments of the present disclosure is shown; Figure 2 A flowchart illustrating a resource recommendation method according to an embodiment of the present disclosure is shown; Figure 3 A schematic diagram of a resource recommendation method according to an embodiment of the present disclosure is shown; Figure 4 A structural block diagram of a resource recommendation apparatus according to an embodiment of the present disclosure is shown; and Figure 5 A structural block diagram of an exemplary electronic device that can be used to implement embodiments of the present disclosure is shown. Detailed Implementation

[0014] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0015] In this disclosure, unless otherwise stated, the use of terms such as "first," "second," etc., to describe various elements is not intended to limit the positional, temporal, or importance relationships of these elements; such terms are merely used to distinguish one element from another. In some examples, the first element and the second element may refer to the same instance of that element, while in other cases, based on the context, they may refer to different instances.

[0016] The terminology used in the description of the various examples described in this disclosure is for the purpose of describing particular examples only and is not intended to be limiting. Unless the context explicitly indicates otherwise, an element may be one or more unless the number of elements is specifically limited. Furthermore, the term "and / or" as used in this disclosure covers any one of the listed items and all possible combinations thereof.

[0017] The embodiments of this disclosure will now be described in detail with reference to the accompanying drawings.

[0018] Figure 1A schematic diagram of an exemplary system 100 in which the various methods and apparatus described herein can be implemented according to embodiments of this disclosure is shown. Reference Figure 1 The system 100 includes one or more client devices 101, 102, 103, 104, 105 and 106, a server 120, and one or more communication networks 110 coupling the one or more client devices to the server 120. The client devices 101, 102, 103, 104, 105 and 106 can be configured to execute one or more applications.

[0019] In embodiments of this disclosure, server 120 may run one or more services or software applications that enable the execution of resource recommendation methods.

[0020] In some embodiments, server 120 may also provide other services or software applications, which may include non-virtual and virtual environments. In some embodiments, these services may be provided as web-based services or cloud services, such as to users of client devices 101, 102, 103, 104, 105 and / or 106 under a Software as a Service (SaaS) model.

[0021] exist Figure 1 In the configuration shown, server 120 may include one or more components that implement the functions performed by server 120. These components may include software components, hardware components, or combinations thereof that can be executed by one or more processors. Users operating client devices 101, 102, 103, 104, 105, and / or 106 can sequentially interact with server 120 using one or more client applications to utilize the services provided by these components. It should be understood that various different system configurations are possible and may differ from system 100. Therefore, Figure 1 This is an example of a system used to implement the various methods described herein, and is not intended to be limiting.

[0022] Users can use client devices 101, 102, 103, 104, 105, and / or 106 to obtain and browse recommended resources. The client devices can provide interfaces that allow users to interact with them. The client devices can also output information to the user through these interfaces. Although... Figure 1 Only six client devices are described, but those skilled in the art will understand that this disclosure can support any number of client devices.

[0023] Client devices 101, 102, 103, 104, 105, and / or 106 may include various types of computer devices, such as portable handheld devices, general-purpose computers (such as personal computers and laptops), workstation computers, wearable devices, smart screen devices, self-service terminal devices, service robots, gaming systems, thin clients, various messaging devices, sensors, or other sensing devices. These computer devices can run various types and versions of software applications and operating systems, such as Microsoft Windows, Apple iOS, UNIX-like operating systems, Linux or Linux-like operating systems (such as Google Chrome OS); or include various mobile operating systems, such as Microsoft Windows Mobile OS, iOS, Windows Phone, and Android. Portable handheld devices may include cellular phones, smartphones, tablets, personal digital assistants (PDAs), etc. Wearable devices may include head-mounted displays (such as smart glasses) and other devices. Gaming systems may include various handheld gaming devices, internet-enabled gaming devices, etc. Client devices are capable of executing various applications, such as various internet-related applications, communication applications (such as email applications), short message service (SMS) applications, and can use various communication protocols.

[0024] Network 110 can be any type of network well known to those skilled in the art, and can support data communication using any of a variety of available protocols (including but not limited to TCP / IP, SNA, IPX, etc.). By way of example only, one or more networks 110 can be a local area network (LAN), an Ethernet-based network, a token ring network, a wide area network (WAN), the Internet, a virtual network, a virtual private network (VPN), an intranet, an extranet, a blockchain network, a public switched telephone network (PSTN), an infrared network, a wireless network (e.g., Bluetooth, WIFI), and / or any combination of these and / or other networks.

[0025] Server 120 may include one or more general-purpose computers, special-purpose server computers (e.g., PC (personal computer) servers, UNIX servers, mid-range servers), blade servers, mainframe computers, server clusters, or any other suitable arrangement and / or combination. Server 120 may include one or more virtual machines running a virtual operating system, or other computing architectures involving virtualization (e.g., one or more flexible pools of logical storage devices that can be virtualized to maintain virtual storage devices for servers). In various embodiments, server 120 may run one or more services or software applications that provide the functionality described below.

[0026] The computing unit in server 120 can run one or more operating systems, including any of the aforementioned operating systems and any commercially available server operating system. Server 120 can also run any of a variety of additional server applications and / or middleware applications, including HTTP servers, FTP servers, CGI servers, JAVA servers, database servers, etc.

[0027] In some implementations, server 120 may include one or more applications to analyze and merge data feeds and / or event updates received from users of client devices 101, 102, 103, 104, 105, and 106. Server 120 may also include one or more applications to display data feeds and / or real-time events via one or more display devices of client devices 101, 102, 103, 104, 105, and 106.

[0028] In some implementations, server 120 can be a server for a distributed system or a server integrated with blockchain. Server 120 can also be a cloud server, or an intelligent cloud computing server or intelligent cloud host with artificial intelligence technology. A cloud server is a host product in the cloud computing service system, designed to address the shortcomings of traditional physical hosts and Virtual Private Server (VPS) services, such as high management difficulty and weak business scalability.

[0029] System 100 may also include one or more databases 130. In some embodiments, these databases may be used to store data and other information. For example, one or more of the databases 130 may be used to store semantic vectors, reference information resources, first point of interest information, etc. Databases 130 may reside in various locations. For example, a database used by server 120 may be local to server 120, or it may be located away from server 120 and may communicate with server 120 via a network-based or dedicated connection. Databases 130 may be of different types. In some embodiments, the database used by server 120 may be, for example, a relational database. One or more of these databases may store, update, and retrieve data from and from the databases in response to commands.

[0030] In some embodiments, one or more of the databases 130 may also be used by an application to store application data. The databases used by the application may be of different types, such as key-value stores, object stores, or regular stores supported by a file system.

[0031] Figure 1The system 100 can be configured and operated in various ways to enable the application of the various methods and apparatus described in this disclosure.

[0032] Typically, user profiles and consumption history can be used to filter and recommend resources within the site that are similar to the user profile and consumption history. However, the lack of real-time external knowledge supplementation (i.e., popular content / trends on other sites) makes it difficult to accurately capture the user's dynamically changing external interests, resulting in a mismatch between recommended content and user needs, and a poor user experience.

[0033] Therefore, an embodiment of this disclosure provides a resource recommendation method. Figure 2 A flowchart illustrating a resource recommendation method according to an embodiment of the present disclosure is shown, such as... Figure 2 As shown, method 200 includes: obtaining first point of interest information corresponding to a target user, wherein the first point of interest information is determined based on at least one historical resource browsed by the target user (step 210); obtaining reference point of interest information, wherein the reference point of interest information is determined based on at least one reference resource, and the at least one reference resource and the at least one historical resource come from different platforms (step 220); generating at least one recommendation description based on the first point of interest information and the reference point of interest information through a large language model, wherein the at least one recommendation description is information generated by the large language model based on the reference point of interest information and associated with the first point of interest information (step 230); and determining at least one resource to be recommended to the target user based on the at least one recommendation description, wherein the at least one resource to be recommended and the at least one historical resource come from the same platform (step 240).

[0034] In this disclosure, reference resources and historical resources come from different platforms. For example, historical resources viewed by the target user can be resources from the current platform. That is, the resource recommendation method described in this disclosure can be used to recommend resources to users on the current platform, therefore the historical resources viewed by the target user are resources from the current platform. Reference resources, however, can be resources from other platforms besides the current platform.

[0035] Therefore, according to the embodiments of this disclosure, by utilizing the generation capabilities of cross-platform popular resource data and large language models, it is possible to accurately capture the dynamic changes in users' interests and needs, improve the matching degree between recommended resources and user preferences, and enrich the diversity and timeliness of recommended content.

[0036] According to an embodiment of this disclosure, the first point of interest information includes: the vertical category corresponding to the historical resource, wherein obtaining the first point of interest information corresponding to the target user includes: determining at least one historical resource that the target user has browsed; determining the vertical category corresponding to each of the at least one historical resource; and determining a preset number of vertical categories with the highest browsing volume of the target user, as the first point of interest information.

[0037] In some embodiments, the first point of interest (POI) information can be determined based on the target user's historical resources viewed on the current platform, used to characterize the user's established interests and preferences. Reference POI information can, for example, come from reference resources provided by other platforms, supplementing the current platform's difficulty in capturing dynamic user interest signals by capturing real-time popular content and the interests and preferences of similar groups on other platforms. Specifically, the target user's historical resources viewed can be obtained, and based on a preset resource classification system, the corresponding vertical category for each historical resource can be determined. Then, the target user's view count for historical resources under each vertical category can be counted to filter out the vertical category with the highest preset number of views as the first POI information.

[0038] Therefore, by using vertical category statistics and high pageview volume filtering, we can accurately focus on user interests, avoid the generalization and dispersion of interest points, and improve the targeting of interest matching and the accuracy of recommended content.

[0039] According to an embodiment of this disclosure, obtaining first user attribute data corresponding to a target user, wherein generating at least one recommendation description based on the first point of interest information and the reference point of interest information through a large language model includes: generating at least one recommendation description based on the first user attribute information, the first point of interest information and the reference point of interest information through a large language model, wherein the at least one recommendation description is information generated by the large language model based on the reference point of interest information and associated with the first user attribute information and the first point of interest information.

[0040] In some embodiments, the first user attribute data can represent the basic characteristic information of the target user, such as a user profile. When generating recommendation description information, the first user attribute data, the first point of interest information, and the reference point of interest information can be jointly input into the large language model. The first point of interest information can be used to represent the target user's interest preferences formed based on historical resources, the reference point of interest information can be used to supplement real-time trending information from other platforms, and the first user attribute data can assist in interest matching.

[0041] Therefore, by introducing first-user attribute data, the integration of user interests and preferences, user attribute features and real-time popular reference interest points is achieved, which improves the accuracy of recommendation description information and optimizes the distribution effect and user experience of the recommendation system.

[0042] According to an embodiment of this disclosure, the first point of interest information includes text information, wherein obtaining reference point of interest information includes: determining the semantic vector corresponding to the text information; and retrieving the reference point of interest information from a preset reference information library based on the semantic vector, wherein the reference information library includes multiple reference information and vectors corresponding to each of the multiple reference information, wherein the reference point of interest information is determined based on reference information similar to the text information among the multiple reference information.

[0043] According to embodiments of this disclosure, the text information is obtained based on the following operations: acquiring the main text content of the corresponding historical resources; and summarizing the main text content through a large language model to obtain the text information output by the large language model.

[0044] In some embodiments, the text information in the first point of interest information can be used to characterize the semantic features corresponding to the resources browsed historically by the target user. This can be the complete text content of the historical resources browsed by the target user, such as scripts extracted from historical videos or descriptions of cooking steps in food resources; or it can be a summary text obtained by using a large language model to specifically summarize the text content. The aforementioned text information can be converted into a semantic vector through feature encoding using, for example, a semantic embedding model. This semantic vector can then be used as the retrieval basis to perform similarity searches in a pre-defined reference information database.

[0045] In some embodiments, the reference information repository can store a massive amount of reference information from other platforms. For example, it can acquire popular resources from other platforms, such as resources on trending topics lists. Based on the acquired resources from other platforms, the reference information repository can be generated, where each piece of reference information in the repository can be determined based on the corresponding resource. For example, data such as the title and / or cover image of each resource can be acquired as the reference information corresponding to that resource. In some examples, information obtained by summarizing the main text content of the corresponding resource through a large language model can also be used as the reference information corresponding to that resource.

[0046] Furthermore, in some embodiments, the reference information database may also include a vector corresponding to each piece of reference information. During the retrieval process, vector similarity, such as cosine similarity, can be calculated between the first point of interest information and each piece of reference information in the reference information database to filter out reference information similar to the first point of interest. Then, reference point of interest information can be determined based on these similar reference information. For example, reference information in the reference information database (such as the title, body text, etc. corresponding to the relevant reference resource) can be directly used as reference point of interest information, or text information (text used to describe reference information of image types) converted from reference information in the reference information database can be used as reference point of interest information.

[0047] Therefore, by utilizing large language models to generate text information, transform semantic vectors, and perform similarity retrieval, cross-platform information matching was achieved, significantly improving the accuracy and suitability of reference information related to user interests.

[0048] According to embodiments of this disclosure, the semantic vector includes a first semantic vector and a second semantic vector. Determining the semantic vector corresponding to the text information includes: obtaining the first semantic vector corresponding to the text information through a semantic embedding model; and obtaining the second semantic vector corresponding to the text information through a multimodal model. The reference information library includes text reference information and image reference information, wherein the text reference information generates a corresponding first vector through the semantic embedding model, and the image reference information generates a corresponding first vector through the multimodal model. The dimension of the first semantic vector is greater than the dimension of the second semantic vector.

[0049] In some embodiments, when determining the text semantic vector corresponding to the first point of interest information, multi-dimensional feature extraction can be achieved through dual-model co-coding. That is, firstly, a semantic embedding model can be used to parse the text information, transforming the text content into a high-dimensional first semantic vector. Simultaneously, a multimodal model can be used to parse the text information to generate a second semantic vector with a relatively low dimension. A pre-defined reference information library can synchronously support multimodal data storage. The text reference information within it can also be encoded using the aforementioned semantic embedding model to generate corresponding vectors, and the image reference information can be encoded using the same multimodal model to generate corresponding vectors, ensuring that the vectors in the reference information library are encoded logically consistent with the semantic vectors generated from the target user's first point of interest information. Then, the first semantic vector can be matched with the text reference information vectors in the reference information library, and the second semantic vector can be matched with the image reference information vectors in the reference information library, respectively, to achieve cross-modal retrieval of similar reference information, thereby comprehensively determining the reference point of interest information.

[0050] In some embodiments, the first point of interest information and the reference point of interest information of the text type can be directly written into the prompt so that at least one recommendation description can be obtained by inputting the prompt into a large language model.

[0051] Therefore, by using a dual-model, bidirectional, multimodal coding and retrieval design, the limitations of single text or image features are overcome, cross-modal reference resource matching is achieved, and the retrieval accuracy and user experience of cross-platform reference point of interest information are improved.

[0052] According to embodiments of this disclosure, the plurality of reference information in the reference information database is determined based on the following operations: obtaining a first vector corresponding to each of the plurality of reference information and second user attribute information corresponding to each of the plurality of reference information, wherein the second user attribute information is used to characterize the user characteristics of the reference resources corresponding to the browsed reference information; clustering the plurality of reference information based on the first vector corresponding to each of the plurality of reference information to obtain a plurality of clusters; determining a third user attribute information corresponding to each of the plurality of clusters based on the second user attribute information, wherein the third user attribute information is used to characterize the user statistical characteristics corresponding to each of the clusters; and determining a preset number of clusters in the plurality of clusters based on the third user attribute information, so as to determine the plurality of reference information according to the preset number of clusters.

[0053] In some embodiments, the first vector of the reference information may refer to a text vector encoded by a semantic embedding model or an image vector encoded by a multimodal model. The second user attribute information may refer to user characteristics of users who have viewed the reference resources corresponding to the reference information, reflecting the user attributes adapted to the corresponding reference resources. For example, the second user attribute information may be the percentage of users with a certain preset user attribute who have viewed the corresponding reference resources, such as 51% based on user IP addresses belonging to region XX. Therefore, the third user attribute information may also be the percentage of users with a certain preset user attribute who have viewed the corresponding reference resources within the cluster.

[0054] Then, specifically, all the pre-acquired reference information can be clustered based on the first vector corresponding to the reference information to obtain multiple reference information clusters. Afterwards, for each reference information cluster, the second user attribute information corresponding to all reference information within the cluster can be statistically analyzed to form third user attribute information characterizing the statistical features of the users in that cluster.

[0055] In some examples, the third user attribute information can be determined based on the following method: the third user attribute information of a cluster = the sum of the exposure of reference resources corresponding to reference information with the same second user attribute information / the sum of the exposure of reference resources corresponding to all reference information within the cluster. Thus, after statistically analyzing all second user attribute information within the cluster, the corresponding third user attribute information can be determined. For example, in a cluster, the second user attribute corresponding to reference information 1 is: 51% of users with IP addresses from region XX, 49% of users with IP addresses from region CC, and the exposure of the reference resource corresponding to reference information 1 is 1000; the second user attribute corresponding to reference information 2 is: 31% of users with IP addresses from region XX, 69% of users with IP addresses from region CC, and the exposure of the reference resource corresponding to reference information 2 is 1200. Then, the third user attribute information corresponding to this cluster can be calculated as: the percentage of users with IP addresses from region XX: (51%) 1000+31% (1200) / (1000 + 1200) = 40.1%; the percentage of users with IP addresses in the CC region: (49%) 1000+69% (1200) / (1000 + 1200) = 59.9%. Therefore, based on this third user attribute information, a preset number of clusters can be determined from multiple clusters, such as filtering clusters where the percentage of users with IP addresses from region XX is greater than 50%. Thus, based on the reference information in the determined preset number of clusters, this reference information database is generated.

[0056] Therefore, by combining attribute association and clustering filtering, the matching degree between reference point of interest information and the interests of users to be recommended is improved, as well as the retrieval efficiency, thereby further improving the recommendation effect and enhancing the user experience.

[0057] According to embodiments of this disclosure, determining at least one resource to be recommended to the target user based on the at least one recommendation description information includes: obtaining a second vector corresponding to each of the at least one recommendation description information and a third vector corresponding to each of the plurality of resources to be recommended; and determining, based on the second vector and the third vector, at least one resource to be recommended that is similar to the at least one recommendation description information among the plurality of resources to be recommended.

[0058] In some embodiments, the second vector corresponding to the recommendation description information can be a high-dimensional semantic vector generated by encoding each recommendation description information through the semantic embedding model described above. The third vector corresponding to multiple resources to be recommended can be a vector obtained by pre-encoding the features of all candidate resources within the current platform. After obtaining the second and third vectors, their semantic correlation can be quantified by calculating their similarity. For example, if the similarity between the second vector of a certain recommendation description information and the third vector of a resource to be recommended on this platform is higher than a preset threshold, it can be determined that the resource on this platform is highly related to the recommendation description information. Based on this similarity screening result, resources similar to the recommendation description information can be selected from multiple resources to be recommended as the final recommended resources to the target user.

[0059] In the above embodiments, the third vector may be the same as or different from the semantic vector corresponding to the first point of interest information. For example, the third vector may be the semantic vector corresponding to the first point of interest information.

[0060] Therefore, by using a unified standard vector encoding and similarity matching mechanism, the semantic alignment between the recommendation description information and the resources within the platform is ensured, the accuracy of resource selection is improved, and users' dynamic interests across platforms are effectively transformed into directly distributable resources within the platform, further enhancing the personalized adaptation effect of the recommendation system.

[0061] Figure 3 A schematic diagram of a resource recommendation method according to an embodiment of the present disclosure is shown. Figure 3 As shown, the text information (i.e., the first point of interest information) corresponding to the historical resources browsed by the user in Platform 1 can generate a first semantic vector and a second semantic vector respectively through a semantic embedding model and a multimodal model. Correspondingly, the image reference information and text reference information corresponding to the reference resources in Platform 2 can each generate a first vector through their respective models. By calculating the similarity between vectors, reference point of interest information can be filtered out from multiple reference information. Based on the reference point of interest information, text information, and first user attribute information, at least one recommendation description can be obtained through a large language model. Therefore, based on this at least one recommendation description, the resources to be recommended to the user can be determined in Platform 1.

[0062] According to embodiments of this disclosure, such as Figure 4As shown, a resource recommendation device 400 is also provided, comprising: a first acquisition module 410 configured to acquire first point of interest information corresponding to a target user, wherein the first point of interest information is determined based on at least one historical resource browsed by the target user; a second acquisition module 420 configured to acquire reference point of interest information, wherein the reference point of interest information is determined based on at least one reference resource, and the at least one reference resource and the at least one historical resource come from different platforms; a generation module 430 configured to generate at least one recommendation description based on the first point of interest information and the reference point of interest information through a large language model, wherein the at least one recommendation description is information generated by the large language model based on the reference point of interest information and associated with the first point of interest information; and a determination module 440 configured to determine at least one resource to be recommended to the target user based on the at least one recommendation description, wherein the at least one resource to be recommended and the at least one historical resource come from the same platform.

[0063] Here, the operation of each of the above units 410 to 440 of the resource recommendation device 400 is similar to the operation of steps 210 to 240 described above, and will not be repeated here.

[0064] The collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0065] According to embodiments of this disclosure, an electronic device, a readable storage medium, and a computer program product are also provided.

[0066] refer to Figure 5 The present invention describes a structural block diagram of an electronic device 500 that can serve as a server or client of the present disclosure, which is an example of a hardware device that can be applied to various aspects of the present disclosure. The electronic device is intended to represent various forms of digital electronic computer devices, 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 processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0067] like Figure 5As shown, the electronic device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. The RAM 503 may also store various programs and data required for the operation of the electronic device 500. The computing unit 501, ROM 502, and RAM 503 are interconnected via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0068] Multiple components in electronic device 500 are connected to I / O interface 505, including: input unit 506, output unit 507, storage unit 508, and communication unit 509. Input unit 506 can be any type of device capable of inputting information to electronic device 500. Input unit 506 can receive input digital or character information and generate key signal inputs related to user settings and / or function control of electronic device, and may include, but is not limited to, a mouse, keyboard, touchscreen, trackpad, trackball, joystick, microphone, and / or remote control. Output unit 507 can be any type of device capable of presenting information, and may include, but is not limited to, a monitor, speaker, video / audio output terminal, vibrator, and / or printer. Storage unit 508 may include, but is not limited to, disk and optical disk. Communication unit 509 allows electronic device 500 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks, and may include, but is not limited to, modems, network cards, infrared communication devices, wireless communication transceivers, and / or chipsets, such as Bluetooth devices, 802.11 devices, WiFi devices, WiMax devices, cellular communication devices, and / or the like.

[0069] The computing unit 501 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 501 performs the various methods and processes described above, such as method 200. For example, in some embodiments, method 200 may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 508. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 500 via ROM 502 and / or communication unit 509. When the computer program is loaded into RAM 503 and executed by the computing unit 501, one or more steps of method 200 described above may be performed. Alternatively, in other embodiments, the computing unit 501 may be configured to perform method 200 by any other suitable means (e.g., by means of firmware).

[0070] Various embodiments 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-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0071] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0072] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0073] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, 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 sound input, voice input, or tactile input).

[0074] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), the Internet, and blockchain networks.

[0075] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0076] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0077] While embodiments or examples of this disclosure have been described with reference to the accompanying drawings, it should be understood that the methods, systems, and devices described above are merely exemplary embodiments or examples, and the scope of the invention is not limited by these embodiments or examples, but only by the granted claims and their equivalents. Various elements in the embodiments or examples may be omitted or replaced by their equivalents. Furthermore, the steps may be performed in a different order than that described in this disclosure. Further, various elements in the embodiments or examples may be combined in various ways. Importantly, as the technology evolves, many elements described herein can be replaced by equivalents that appear after this disclosure.

Claims

1. A resource recommendation method, comprising: Obtain the first point of interest information corresponding to the target user, wherein the first point of interest information is determined based on at least one historical resource browsed by the target user; Obtain reference point of interest information, wherein the reference point of interest information is determined based on at least one reference resource, and the at least one reference resource and the at least one historical resource come from different platforms; Based on the first point of interest information and the reference point of interest information, at least one recommended description is generated through a large language model, wherein the at least one recommended description is information generated by the large language model based on the reference point of interest information and associated with the first point of interest information; and Based on the at least one recommendation description, at least one resource to be recommended to the target user is determined, wherein the at least one resource to be recommended comes from the same platform as the at least one historical resource.

2. The method as described in claim 1, wherein, Obtaining the first user attribute data corresponding to the target user, wherein, based on the first point of interest information and the reference point of interest information, generating at least one recommendation description through a large language model includes: Based on the first user attribute information, the first point of interest information, and the reference point of interest information, at least one recommended description is generated by a large language model, wherein the at least one recommended description is information generated by the large language model based on the reference point of interest information and associated with the first user attribute information and the first point of interest information.

3. The method of claim 1, wherein, The first point of interest information includes text information, wherein obtaining reference point of interest information includes: Determine the semantic vector corresponding to the text information; and Based on the semantic vector, the reference interest point information is retrieved from a preset reference information database. The reference information database includes multiple reference information and a first vector corresponding to each of the multiple reference information. The multiple reference information is determined based on multiple reference resources, and the reference interest point information is determined based on reference information similar to the text information among the multiple reference information.

4. The method of claim 3, wherein, The semantic vector includes a first semantic vector and a second semantic vector, wherein determining the semantic vector corresponding to the text information includes: The first semantic vector corresponding to the text information is obtained through a semantic embedding model; and The second semantic vector corresponding to the text information is obtained through a multimodal model, and The reference information database includes text reference information and image reference information. The text reference information generates a corresponding first vector through the semantic embedding model, and the image reference information generates a corresponding first vector through the multimodal model. The dimension of the first semantic vector is greater than the dimension of the second semantic vector.

5. The method as described in claim 3 or 4, wherein, The plurality of reference information in the reference information database is determined based on the following operations: Obtain a first vector corresponding to each of the multiple reference information and a second user attribute information corresponding to each of the multiple reference information, wherein the second user attribute information is used to characterize the user characteristics of the reference resource corresponding to the browsed reference information; Based on the first vector corresponding to each of the multiple reference information, the multiple reference information is clustered to obtain multiple clusters; Based on the second user attribute information, the third user attribute information corresponding to each of the plurality of clusters is determined, wherein the third user attribute information is used to characterize the user statistical characteristics corresponding to each cluster. as well as Based on the third user attribute information, a preset number of clusters are determined from the plurality of clusters, so as to determine the plurality of reference information according to the preset number of clusters.

6. The method of claim 3, wherein, The text information was obtained based on the following operations: To obtain the main text content of the corresponding historical resources; and The main text content is summarized using a large language model to obtain the text information output by the large language model.

7. The method according to any one of claims 1-4, wherein, The first point of interest information includes: the vertical category corresponding to the historical resources, wherein obtaining the first point of interest information corresponding to the target user includes: Identify at least one historical resource that the target user has viewed; Determine the vertical category corresponding to each of the at least one historical resource; and The vertical category with the highest number of views from the target user is determined as the first point of interest information.

8. The method of claim 1, wherein, Based on the at least one recommendation description, at least one resource to be recommended to the target user is determined, including: Obtain the second vector corresponding to each of the at least one recommendation description and the third vector corresponding to each of the multiple resources to be recommended; and Based on the second vector and the third vector, at least one resource to be recommended that is similar to the at least one recommendation description is determined from the plurality of resources to be recommended.

9. A resource recommendation device, comprising: The first acquisition module is configured to acquire first point of interest information corresponding to the target user, wherein the first point of interest information is determined based on at least one historical resource browsed by the target user. The second acquisition module is configured to acquire reference point of interest information, wherein the reference point of interest information is determined based on at least one reference resource, and the at least one reference resource and the at least one historical resource come from different platforms; The generation module is configured to generate at least one recommended description based on the first point of interest information and the reference point of interest information using a large language model, wherein the at least one recommended description is information generated by the large language model based on the reference point of interest information and associated with the first point of interest information; and The determination module is configured to determine at least one resource to be recommended to the target user based on the at least one recommendation description information, wherein the at least one resource to be recommended comes from the same platform as the at least one historical resource.

10. An electronic device, comprising: At least one processor; as well as A memory that is communicatively connected to the at least one processor; in The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-8.

11. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-8.

12. A computer program product comprising a computer program, wherein, When the computer program is executed by a processor, it implements the method of any one of claims 1-8.

Images