Information recommendation methods, devices, electronic equipment and media

Abstract

This disclosure provides an information recommendation method, apparatus, electronic device, computer-readable storage medium, and computer program product, relating to the field of artificial intelligence, and particularly to the fields of data processing, intelligent recommendation, and deep learning. The implementation scheme is as follows: For each consumption time period among multiple consumption time periods, information consumed by a first user is obtained; a first feature vector for each piece of information, a second feature vector for the first user, and a third feature vector for the information to be recommended are determined; for each consumption time period, the first feature vector within that consumption time period is input into a preset self-attention layer to obtain a fourth feature vector corresponding to that consumption time period; the first feature vectors are combined and input into a preset neural network model to obtain a predicted value, wherein the first feature vector combination includes: the fourth feature vector, the second feature vector, and the third feature vector; based on the predicted value, it is determined whether to recommend the information to be recommended to the first user.

Description

Technical Field

[0001] This disclosure relates to the field of artificial intelligence, and in particular to the fields of data processing, intelligent recommendation, and deep learning technology, specifically to an information 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] With the development of the internet and the increasing popularity of various online services and applications, while people enjoy the convenience brought by the growth in information, the explosive growth of information has also led to serious information overload. Recommendation systems can analyze users' historical behavior, such as browsing and purchasing records, to infer users' preferences and needs, and then recommend suitable items to users based on this information. Summary of the Invention

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

[0005] According to one aspect of this disclosure, an information recommendation method is provided, comprising: for each of a plurality of consumption time periods, obtaining a first information identifier of information consumed by a first user within that consumption time period; determining a first feature vector corresponding to each of the first information identifiers and a second feature vector corresponding to a first user identifier of the first user; determining a third feature vector corresponding to the second information identifier of the information to be recommended; for each consumption time period, inputting each of the first feature vectors obtained within that consumption time period into a preset self-attention layer to obtain an output fourth feature vector corresponding to that consumption time period; combining the first feature vectors and inputting them into a preset neural network model to obtain an output predicted value, wherein the predicted value is used to represent the degree of interest of the first user in the information to be recommended, wherein the combination of the first feature vectors includes: the fourth feature vector, the second feature vector, and the third feature vector; and determining whether to recommend the information to be recommended to the first user based on the predicted value.

[0006] According to another aspect of this disclosure, a model training method is provided, comprising: for each of a plurality of consumption time periods, obtaining a first information identifier of information consumed by a first user within that consumption time period; determining a first feature vector corresponding to each of the first information identifiers, and a second feature vector corresponding to a first user identifier corresponding to the first user; determining a third feature vector corresponding to the second information identifier of information to be recommended, and preset tag data for determining whether to recommend the information to be recommended to the first user; for each consumption time period, inputting each of the first feature vectors obtained within that consumption time period into a preset self-attention layer to obtain an output fourth feature vector corresponding to that consumption time period; inputting a combination of the first feature vectors into a neural network model to obtain a predicted value, wherein the predicted value is used to represent the degree of interest of the first user in the information to be recommended, wherein the combination of the first feature vectors includes: the fourth feature vector, the second feature vector, and the third feature vector; determining a loss value based on the predicted value and the tag data using a preset loss function; and training the neural network model based on the loss value.

[0007] According to another aspect of this disclosure, an information recommendation device is provided, comprising: a first acquisition unit configured to acquire, for each of a plurality of consumption time periods, a first information identifier of information consumed by a first user within that consumption time period; a first determination unit configured to determine a first feature vector corresponding to each of the first information identifiers and a second feature vector corresponding to a first user identifier of the first user; a second determination unit configured to determine a third feature vector corresponding to the second information identifier of information to be recommended; a first input unit configured to, for each consumption time period, input each of the first feature vectors acquired within that consumption time period into a preset self-attention layer to obtain an output fourth feature vector corresponding to that consumption time period; a second input unit configured to input a combination of the first feature vectors into a preset neural network model to obtain an output predicted value, wherein the predicted value is used to represent the degree of interest of the first user in the information to be recommended, wherein the combination of the first feature vectors includes: the fourth feature vector, the second feature vector, and the third feature vector; and a recommendation unit configured to determine, based on the predicted value, whether to recommend the information to be recommended to the first user.

[0008] According to another aspect of this disclosure, a model training apparatus is provided, comprising: a second acquisition unit configured to acquire, for each of a plurality of consumption time periods, a first information identifier of information consumed by a first user within that consumption time period; a third determination unit configured to determine a first feature vector corresponding to each of the information identifiers, and a second feature vector corresponding to a first user identifier of the first user; a fourth determination unit configured to determine a third feature vector corresponding to the second information identifier of information to be recommended, and preset tag data for determining whether to recommend the information to be recommended to the first user; a third input unit configured to, for each of the consumption time periods, input each of the first feature vectors acquired within that consumption time period into a preset self-attention layer to obtain an output fourth feature vector corresponding to that consumption time period; a fourth input unit configured to input a combination of the first feature vectors into a neural network model to obtain a predicted value, wherein the predicted value is used to represent the degree of interest of the first user in the information to be recommended, wherein the combination of the first feature vectors includes: a fourth feature vector, a second feature vector, and a third feature vector; a calculation unit configured to determine a loss value based on the predicted value and the tag data using a preset loss function; and an adjustment unit configured to train the neural network model based on the loss value.

[0009] 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.

[0010] 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.

[0011] 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.

[0012] According to one or more embodiments of this disclosure, by extracting user consumption information within multiple consumption time periods, the user historical profile is expanded, and feature extraction of the user historical profile for each time period is achieved through a self-attention layer, thereby solving the problem of long-term user interest forgetting and improving the accuracy of information recommendation.

[0013] 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

[0014] 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.

[0015] 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;

[0016] Figure 2 A flowchart of an information recommendation method according to an embodiment of the present disclosure is shown;

[0017] Figure 3 A schematic diagram illustrating information recommendations according to embodiments of the present disclosure is shown;

[0018] Figure 4 A flowchart illustrating model training according to an embodiment of the present disclosure is shown;

[0019] Figure 5 A structural block diagram of an information recommendation device according to an embodiment of the present disclosure is shown;

[0020] Figure 6 A structural block diagram of a model training apparatus according to an embodiment of the present disclosure is shown; and

[0021] Figure 7 A structural block diagram of an exemplary electronic device that can be used to implement embodiments of the present disclosure is shown. Detailed Implementation

[0022] 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.

[0023] 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.

[0024] 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.

[0025] The embodiments of this disclosure will now be described in detail with reference to the accompanying drawings.

[0026] Figure 1 A 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.

[0027] In embodiments of this disclosure, server 120 may run one or more services or software applications that enable methods for performing information recommendation or model training.

[0028] 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.

[0029] 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.

[0030] Users can use client devices 101, 102, 103, 104, 105, and / or 106 to obtain recommendation information. The client devices can provide an interface that allows users to interact with the client devices. The client devices can also output information to the users through this interface. 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.

[0031] 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.

[0032] Network 110 can be any type of network well known to those skilled in the art, and can use any of a variety of available protocols (including but not limited to TCP / IP, SNA, IPX, etc.) to support data communication. 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.

[0033] 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.

[0034] 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.

[0035] 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.

[0036] 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.

[0037] 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 information such as user information, recommendations, consumed 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.

[0038] 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.

[0039] Figure 1 The system 100 can be configured and operated in various ways to enable the application of the various methods and apparatus described in this disclosure.

[0040] Typically, ranking models in recommendation systems are designed with features based on user information and the content to be recommended. They also employ a training mode that decays long-term features during model training. This makes it easier for the model to remember short-term interests and gradually forget the user's long-term interests, causing the user's interests to become trapped in a cocoon.

[0041] Therefore, an information recommendation method is provided according to embodiments of the present disclosure. Figure 2 A flowchart of an information recommendation method according to an embodiment of the present disclosure is shown, such as... Figure 2As shown, method 200 includes: for each of a plurality of consumption time periods, obtaining a first information identifier of information consumed by a first user within that consumption time period (step 210); determining a first feature vector corresponding to each first information identifier and a second feature vector corresponding to the first user identifier of the first user (step 220); determining a third feature vector corresponding to the second information identifier of the information to be recommended (step 230); for each consumption time period, inputting each of the first feature vectors obtained within that consumption time period into a preset self-attention layer to obtain an output fourth feature vector corresponding to that consumption time period (step 240); inputting the first feature vector combination into a preset neural network model to obtain an output prediction value, wherein the prediction value is used to represent the degree of interest of the first user in the information to be recommended, wherein the first feature vector combination includes: the fourth feature vector, the second feature vector, and the third feature vector (step 250); and determining whether to recommend the information to be recommended to the first user based on the prediction value (step 260).

[0042] According to embodiments of this disclosure, by extracting user consumption information within multiple consumption time periods, the user's historical profile is expanded, and feature extraction of the user's historical profile for each time period is achieved through a self-attention layer, thereby solving the problem of users forgetting long-term interests and improving the accuracy of information recommendation.

[0043] In this disclosure, information can be any suitable information for user consumption, such as text information, image information, audio information, video information, graphic and text information, product information, etc., without any limitation.

[0044] In this disclosure, "consumption" can refer to the action or operation of a user acquiring information through a media. The system recommends information to users through this media. For example, when the system pushes videos, images, or text to a user through a user interface, it is considered to be recommending such information; once a user browses such information through the user interface, the information viewed can be referred to as consumed information. Alternatively, when the system pushes product information to a user through an e-commerce platform, it is considered to be recommending such product information; once a user purchases a product through the e-commerce platform, the information about the purchased product can be referred to as consumed information.

[0045] In some examples, user historical consumption information can be obtained, which has a unique content ID that identifies the information. In some examples, the information identifier may include at least one of the following: content ID, primary category, secondary category, creator ID, tag information, etc., without limitation.

[0046] In some examples, the primary category, secondary category, and tag information are used to identify the information respectively. For example, for a picture and text post about food preparation, its primary category could be "food," its secondary category could be "food preparation," and its tag information could be "Sichuan cuisine," "recipe," etc.

[0047] In this disclosure, multiple consumption time periods can be determined according to the consumption history. For example, based on the user's consumption history, the consumption time periods can be categorized in ascending order of recent to distant past as short-term consumption time periods, medium-term consumption time periods, and long-term consumption time periods. For instance, a short-term consumption time period can be represented as the last 7 days, a medium-term consumption time period as the period from 7 days ago to the previous 3 months, and a long-term consumption time period as the period from 3 months ago to the previous 12 months.

[0048] In some embodiments, in step 210, for each of the multiple consumption time periods, the first information identifier of one or more pieces of information consumed by the first user during that consumption time period is obtained.

[0049] According to some embodiments, obtaining the first information identifier of each of one or more pieces of information consumed by the first user during the consumption period includes: for each of a plurality of information categories, obtaining the first information identifier of each of one or more pieces of information consumed by the first user during the consumption period and under that information category.

[0050] In some examples, the information category can be either a first-level category or a second-level category. A second-level category refers to a more specific subcategory under a first-level category. For example, on an art website, there might be a first-level category "Art," and then several second-level categories under this level, such as "Painting," "Sculpture," and "Photography." In this case, "Painting" is a second-level category of the first-level category "Art."

[0051] According to some embodiments, the consumed information refers to one or more pieces of information that the first user is most interested in. The first user's interest in the consumed information is determined based on at least one of the following: the cumulative number of clicks on the information by the first user within the corresponding consumption time period, and the maximum single consumption duration of the information by the first user within the consumption time period.

[0052] Understandably, the information users are most interested in best reflects their consumption profile. In some examples, the number of clicks a user makes on a given piece of information within a specific consumption period can be used to characterize their level of interest. For instance, more clicks indicate a higher level of interest.

[0053] In some examples, a user's level of interest in the information can be characterized by the longest single consumption period for that information within that consumption timeframe. Here, "single consumption" can refer to a single consumption process, such as a user browsing an article, which is considered a single consumption process, and the browsing time is the single consumption duration; when a user exits the article browsing interface and then re-enters the article browsing interface, it is considered a new consumption process.

[0054] In some embodiments, for information such as video or audio, the maximum single consumption duration of the information by a first user within the consumption period can be determined based on both the user's browsing time and the physical duration of the video or audio information. For example, the maximum single consumption duration can be normalized to the range [0,1], such as by determining the ratio of the user's browsing time to its physical duration. If the user's browsing time is the same as its physical duration, then it is 1; if the user's browsing time is 0, then it is 0.

[0055] In some examples, a user's level of interest in the information can be characterized by both the cumulative number of clicks made by the user within a given consumption period and the maximum duration of a single click within that period. For instance, the user's level of interest can be obtained by multiplying the cumulative number of clicks by the maximum duration of a single click.

[0056] It is understandable that the degree of interest of the first user in the relevant information can be determined in any suitable manner, and no limitation is made here.

[0057] In some embodiments, in step 240, for each consumption time period, the first feature vector corresponding to each piece of information acquired within that time period is input into a preset self-attention layer to obtain multiple fourth feature vectors corresponding one-to-one with the multiple consumption cycles. According to some embodiments, the first user identifier includes a marker slot. Further, the method may also include: obtaining one or more second user identifiers by changing the marker slot of the first user identifier, wherein the number of user identifiers in the combination formed by the one or more second user identifiers and the first user identifier is the same as the number of the multiple consumption time periods; determining a fifth feature vector corresponding to each of the one or more second user identifiers. In this case, the first feature vector combination further includes: the fifth feature vector corresponding to each of the second user identifiers.

[0058] For example, suppose the obtained first user identifier is {240136699}:1, where the "1" after the ":" represents its tag slot information. By changing the tag slot, one or more second user identifiers can be obtained, such as {240136699}:20001 and {240136699}:30001. That is to say, the only difference between the first user identifier and the second user identifier is the tag slot information.

[0059] By using the aforementioned mirrored user identifier operation, each consumption time period corresponds to a specific user identifier, enabling subsequent user characteristics to represent the user's profile in each consumption time period, thereby enhancing the user's interest characteristics in each period.

[0060] According to some embodiments, the first feature vector combination further includes: a cross feature vector, wherein the cross feature vector is determined based on the first user identifier and the tag information corresponding to the information to be recommended. The tag information is determined based on the content of the information to be recommended.

[0061] In the above embodiments, the tag information is determined based on the content of the information to be recommended. For example, for a text and image post introducing food preparation, the tag information could be: Sichuan cuisine, recipe, etc. This tag information can be multi-dimensional, including tags based on theme, topic, entity, etc., to cover key information in information such as articles.

[0062] In some examples, the tag information may be extracted using keyword extraction techniques or deep learning techniques, and there are no restrictions on this.

[0063] By designing cross-features, the fit between the first user and the information to be recommended is further evaluated, thereby improving the accuracy of subsequent predictions.

[0064] According to some embodiments, the preset neural network model includes: a densely connected network input layer and a first fully connected layer. Inputting the first feature vector combination into the preset neural network model to obtain a predicted value includes: inputting the first feature vector combination into the densely connected network input layer to obtain a sixth feature vector; and obtaining the predicted value based on the sixth feature vector and passing it through the first fully connected layer.

[0065] In some embodiments, obtaining the predicted value based on the sixth feature vector and passing it through the first fully connected layer includes: inputting the sixth feature vector into the first fully connected layer to obtain the output predicted value. In some examples, when multiple first fully connected layers exist, the sixth feature vector is sequentially input into the one or more first fully connected layers to obtain the predicted value.

[0066] In some examples, a user's level of interest in a recommended message can be determined by its completion rate. Completion rate is an important metric for measuring the attractiveness of the recommended message content and the quality of user viewing. For example, completion rate can be the percentage of the predicted content a user is likely to watch. For instance, for a text / image message, if a user is likely to watch half of it, the completion rate is 50%. For a video message, completion rate can be determined based on the ratio of the predicted user viewing time to the video's physical duration; if the user viewing time equals the physical duration, the completion rate is 1.

[0067] In some examples, the densely connected network input layer and the first fully connected layer can together form a densely connected network. In this case, based on the core idea of ​​dense connections in densely connected networks, when multiple first fully connected layers are included, the input of the corresponding fully connected layer contains not only the output of the previous layer but also the outputs of all preceding layers in the densely connected network. Through densely connected networks, gradient vanishing is mitigated, feature propagation is enhanced, thus utilizing features more effectively and reducing the number of parameters to some extent.

[0068] According to some embodiments, the preset neural network model further includes: a gate network and a second fully connected layer. The method according to this disclosure may further include: sequentially inputting a second feature vector combination into the gate network and the second fully connected layer to obtain an output seventh feature vector. In this case, obtaining the predicted value based on the sixth feature vector and through the first fully connected layer may further include: performing feature fusion on the sixth and seventh feature vectors to obtain an eighth feature vector; and inputting the eighth feature vector into the first fully connected layer to obtain the output predicted value. The second feature vector combination includes: other feature vectors in the first feature vector combination besides the feature vector determined based on the information to be recommended.

[0069] In the above embodiments, the second feature vector combination includes all feature vectors in the first feature vector combination except for the feature vectors determined based on the information to be recommended. Specifically, when the first feature vector combination includes the plurality of fourth feature vectors, the second feature vector, and the third feature vector, the second feature vector combination includes the plurality of fourth feature vectors and the second feature vector. When the first feature vector combination includes the plurality of fourth feature vectors, the second feature vector, the third feature vector, and the fifth feature vector corresponding to each second user identifier (or further includes a cross feature vector), the second feature vector combination includes the plurality of fourth feature vectors, the second feature vector, and the fifth feature vector corresponding to each second user identifier.

[0070] In some embodiments, the gate network may include one or more fully connected layers. Alternatively, the gate network may include one or more fully connected layers, a sigmoid function layer, and / or a ReLU function layer.

[0071] Through the above embodiments, user characteristics and consumption profile features of users in different consumption periods are enhanced. This achieves the addition of personalized bias structures to the consumption profile features of users in different consumption periods, and then the output features are connected to the main network (i.e., fused with the output features of the input layer of the densely connected network), thereby enhancing the user's interest in different consumption periods and enabling accurate recommendations to users.

[0072] Figure 3 A schematic diagram illustrating information recommendations according to embodiments of the present disclosure is shown, such as... Figure 3 As shown, s1, s2…s5k represent the first feature vectors of the first user's consumed information in each consumption time period. The first feature vectors of user consumption information in the long-term, short-term, and medium-term periods are input into the serialization feature extraction module (i.e., the self-attention layer) to obtain three fourth feature vectors. The second feature vector corresponding to the first user (uid-1-0), along with the two fifth feature vectors obtained by mirroring (uid-1-1, uid-1-2), the third feature vector of the information to be recommended (nid), the cross feature vector, and the three fourth feature vectors are input into the input layer of the Dense network to obtain the sixth feature vector. The three fourth feature vectors, the second feature vector corresponding to the first user (uid-1-0), and the two fifth feature vectors obtained by mirroring (uid-1-1, uid-1-2) are sequentially input into the gate network and fully connected layer of the personalized bias structure to obtain the seventh feature vector. After feature fusion (such as inner product operation) of the sixth and seventh feature vectors, the output eighth feature vector is sequentially input into the fully connected layer of the Dense network to obtain the output prediction value. This allows the system to determine whether to recommend the information to the first user based on the predicted value.

[0073] According to embodiments of this disclosure, such as Figure 4As shown, a model training method 400 is also provided, comprising: for each of multiple consumption time periods, obtaining a first information identifier of information consumed by a first user within that consumption time period (step 410); determining a first feature vector corresponding to each first information identifier and a second feature vector corresponding to the first user identifier of the first user (step 420); determining a third feature vector corresponding to the second information identifier of the information to be recommended and preset tag data for determining whether to recommend the information to the first user (step 430); for each consumption time period, inputting each first feature vector obtained within that consumption time period into a preset self-attention layer to obtain an output fourth feature vector corresponding to that consumption time period (step 440); inputting the combination of the first feature vectors into a neural network model to obtain a predicted value, wherein the predicted value is used to represent the degree of interest of the first user in the information to be recommended, wherein the combination of the first feature vectors includes: the fourth feature vector, the second feature vector, and the third feature vector (step 450); determining a loss value based on the predicted value and the tag data using a preset loss function (step 460); and training the neural network model based on the loss value (step 470).

[0074] In some embodiments, the level of interest a first user has in the recommended information can be determined by the completion rate of the recommended information. The completion rate is an important indicator for measuring the attractiveness of the recommended information content and the quality of user viewing. For example, the completion rate can be the percentage of the information content that the first user is predicted to watch. For instance, for a text and image message, if the user is likely to watch half of the message, the completion rate is 50%. For example, for a video message, the completion rate can be determined based on the ratio of the predicted user viewing time to the physical duration of the video; if the user viewing time equals the physical duration, the completion rate is 1. Therefore, in some embodiments, the tag data can be a preset completion rate for the recommended information.

[0075] In some embodiments, in step 410, for each of the multiple consumption time periods, the first information identifier of one or more pieces of information consumed by the first user during that consumption time period is obtained.

[0076] According to some embodiments, obtaining the first information identifier of each of one or more pieces of information consumed by the first user during the consumption period includes: for each of a plurality of information categories, obtaining the first information identifier of each of one or more pieces of information consumed by the first user during the consumption period and under that information category.

[0077] According to some embodiments, the consumed information is one or more pieces of information that the first user is most interested in, wherein the first user's interest in the consumed information is determined based on at least one of the following: the cumulative number of clicks on the information by the first user within the corresponding consumption time period, and the maximum single consumption duration of the information by the first user within the consumption time period.

[0078] In some embodiments, in step 440, for each consumption time period, the first feature vector corresponding to each piece of information acquired within that time period is input into a preset self-attention layer, thereby obtaining multiple fourth feature vectors that correspond one-to-one with the multiple consumption cycles.

[0079] According to some embodiments, the first user identifier includes a marker slot, and the method further includes: obtaining one or more second user identifiers by changing the marker slot of the first user identifier, wherein the number of user identifiers in the combination formed by the one or more second user identifiers and the first user identifier is the same as the number of the plurality of consumption time periods; determining a fifth feature vector corresponding to each of the one or more second user identifiers; wherein the first feature vector combination further includes: the fifth feature vector corresponding to each of the second user identifiers.

[0080] According to some embodiments, the first feature vector combination further includes: a cross feature vector, wherein the cross feature vector is determined based on the first user identifier and the tag information corresponding to the information to be recommended, wherein the tag information is determined based on the content of the information to be recommended.

[0081] According to some embodiments, the neural network model includes: a densely connected network input layer and a first fully connected layer, wherein combining the first feature vectors and inputting them into the neural network model to obtain a predicted value includes: combining the first feature vectors and inputting them into the densely connected network input layer to obtain a sixth feature vector; and obtaining the predicted value based on the sixth feature vector and passing it through the first fully connected layer.

[0082] In some embodiments, obtaining the predicted value based on the sixth feature vector and passing it through the first fully connected layer includes: inputting the sixth feature vector into the first fully connected layer to obtain the output predicted value. In some examples, when multiple first fully connected layers exist, the sixth feature vector is sequentially input into the one or more first fully connected layers to obtain the predicted value.

[0083] According to some embodiments, the neural network model further includes a gate network and a second fully connected layer. The method further includes sequentially inputting a second feature vector combination into the gate network and the second fully connected layer to obtain an output seventh feature vector. Obtaining the predicted value based on the sixth feature vector and through the first fully connected layer may further include: fusing the sixth and seventh feature vectors to obtain an eighth feature vector; and inputting the eighth feature vector into the first fully connected layer to obtain the predicted value. The second feature vector combination includes all feature vectors in the first feature vector combination other than the feature vector determined based on the information to be recommended.

[0084] According to some embodiments, training the neural network model based on the loss value includes: adjusting the parameters of the neural network model using multiple optimizers based on the loss value, wherein the number of multiple optimizers is the same as the number of multiple consumption time periods.

[0085] According to some embodiments, the plurality of consumption time periods include: short-term consumption time periods, medium-term consumption time periods, and long-term consumption time periods determined according to consumption history. The plurality of optimizers includes: an optimizer with a decay factor and an optimizer with an automatically decaying learning rate.

[0086] Therefore, according to some embodiments, the plurality of optimizers includes: RMSProp optimizer, AdaGrad optimizer, and Adam optimizer.

[0087] Understandably, optimizers with decay factors converge faster, which helps to emphasize the importance of recent samples. Optimizers with automatic learning rate decay designs have higher initial learning rates, converge faster, and become smoother in the mid-term, thus making little difference in the utilization of mid-term and recent samples, which helps to remember mid-term interests. Furthermore, the Adam optimizer is used to warm up the model and utilize longer-term samples. The Adam optimizer is currently a relatively stable and widely used optimizer.

[0088] It is understandable that in the backpropagation parameter update process, different optimizers have different decay factors (or learning rates) designed. By obtaining the update parameter values ​​corresponding to their respective feature vectors through different optimizers, the user features controlled by multiple optimizers can express user interests in different periods.

[0089] According to some embodiments, determining a loss value based on the predicted value and the label data using a preset loss function includes: determining a first loss value based on the predicted value and the label data using a preset first loss function; and determining a second loss value based on the second feature vector and the fifth feature vector using a preset second loss function. Adjusting the parameters of the neural network model based on the loss value includes: training the neural network model based on the first loss value and the second loss value to minimize the first loss value and make the second loss value greater than a preset threshold.

[0090] In some examples, when there are multiple second user identifiers, a second loss value is determined based on the second feature vector and the fifth feature vectors corresponding to the multiple second user identifiers, using a preset second loss function.

[0091] As described above, since the second feature vector and the fifth feature vector corresponding to each second user identifier are both determined based on the first user identifier, in order to achieve further differentiation, a vector distance constraint is added between the feature vectors to make the distance between the feature vectors as large as possible, thereby realizing the separate representation of user interests across multiple consumption time periods.

[0092] According to some embodiments, the preset second loss function is determined based on the following formula:

[0093] ||HH T -I|| F →Ω

[0094] Where H represents the vector matrix composed of the second and fifth eigenvectors, I represents the identity matrix, ||·|| F Ω represents the matrix norm, and Ω represents the preset threshold.

[0095] In some examples, when there are multiple second user identifiers, H represents a vector matrix composed of the second feature vector and the fifth feature vectors corresponding to the multiple second user identifiers. That is, if there are two second user identifiers, the second feature vector and the two fifth feature vectors (each corresponding to a corresponding second user identifier) ​​form the vector matrix H.

[0096] It is understood that the form of the loss function described above is merely illustrative and is not intended to be limiting.

[0097] In this disclosure, the model trained by the model training method described in any of the above embodiments can be used to implement the information recommendation method described in any of the embodiments of this disclosure.

[0098] Here, the embodiments for implementing the model training method and the embodiments for implementing the information recommendation method are similar in operation and feature, and will not be described again here.

[0099] According to embodiments of this disclosure, such as Figure 5 As shown, an information recommendation device 500 is also provided, comprising: a first acquisition unit 510 configured to acquire, for each of a plurality of consumption time periods, a first information identifier of information consumed by a first user within that consumption time period; a first determination unit 520 configured to determine a first feature vector corresponding to each of the first information identifiers and a second feature vector corresponding to a first user identifier of the first user; a second determination unit 530 configured to determine a third feature vector corresponding to the second information identifier of the information to be recommended; a first input unit 540 configured to, for each of the consumption time periods, input each of the first feature vectors acquired within that consumption time period into a preset self-attention layer to obtain an output fourth feature vector corresponding to that consumption time period; a second input unit 550 configured to input a combination of the first feature vectors into a preset neural network model to obtain an output predicted value, wherein the predicted value is used to represent the degree of interest of the first user in the information to be recommended, wherein the combination of the first feature vectors includes: the fourth feature vector, the second feature vector, and the third feature vector; and a recommendation unit 560 configured to determine, based on the predicted value, whether to recommend the information to be recommended to the first user.

[0100] Here, the operation of each of the above-mentioned units 510 to 560 of the information recommendation device 500 is similar to the operation of steps 210 to 260 described above, and will not be repeated here.

[0101] According to embodiments of this disclosure, such as Figure 6As shown, a model training device 600 is also provided, including: a second acquisition unit 610, configured to acquire, for each of a plurality of consumption time periods, a first information identifier of information consumed by a first user within that consumption time period; a third determination unit 620, configured to determine a first feature vector corresponding to each of the information identifiers, and a second feature vector corresponding to the first user identifier of the first user; a fourth determination unit 630, configured to determine a third feature vector corresponding to the second information identifier of the information to be recommended, and preset tag data for determining whether to recommend the information to be recommended to the first user; and a third input unit 640, configured to, for each consumption time period, Each of the first feature vectors acquired during the consumption period is input into a preset self-attention layer to obtain an output fourth feature vector corresponding to the consumption period; the fourth input unit 650 is configured to input the combination of the first feature vectors into a neural network model to obtain a predicted value, wherein the predicted value is used to represent the degree of interest of the first user in the information to be recommended, wherein the combination of the first feature vectors includes: the fourth feature vector, the second feature vector, and the third feature vector; the calculation unit 660 is configured to determine a loss value based on the predicted value and the tag data through a preset loss function; and the adjustment unit 670 is configured to train the neural network model based on the loss value.

[0102] Here, the operation of each of the above units 610 to 670 of the model training device 600 is similar to the operation of steps 410 to 470 described above, and will not be repeated here.

[0103] 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.

[0104] According to embodiments of this disclosure, an electronic device, a readable storage medium, and a computer program product are also provided.

[0105] refer to Figure 7The present invention describes a structural block diagram of an electronic device 700 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.

[0106] like Figure 7 As shown, the electronic device 700 includes a computing unit 701, which can perform various appropriate actions and processes based on a computer program stored in a read-only memory (ROM) 702 or a computer program loaded from a storage unit 708 into a random access memory (RAM) 703. The RAM 703 may also store various programs and data required for the operation of the electronic device 700. The computing unit 701, ROM 702, and RAM 703 are interconnected via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0107] Multiple components in electronic device 700 are connected to I / O interface 705, including: input unit 706, output unit 707, storage unit 708, and communication unit 709. Input unit 706 can be any type of device capable of inputting information to electronic device 700. Input unit 706 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 707 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 708 may include, but is not limited to, disk and optical disk. Communication unit 709 allows electronic device 700 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.

[0108] The computing unit 701 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 701 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 701 performs the various methods and processes described above, such as method 200 or 400. For example, in some embodiments, method 200 or 400 may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 708. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 700 via ROM 702 and / or communication unit 709. When the computer program is loaded into RAM 703 and executed by the computing unit 701, one or more steps of method 200 or 400 described above may be performed. Alternatively, in other embodiments, the computing unit 701 may be configured to execute method 200 or 400 by any other suitable means (e.g., by means of firmware).

[0109] 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.

[0110] 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.

[0111] 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.

[0112] 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).

[0113] 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.

[0114] 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.

[0115] 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.

[0116] 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. An information recommendation method, comprising: For each of the multiple consumption time periods, obtain the first information identifier of the information consumed by the first user within that consumption time period; Determine a first feature vector corresponding to each of the first information identifiers, and a second feature vector corresponding to the first user identifier corresponding to the first user, wherein the first user identifier includes a marker slot; A second user identifier is obtained by changing the tag slot of the first user identifier, wherein the number of user identifiers in the combination formed by the second user identifier and the first user identifier is the same as the number of the plurality of consumption time periods; Determine a fifth feature vector corresponding to each second user identifier, wherein each fifth feature vector and the second feature vector are used to characterize the user profile of the first user in each consumption time period; Determine the third feature vector corresponding to the second information identifier of the information to be recommended; For each consumption time period, each of the first feature vectors obtained within that consumption time period is input into a preset self-attention layer to obtain the output fourth feature vector corresponding to that consumption time period. The first feature vector combination is input into a preset neural network model to obtain a predicted output value, wherein the predicted value represents the degree of interest of the first user in the information to be recommended, and wherein the first feature vector combination includes: the fourth feature vector, the second feature vector, the third feature vector, and the fifth feature vector; and Based on the predicted value, determine whether to recommend the information to be recommended to the first user.

2. The method as described in claim 1, wherein, The first information identifier for obtaining the information on the first user's consumption during the consumption period includes: For each of the multiple information categories, obtain the first information identifier of the first user within the consumption time period and under that information category.

3. The method as described in claim 1 or 2, wherein, The consumed information is the information that the first user is most interested in, wherein the degree of interest is determined based on at least one of the following: The first user's cumulative number of clicks on the information within the corresponding consumption period, and the first user's maximum single consumption duration of the information within the consumption period.

4. The method of claim 1, wherein, The first feature vector combination further includes: a cross feature vector, wherein the cross feature vector is determined based on the first user identifier and the tag information corresponding to the information to be recommended, wherein the tag information is determined based on the content of the information to be recommended.

5. The method of claim 1, wherein, The preset neural network model includes: a densely connected network input layer and a first fully connected layer; The step of combining the first feature vectors and inputting them into a preset neural network model to obtain the predicted output value includes: The first feature vector is combined and input into the input layer of the densely connected network to obtain the output sixth feature vector; and The predicted value is obtained based on the sixth feature vector and through the first fully connected layer.

6. The method of claim 5, wherein, The preset neural network model also includes: a gate network and a second fully connected layer; The method further includes: The second feature vector combination is sequentially input into the gate network and the second fully connected layer to obtain the output seventh feature vector, wherein, The step of obtaining the predicted value based on the sixth feature vector and through the first fully connected layer includes any one of the following: The sixth feature vector is input into the first fully connected layer to obtain the predicted output value; The sixth and seventh feature vectors are fused to obtain an eighth feature vector; the eighth feature vector is then input into the first fully connected layer to obtain the output predicted value. The second feature vector combination includes: other feature vectors in the first feature vector combination besides the feature vectors determined based on the information to be recommended.

7. A model training method, comprising: For each of the multiple consumption time periods, obtain the first information identifier of the information consumed by the first user within that consumption time period; Determine a first feature vector corresponding to each of the first information identifiers, and a second feature vector corresponding to the first user identifier corresponding to the first user, wherein the first user identifier includes a marker slot; A second user identifier is obtained by changing the tag slot of the first user identifier, wherein the number of user identifiers in the combination formed by the second user identifier and the first user identifier is the same as the number of the plurality of consumption time periods; Determine a fifth feature vector corresponding to each second user identifier, wherein each fifth feature vector and the second feature vector are used to characterize the user profile of the first user in each consumption time period; The third feature vector corresponding to the second information identifier of the information to be recommended is determined, as well as the preset tag data used to determine whether to recommend the information to the first user; For each consumption time period, each of the first feature vectors obtained within that consumption time period is input into a preset self-attention layer to obtain the output fourth feature vector corresponding to that consumption time period. The first feature vector combination is input into the neural network model to obtain a predicted value, wherein the predicted value is used to represent the degree of interest of the first user in the information to be recommended, wherein the first feature vector combination includes: a fourth feature vector, a second feature vector, a third feature vector, and a fifth feature vector; Based on the predicted value and the label data, a loss value is determined using a preset loss function; and The neural network model is trained based on the loss value.

8. The method of claim 7, wherein, The first information identifier for obtaining the information consumed by the first user during the consumption period includes: for each of the multiple information categories, obtaining the first information identifier of the first user during the consumption period and under that information category.

9. The method of claim 7, wherein, The consumed information is the information that the first user is most interested in, wherein the degree of interest is determined based on at least one of the following: The first user's cumulative number of clicks on the information within the corresponding consumption period, and the first user's maximum single consumption duration of the information within the consumption period.

10. The method of claim 7, wherein, The first feature vector combination further includes: a cross feature vector, wherein the cross feature vector is determined based on the first user identifier and the tag information corresponding to the information to be recommended, wherein the tag information is determined based on the content of the information to be recommended.

11. The method of claim 7, wherein, The neural network model includes: a densely connected network input layer and a first fully connected layer; The step of combining the first feature vectors and inputting them into the neural network model to obtain the predicted output value includes: The first feature vector is combined and input into the input layer of the densely connected network to obtain the output sixth feature vector; and The predicted value is obtained based on the sixth feature vector and through the first fully connected layer.

12. The method of claim 11, wherein, The neural network model also includes: a gate network and a second fully connected layer; The method further includes: The second feature vector combination is sequentially input into the gate network and the second fully connected layer to obtain the output seventh feature vector, wherein, The step of obtaining the predicted value by sequentially passing the sixth feature vector through the first fully connected layer includes any one of the following: The sixth feature vector is input into the first fully connected layer to obtain the predicted output value; The sixth and seventh feature vectors are fused to obtain an eighth feature vector; the eighth feature vector is then input into the first fully connected layer to obtain the output predicted value. The second feature vector combination includes: other feature vectors in the first feature vector combination besides the feature vectors determined based on the information to be recommended.

13. The method according to any one of claims 7-11, wherein, Adjusting the parameters of the neural network model based on the loss value includes: Based on the loss value, the neural network model is trained using multiple optimizers, wherein the number of optimizers is the same as the number of consumption time periods.

14. The method of claim 13, wherein, The multiple consumption time periods include short-term consumption time periods, medium-term consumption time periods, and long-term consumption time periods determined according to consumption history time, and the multiple optimizers include: optimizers with decay factors and optimizers with automatic decay learning rates.

15. The method of claim 14, wherein, The multiple optimizers include: RMSProp optimizer, AdaGrad optimizer, and Adam optimizer.

16. The method of claim 7, wherein, The process of determining the loss value based on the predicted value and the label data using a preset loss function includes: Based on the predicted value and the label data, a first loss value is determined using a preset first loss function; Based on the second feature vector and the fifth feature vector, a second loss value is determined using a preset second loss function. Training the neural network model based on the loss value includes: The neural network model is trained based on the first loss value and the second loss value, so as to minimize the first loss value and make the second loss value greater than a preset threshold.

17. The method of claim 16, wherein, The preset second loss function is determined based on the following formula: in, This represents a vector matrix composed of the second eigenvector and the fifth eigenvector. Represents the identity matrix. Represents the matrix norm. This refers to the preset threshold.

18. An information recommendation device, comprising: The first acquisition unit is configured to acquire, for each of the multiple consumption time periods, a first information identifier of the information consumed by the first user within that consumption time period. The first determining unit is configured to determine a first feature vector corresponding to each of the first information identifiers and a second feature vector corresponding to the first user identifier corresponding to the first user, wherein the first user identifier includes a marker slot; A unit for obtaining a second user identifier by changing the marker slot of the first user identifier, wherein the number of user identifiers in the combination formed by the second user identifier and the first user identifier is the same as the number of the plurality of consumption time periods; A unit for determining a fifth feature vector corresponding to each second user identifier, wherein each fifth feature vector and the second feature vector are used to characterize the user profile of the first user in each consumption time period; The second determining unit is configured to determine the third feature vector corresponding to the second information identifier of the information to be recommended; The first input unit is configured to input each of the first feature vectors obtained within each consumption time period into a preset self-attention layer for each consumption time period, so as to obtain the output fourth feature vector corresponding to the consumption time period. The second input unit is configured to input the first feature vector combination into a preset neural network model to obtain a predicted output value, wherein the predicted value is used to represent the degree of interest of the first user in the information to be recommended, and wherein the first feature vector combination includes: the fourth feature vector, the second feature vector, the third feature vector, and the fifth feature vector; and The recommendation unit is configured to determine whether to recommend the information to be recommended to the first user based on the predicted value.

19. A model training device, comprising: The second acquisition unit is configured to acquire, for each of the multiple consumption time periods, a first information identifier of the information consumed by the first user within that consumption time period. The third determining unit is configured to determine a first feature vector corresponding to each of the information identifiers and a second feature vector corresponding to the first user identifier corresponding to the first user, wherein the first user identifier includes a marker slot; A unit for obtaining a second user identifier by changing the marker slot of the first user identifier, wherein the number of user identifiers in the combination formed by the second user identifier and the first user identifier is the same as the number of the plurality of consumption time periods; A unit for determining a fifth feature vector corresponding to each second user identifier, wherein each fifth feature vector and the second feature vector are used to characterize the user profile of the first user in each consumption time period; The fourth determining unit is configured to determine the third feature vector corresponding to the second information identifier of the information to be recommended, and preset tag data for determining whether to recommend the information to be recommended to the first user; The third input unit is configured to input each of the first feature vectors obtained within each consumption time period into a preset self-attention layer for each consumption time period, so as to obtain the output fourth feature vector corresponding to the consumption time period. The fourth input unit is configured to input the first feature vector combination into the neural network model to obtain a predicted value, wherein the predicted value is used to represent the degree of interest of the first user in the information to be recommended, and wherein the first feature vector combination includes: a fourth feature vector, a second feature vector, and the third feature vector; The calculation unit is configured to determine a loss value based on the predicted value and the label data using a preset loss function; and The adjustment unit is configured to train the neural network model based on the loss value.

20. 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 1-17.

21. 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 1-17.

22. 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 1-17.

Images