Data Processing Method, Apparatus, Electronic Device, Medium, and Program Product

Abstract

The present disclosure provides a data processing method, apparatus, electronic device, medium, and program product, which relate to the technical field of data processing, and particularly to the fields of artificial intelligence and personalized recommendation. The implementation solution is as follows: determining a first query feature corresponding to first query information and a user feature corresponding to a user; processing the first query feature and the user feature by using a gain model to obtain a predicted interest score of the user in relevant content related to the first query information; and determining first recommended content for a relevant page of the first query information, wherein the number of relevant content related to the first query information in the first recommended content is determined based on the predicted interest score.

Description

Technical Field

[0001] The present disclosure relates to the field of data processing technologies, and in particular, to the fields of artificial intelligence and personalized recommendation. Specifically, the present disclosure relates to a data processing method, apparatus, electronic device, computer-readable storage medium, and computer program product. Background Art

[0002] Artificial intelligence is a discipline that studies how to make a computer simulate certain human thinking processes and intelligent behaviors (such as learning, reasoning, thinking, planning, etc.). It includes both hardware-level technologies and software-level technologies. Artificial intelligence hardware technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, and big data processing; artificial intelligence software technologies mainly include several major directions such as computer vision technology, speech recognition technology, natural language processing technology, and machine learning / deep learning, big data processing technology, and knowledge graph technology.

[0003] The methods described in this section are not necessarily methods that have been previously envisioned or adopted. Unless otherwise specified, no method described in this section should be considered prior art solely because it is included in this section. Similarly, unless otherwise specified, the problems mentioned in this section should not be considered to have been recognized in any prior art. Summary of the Invention

[0004] The present disclosure provides a data processing method, apparatus, electronic device, computer-readable storage medium, and computer program product.

[0005] According to one aspect of the present disclosure, there is provided a data processing method, including: determining a first query feature corresponding to a first query information and a user feature corresponding to a user; processing the first query feature and the user feature by using a gain model to obtain a predicted interest score of the user for relevant content related to the first query information; and determining a first recommended content for a relevant page of the first query information, wherein the number of relevant content related to the first query information in the first recommended content is determined based on the predicted interest score.

[0006] According to another aspect of the present disclosure, there is provided a data processing apparatus, including: a feature extraction unit configured to determine a first query feature corresponding to a first query information and a user feature corresponding to a user; a prediction unit configured to process the first query feature and the user feature by using a gain model to obtain a predicted interest score of the user for relevant content related to the first query information; and a recommendation unit configured to determine a first recommended content for a relevant page of the first query information, wherein the number of relevant content related to the first query information in the first recommended content is determined based on the predicted interest score.

[0007] According to another aspect of the present disclosure, there is also provided an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and when the instructions are executed by the at least one processor, the at least one processor is enabled to execute the method according to the embodiments of the present disclosure.

[0008] According to another aspect of the present disclosure, there is also provided a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause the computer to execute the method according to the embodiments of the present disclosure.

[0009] According to another aspect of the present disclosure, there is also provided a computer program product, including a computer program, wherein the computer program, when executed by a processor, implements the method according to the embodiments of the present disclosure.

[0010] According to one or more embodiments of the present disclosure, by using a gain model to predict a user's relevance demand based on query information and user information, a targeted prediction result can be provided for each user's query behavior, thereby providing more accurate personalized recommended content for the user.

[0011] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The drawings exemplarily illustrate embodiments and constitute a part of the specification, and are used together with the written description of the specification to explain the exemplary embodiments of the embodiments. The illustrated embodiments are only for illustrative purposes and do not limit the scope of the claims. In all 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 can be implemented according to an embodiment of the present disclosure;

[0014] Figure 2 An exemplary process of a data processing method according to an embodiment of the present disclosure;

[0015] Figure 3 An exemplary process of training a gain model according to an embodiment of the present disclosure;

[0016] Figure 4 Shows an exemplary structure of a gain model according to an embodiment of the present disclosure;

[0017] Figure 5 Shows an exemplary block diagram of a data processing device according to an embodiment of the present disclosure;

[0018] Figure 6 Shows a structural block diagram of an exemplary electronic device that can be used to implement an embodiment of the present disclosure. Detailed implementation manners

[0019] The following describes exemplary embodiments of the present disclosure with reference to the accompanying drawings. Various details of the embodiments of the present disclosure are included to assist understanding, and they should be considered merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0020] In the present disclosure, unless otherwise specified, the terms "first", "second", etc. are used to describe various elements and are not intended to limit the positional relationship, timing relationship, or importance relationship of these elements. Such terms are only used to distinguish one element from another. In some examples, the first element and the second element may refer to the same instance of the element, and in certain cases, based on the context description, they may also refer to different instances.

[0021] In the description of various examples in the present disclosure, the terms used are only for the purpose of describing specific examples and are not intended to be limiting. Unless the context clearly indicates otherwise, if the number of elements is not specifically limited, the element may be one or more. In addition, the term "and / or" used in the present disclosure covers any one of the listed items and all possible combinations.

[0022] Embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.

[0023] Figure 1 Shows a schematic diagram of an exemplary system 100 in which various methods and devices described herein can be implemented according to an embodiment of the present disclosure. Referring to 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 that couple 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.

[0024] In an embodiment of the present disclosure, the server 120 can run one or more services or software applications that enable the execution of a data processing method according to an embodiment of the present disclosure.

[0025] In some embodiments, server 120 may also provide other services or software applications, which may include non-virtual environments and virtual environments. In some embodiments, these services may be provided as web-based services or cloud services, such as provided to users of client devices 101, 102, 103, 104, 105, and / or 106 under a software as a service (SaaS) model.

[0026] In 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 a combination thereof that may be executed by one or more processors. Users operating client devices 101, 102, 103, 104, 105, and / or 106 may in turn utilize one or more client applications to interact with server 120 to utilize the services provided by these components. It should be understood that various different system configurations are possible, which may be different from system 100. Thus, Figure 1 is an example of a system for implementing the various methods described herein and is not intended to be limiting.

[0027] Users may use client devices 101, 102, 103, 104, 105, and / or 106 to perform information searches and queries. The client device may provide an interface that enables a user of the client device to interact with the client device. The client device may also output information to the user via the interface. Although Figure 1 only six client devices are depicted, those skilled in the art will be able to understand that the present disclosure may support any number of client devices.

[0028] Client devices 101, 102, 103, 104, 105, and / or 106 may include various types of computing devices such as portable handheld devices, general-purpose computers (such as personal computers and laptop computers), 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, etc. These computing devices may 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 WindowsMobile OS, iOS, Windows Phone, 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 different applications such as various Internet-related applications, communication applications (such as email applications), short message service (SMS) applications, and may use various communication protocols.

[0029] Network 110 can be any type of network known to those skilled in the art, which 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, token ring, 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 (such as Bluetooth, WIFI), and / or any combination of these and / or other networks.

[0030] Server 120 may include one or more general-purpose computers, dedicated server computers (such as PC (personal computer) servers, UNIX servers, midrange servers), blade servers, mainframes, 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 (such as one or more flexible pools of logical storage devices that can be virtualized to maintain virtual storage devices for the server). In various embodiments, server 120 may run one or more services or software applications that provide the functions described below.

[0031] The computing unit in server 120 can run one or more operating systems including any of the above - mentioned operating systems and any commercially available server operating systems. Server 120 can also run any one of a variety of additional server applications and / or middleware applications, including HTTP servers, FTP servers, CGI servers, JAVA servers, database servers, etc.

[0032] In some embodiments, server 120 can include one or more applications to analyze and combine data feeds and / or event updates received from users of client devices 101, 102, 103, 104, 105, and / or 106. Server 120 can 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 / or 106.

[0033] In some embodiments, server 120 can be a server of a distributed system or a server incorporating 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, which solves the defects of high management difficulty and weak business scalability existing in traditional physical hosts and virtual private server (VPS) services.

[0034] System 100 can also include one or more databases 130. In certain embodiments, these databases can be used to store data and other information. For example, one or more of databases 130 can be used to store information such as audio files and video files. Databases 130 can reside in various locations. For example, the databases used by server 120 can be local to server 120, or can be remote from server 120 and can communicate with server 120 via a network - based or dedicated connection. Databases 130 can be of different types. In certain embodiments, the databases used by server 120 can be relational databases, for example. One or more of these databases can store, update, and retrieve data to and from the databases in response to commands.

[0035] In certain embodiments, one or more of databases 130 can also be used by applications to store application data. The databases used by applications can be different types of databases, such as key - value repositories, object repositories, or conventional repositories supported by a file system.

[0036] Figure 1The system 100 can be configured and operated in various ways to enable the application of various methods and devices described in the present disclosure.

[0037] Users can use a search engine to search for content of interest. For example, a user can enter query information in a search engine and browse the results related to the query information returned by the search engine. After the user's search needs are met, recommended content can also be provided to the user on the relevant page of the query information, thereby optimizing the search product into a search and recommendation integrated product with personalized recommendation capabilities.

[0038] The recommended content provided to the user on the relevant page can generally be divided into two types. One is relevant content related to the query information, and the other is personalized content related to the user's individual characteristics. For example, relevant content is pictures, articles, videos, etc. that can be queried based on the query information, and personalized content is content that is speculated to be of interest to the user based on the user's historical behavior. The personalized content may be irrelevant to the query information. Different users have different degrees of demand for relevant content for different query information.

[0039] The following several methods are usually adopted in the related art to predict the degree of the user's demand for relevant content:

[0040] 1. Provide a fixed N pieces of relevant content for all query requests in the recommended content;

[0041] 2. Cluster according to the semantic vector of the query request, and count the number of relevant content at the 80th percentile of each cluster center, and further determine the recommended number of relevant content corresponding to the cluster center to which the current query request belongs;

[0042] 3. Calculate the number of relevant content at the 80th percentile in the predetermined category to which the query request belongs, and further determine the recommended number of relevant content corresponding to the category to which the current query request belongs;

[0043] 4. Combine the predetermined category to which the query request belongs and the user's activity for the content of this category to calculate the overall 80th percentile relevant number, and further determine the recommended number of relevant content corresponding to the category and the user's activity to which the current query request belongs.

[0044] However, the prediction of the degree of relevance content requirements for users in related technologies has drawbacks. The first method is the simplest to implement, but it ignores the significant differences in relevance requirements under different query requests. In the second method, limited by the clustering performance and accuracy, it still cannot better distinguish the relevance requirements of different query requests. The third method distinguishes the relevance requirements of different query requests by the category to which they belong, improving the classification accuracy of query requests, but it does not consider that there are still significant differences in the relevance requirements of different users for different query requests. The fourth method considers both user activity and the category to which the query request belongs and cannot accurately estimate the personalized relevance requirements of each user under different query requests.

[0045] To further meet the personalized relevance requirements of users, the present disclosure provides a new data processing method.

[0046] Figure 2 An exemplary process of the data processing method according to an embodiment of the present disclosure is shown.

[0047] In step S202, a first query feature corresponding to the first query information and a user feature corresponding to the user are determined.

[0048] In step S204, the gain model is used to process the first query feature and the user feature to obtain a predicted interest score of the user in the relevant content related to the first query information.

[0049] In step S206, a first recommended content for the relevant page of the first query information is determined, where the number of relevant content related to the first query information in the first recommended content is determined based on the predicted interest score.

[0050] Using the embodiment of the present disclosure, by using the gain model to predict the relevance requirements of users based on query information and user information, targeted prediction results can be provided for each user's query behavior, thereby providing more accurate personalized recommended content for users.

[0051] The principle of the present disclosure will be described in detail below.

[0052] In step S202, a first query feature corresponding to the first query information and a user feature corresponding to the user can be determined.

[0053] The first query feature may include at least one of the semantic feature and the statistical feature of the first query information. Among them, the semantic feature can be used to represent the specific content of the first query information, and the statistical feature may include the statistical results of user behavior in historical query behaviors related to the first query information, which can reflect the degree of interest of the user in the first query information.

[0054] In some examples, the semantic features of the first query information may include the semantic vector of the first query information, and the semantic vector can be obtained by processing the first query information in any natural language vectorization method. In the example, the semantic features of the first query information may also include the classification features of the semantics of the first query information. For example, the classification features indicating whether the first query information belongs to illegal information (such as whether it belongs to pornographic information) and whether it has timeliness can be determined based on the semantics of the first query information.

[0055] In some examples, the statistical features of the first query information may include the statistical data of user behavior in the pages related to the first query information. When determining the statistical features of the first query information, data collection can be performed for all users or randomly selected users to minimize the impact of user personalized behavior on the statistical results of the query information. In the example, the statistical features of the first query information may include the statistical data of the number of clicks and browsing duration of the user in the result page and the landing page of the first query information, which can reflect the user's interest level in the first query information. Among them, the result page refers to the query result list returned by the search engine in response to the first query information, and the landing page refers to the specific page entered by clicking the link in the query result list. For example, the statistical features of the first query information may include the number of searches of the result page of the first query information, the click-through rate of the result page of the first query information, the number of times the user enters the landing page, and the number of times the user repeatedly enters the landing page under the same query information.

[0056] In the example, the statistical features of the first query information may also include the statistical data of the user's clicks and browsing of relevant content in the pages related to the first query information, which can reflect the user's relevant needs under the first query information. For example, the statistical features of the first query information may include the proportion of relevant content distribution in the landing page, the proportion of the playing duration of relevant content in the landing page, the fast-scrolling rate of relevant content in the landing page (i.e., the proportion of the content that the user quickly scrolls past), the long-playing rate of relevant content in the landing page (i.e., the proportion of the user watching the relevant content exceeding a predetermined threshold (such as 10s)), and the browsing step length of relevant content in the landing page (i.e., the number of relevant content that the user watches).

[0057] Without departing from the principles of the present disclosure, those skilled in the art can supplement, delete, or modify the above statistical features according to the actual situation to meet the requirements of actual applications.

[0058] User characteristics may include at least one of the user's historical access characteristics for query content, historical access characteristics for relevant content, and historical access characteristics for personalized content. User characteristics can reflect the user's browsing habits after sending a query request, browsing habits for relevant content, and browsing habits for personalized content, thereby further reflecting the user's different needs for relevant content and personalized content.

[0059] In the example, the historical access characteristics for query content may include the user's historical browsing length, which represents how much content the user will browse each time during historical use. In the case where the application automatically refreshes content for the user, the background refresh quantity for each request can be used to represent the user's historical browsing length. Further, the user's historical access characteristics may also include the frequency of accessing the landing page within different time ranges (such as 30 days, 14 days, etc.), the number of searches within a certain time range (such as 30 days), and the average number of times of repeatedly entering the landing page under the same query information within a certain time range (such as 30 days).

[0060] In the example, the historical access characteristics for relevant content may include the user's fast - swiping rate for relevant content, the proportion of the playing duration of relevant content, the long - playing rate of relevant content, etc.

[0061] In the example, the historical access characteristics for personalized content may include the user's fast - swiping rate for personalized content, the proportion of the playing duration of personalized content, the long - playing rate of personalized content, etc.

[0062] In step S204, a gain model is used to process the first query feature and the user characteristics to obtain the predicted interest score of the user for relevant content related to the first query information.

[0063] The gain model can be used to estimate the impact of an intervention action (treatment) on the user's corresponding behavior. Using the gain model, the difference in behavior of the same individual in different situations of intervention and non - intervention can be determined. In some embodiments, the gain model may be an uplift model. In this disclosure, the uplift model will be used as an example to describe the principle of this disclosure. Without departing from the principle of this disclosure, other gain models with the same function can also be used.

[0064] Figure 3 An exemplary process of training the gain model according to an embodiment of the present disclosure is shown.

[0065] In step S302, a sample user set can be determined. The sample users can be selected from all users in a random manner to cover different types of users as much as possible and improve the applicable range of the trained gain model.

[0066] In step S304, the sample user characteristics of the sample user and the second query characteristics of the second query information input by the sample user can be determined.

[0067] In step S306, second recommended content can be provided on a page related to the second query information. Among the sample data collected during the training process, some sample users can be intervened, and the other part of the sample users can be the non-intervened control group. In the case of being intervened, all the recommended content in the second recommended content is related to the second query information, and in the case of not being intervened, the second recommended content includes relevant content related to the second query information and personalized content for the sample user. Any method provided in the related art can be used to provide recommended content for users who are not intervened. For the uplift model, in the case of intervention, the intervention information (treatment) of the model can be set to 1, and in the case of not being intervened, the intervention information can be set to 0.

[0068] For users in the sample user set, within a certain time range, for at least some sample users, only relevant content is provided as the recommended result. Using this method, feedback from different users on relevant content can be effectively collected, so as to determine the needs of different users for relevant content.

[0069] In step S308, the true interest score of the sample user for the relevant content related to the second query information can be determined based on the browsing time of the sample user for the second recommended content. Among them, the browsing time of the user can be normalized as the interest score. In the gain model provided in this disclosure, the browsing time of the user for the relevant content is used to determine the degree of interest of the user in the relevant content. It can be understood that the higher the demand of the user for the relevant content, the longer the time for the user to browse the relevant content.

[0070] In step S310, the uplift model can be used to process the sample user characteristics and the second query characteristics to obtain the predicted interest score of the sample user for the relevant content. The uplift model can perceive the query information and the characteristics of the user, and predict the degree of interest of the user in the relevant content according to the characteristics of the query information and the characteristics of the user. The predicted interest score has the same representation as the true interest score, for example, represented as the normalized result of the browsing time. Those skilled in the art can also use other types of parameters to represent the predicted interest score according to the actual situation.

[0071] In step S312, the parameters of the uplift model can be adjusted to minimize the error between the predicted interest score and the true interest score. As mentioned above, the true interest score is the actual result obtained by providing all relevant content recommendations to the user, which can reflect the user's true demand for relevant content. The predicted interest score is the prediction result obtained by the model processing the query features and user features. By minimizing the error between the predicted interest score and the true interest score, the ability of the uplift model to perceive query features and user features can be improved, and further, based on the perceived features, the ability to predict the user's interest level in relevant content can be enhanced.

[0072] Figure 4 The exemplary structure of the uplift model according to the embodiments of the present disclosure is shown.

[0073] As Figure 4 shown, the uplift model 400 may include a first fully-connected layer 410 and a second fully-connected layer 420 connected in series.

[0074] For a query request initiated by a user, the query feature 401 of the query information included in the query request and the vector representation of the user feature 402 of the user can be determined and connected together as the input of the first fully-connected layer 410. Further, the input of the first fully-connected layer 410 may also include intervention information 403. When only relevant content recommendations are provided to the user, the intervention information 403 may be set to 1, and in other cases, the intervention information 403 may be set to 0. Those skilled in the art may also set the intervention information to any other appropriate value according to the actual situation.

[0075] Returning to reference Figure 2 , in step S206, the first recommended content for the relevant page of the first query information is determined, where the number of relevant content related to the first query information in the first recommended content is determined based on the predicted interest score. The relevant page of the first query information mentioned here may be the result page or landing page of the first query information, or any other form of page displayed to the user in response to the query request including the first query information. The recommended content refers to the content recommended by the system for the user to further browse in addition to the content of the relevant page itself. The recommended content may include relevant content related to the query information and may also include personalized content related to the user.

[0076] In some examples, the number of relevant content may be expressed as an absolute number or a relative number. The absolute number refers to the actual number of relevant content items, such as numerical values like 5, 10, etc. The relative number refers to the proportion of relevant content in the first recommended content, such as 50%, 60%, etc.

[0077] The recommended quantity of relevant content corresponding to the predicted score of interest can be determined through various predefined mapping methods.

[0078] In some embodiments, the score of interest can be divided into different ranges, and different recommended quantities of relevant content can be determined for different ranges of the score of interest. In an example, taking the predicted score of interest as a normalized result between 0 and 1, the score of interest can be divided into three ranges: 0 to 0.4, 0.4 to 0.7, and 0.7 to 1. For the score of interest from 0 to 0.4, the recommended quantity of relevant content can be determined to be 30%; for the score of interest from 0.4 to 0.7, the recommended quantity of relevant content can be determined to be 50%; for the score of interest from 0.7 to 1, the recommended quantity of relevant content can be determined to be 80%. Those skilled in the art can preset appropriate mapping relationships according to the actual situation, and the solution of the present disclosure is not limited thereto.

[0079] In other embodiments, an adjustment value can be determined according to the reference recommended quantity of query content for the predicted score of interest. Step S206 may include: determining the reference recommended quantity of relevant content based on the first query information; determining the adjustment value of relevant content based on the predicted score of interest; and determining the recommended quantity of relevant content in the first recommended content based on the reference recommended quantity and the adjustment value. In this case, the reference recommended quantity for the query information can be determined according to the content of the query information. Among them, the reference recommended quantity can be determined based on the category of the first query information. For example, the method provided in the related art can be used to determine the reference recommended quantity of relevant content for the query information in this category according to the category to which the query information belongs. Then, the reference recommended quantity can be adjusted according to the score of interest of the user in the relevant content predicted by the gain model. Still taking the predicted score of interest as a normalized result between 0 and 1, and dividing the score of interest into three ranges: 0 to 0.4, 0.4 to 0.7, and 0.7 to 1 as an example. For the score of interest from 0 to 0.4, the adjustment value of relevant content can be determined to be -30%; for the score of interest from 0.4 to 0.7, the adjustment value of relevant content can be determined to be 0; for the score of interest from 0.7 to 1, the adjustment value of relevant content can be determined to be +30%. Those skilled in the art can preset appropriate mapping relationships according to the actual situation, and the solution of the present disclosure is not limited thereto.

[0080] In the case of using an uplift model as a gain model, the first predicted interest score in the case of being intervened and the second predicted interest score in the case of not being intervened can be output by the uplift model. Among them, the first predicted interest score in the case of being intervened is the result of processing the query feature, user feature, and intervention information set to 1 by using the uplift model, and the second predicted interest score in the case of not being intervened is the result of processing the query feature, user feature, and intervention information set to 0 by using the uplift model. An adjustment value can be determined based on the difference between the first predicted interest score and the second predicted interest score. The above difference can be used to represent the different performances of the user in the case of being intervened (for example, only providing recommendations of relevant content to the user) and in the case of not being intervened (for example, providing recommendations of both relevant content and personalized content to the user), and the adjustment value of the relevant content can be determined according to the different performances of the user. For example, if the predicted interest score output by the user in the case of being intervened is higher, it means that the number of relevant content is more, and the time the user stays on the page is also longer. Therefore, in this case, the number of relevant content in the recommended content can be increased. Conversely, the number of relevant content in the recommended content can be reduced. Those skilled in the art can determine the specific mapping relationship between the difference between the first predicted interest score and the second predicted interest score and the adjustment value according to the actual situation.

[0081] Without departing from the principles of the present disclosure, step S206 may further include determining the display manner of the relevant content on the page according to the predicted interest score. As mentioned above, the predicted interest score can reflect the user's interest degree in the relevant content. Therefore, the display features such as the position and size of the relevant content on the page can be adjusted according to the predicted interest score to increase or decrease the user's attention degree to the relevant content on the page.

[0082] Figure 5 An exemplary block diagram of a data processing apparatus according to an embodiment of the present disclosure is shown.

[0083] As Figure 5 shown, the apparatus 500 may include a feature extraction unit 510, a prediction unit 520, and a recommendation unit 530.

[0084] The feature extraction unit 510 may be configured to determine a first query feature corresponding to the first query information and a user feature corresponding to the user.

[0085] The prediction unit 520 may be configured to process the first query feature and the user feature by using a gain model to obtain a predicted interest score of the user for relevant content related to the first query information.

[0086] The recommendation unit 530 may be configured to determine first recommendation content for relevant pages of the first query information, where the number of relevant content related to the first query information in the first recommendation content is determined based on a predicted interest score.

[0087] In some embodiments, the gain model is an uplift model.

[0088] In some embodiments, the uplift model is trained by the following method: determining a set of sample users; determining sample user features of the sample users and second query features of the second query information input by the sample users, and providing second recommendation content in relevant pages of the second query information; determining the true interest score of the sample users for relevant content related to the second query information based on the browsing time of the sample users for the second recommendation content; using the uplift model to process the sample user features and the second query features to obtain a predicted interest score of the sample users for the relevant content; and adjusting the parameters of the uplift model to minimize the error between the predicted interest score and the true interest score.

[0089] In some embodiments, in the case of being intervened, all the recommendation content in the second recommendation content is related to the second query information, and in the case of not being intervened, the second recommendation content includes relevant content related to the second query information and personalized content for the sample users.

[0090] In some embodiments, the first recommendation content further includes personalized content for the user.

[0091] In some embodiments, the recommendation unit is configured to: determine a reference recommendation quantity of relevant content based on the first query information; determine an adjustment value of the relevant content based on the predicted interest score; and determine the recommendation quantity of the relevant content in the first recommendation content based on the reference recommendation quantity and the adjustment value.

[0092] In some embodiments, determining the adjustment value of the relevant content based on the predicted interest score includes: determining a first predicted interest score output by the gain model in the case of being intervened; determining a second predicted interest score output by the gain model in the case of not being intervened; and determining the adjustment value based on the difference between the first predicted interest score and the second predicted interest score.

[0093] In some embodiments, the reference recommendation quantity is determined based on the category of the first query information.

[0094] In some embodiments, the first query feature includes at least one of the semantic feature of the first query information and the statistical feature of the first query information.

[0095] In some embodiments, the user characteristics include at least one of the user's historical access characteristics for query content, historical access characteristics for relevant content, and historical access characteristics for personalized content.

[0096] The units 510-530 shown in Figure 5 can be used to execute Figure 2 the steps S202-S206 shown in, which will not be elaborated here.

[0097] It should be understood that Figure 5 each module or unit of the device 500 shown in can correspond to each step in the method 200 described with reference to Figure 2 Accordingly, the operations, features, and advantages described above for the method 200 also apply to the device 500 and its included modules and units. For the sake of brevity, certain operations, features, and advantages are not elaborated here.

[0098] In the technical solution of the present disclosure, the processing of collection, storage, use, processing, transmission, provision, and disclosure of the user's personal information involved all comply with the provisions of relevant laws and regulations and do not violate public order and good customs.

[0099] According to an embodiment of the present disclosure, there is also provided an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method according to the embodiment of the present disclosure.

[0100] According to an embodiment of the present disclosure, there is also provided a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause the computer to execute the method according to the embodiment of the present disclosure.

[0101] According to an embodiment of the present disclosure, there is also provided a computer program product, including a computer program, wherein the computer program, when executed by a processor, implements the method according to the embodiment of the present disclosure.

[0102] Referring to Figure 6, a block diagram of an electronic device 600 that can be a server or a client of the present disclosure will now be described. It 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, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0103] As Figure 6 shown, the electronic device 600 includes a computing unit 601, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 602 or a computer program loaded from a storage unit 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 600 can also be stored. The computing unit 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0104] Multiple components in the electronic device 600 are connected to the I / O interface 605, including: an input unit 606, an output unit 607, a storage unit 608, and a communication unit 609. The input unit 606 can be any type of device that can input information into the electronic device 600. The input unit 606 can receive input digital or character information, and generate key signal inputs related to the user settings and / or function controls of the electronic device, and can include, but are not limited to, a mouse, a keyboard, a touch screen, a trackpad, a trackball, a joystick, a microphone, and / or a remote control. The output unit 607 can be any type of device that can present information, and can include, but are not limited to, a display, a speaker, a video / audio output terminal, a vibrator, and / or a printer. The storage unit 608 can include, but is not limited to, magnetic disks, optical disks. The communication unit 609 allows the electronic device 600 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks, and can include, but are not limited to, a modem, a network card, an infrared communication device, a wireless communication transceiver, and / or a chipset, such as a Bluetooth™ device, an 802.11 device, a WiFi device, a WiMax device, a cellular communication device, and / or the like.

[0105] The computing unit 601 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 601 executes the various methods and processes described above, such as method 200, 300. For example, in some embodiments, methods 200, 300 can be implemented as computer software programs tangibly embodied in a machine-readable medium, such as the storage unit 608. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 600 via the ROM 602 and / or the communication unit 609. When the computer program is loaded into the RAM 603 and executed by the computing unit 601, one or more steps of the methods 200, 300 described above can be executed. Alternatively, in other embodiments, the computing unit 601 can be configured to execute methods 200, 300 in any other suitable manner (e.g., by means of firmware).

[0106] The various embodiments of the systems and techniques described above in this document can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGA), application-specific integrated circuits (ASIC), application-specific standard products (ASSP), system-on-a-chip systems (SOC), complex programmable logic devices (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a dedicated or general-purpose programmable processor, and can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0107] The program code for implementing the methods of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to the processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the program code is executed by the processor or controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program code can be executed entirely on the machine, partially on the machine, executed partially on the machine as an independent software package and partially on a remote machine, or executed entirely on a remote machine or server.

[0108] In the context of this disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium can include, 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 the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0109] In order to provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, speech input, or tactile input).

[0110] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected to each other by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), the Internet, and blockchain networks.

[0111] A computer system may include a client and a server. The client and the server are generally far away from each other and usually interact through a communication network. The relationship between the client and the server is generated by computer programs running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, or a server of a distributed system, or a server combined with a blockchain.

[0112] It should be understood that the various forms of processes shown above can be used, steps can be reordered, added or deleted. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this is not limited herein.

[0113] Although the embodiments or examples of the present disclosure have been described with reference to the accompanying drawings, it should be understood that the above methods, systems, and devices are merely exemplary embodiments or examples, and the scope of the present invention is not limited by these embodiments or examples, but is only defined by the authorized claims and their equivalent scope. Various elements in the embodiments or examples can be omitted or replaced by their equivalent elements. In addition, the steps can be executed in an order different from that described in this disclosure. Further, the various elements in the embodiments or examples can be combined in various ways. Importantly, with the evolution of technology, many of the elements described herein can be replaced by equivalent elements that emerge after this disclosure.

Claims

1. A data processing method, comprising: Determining a first query feature corresponding to first query information and a user feature corresponding to a user; Processing the first query feature and the user feature by using a gain model to obtain a predicted interest score of the user for relevant content related to the first query information, wherein the gain model is used to determine a first predicted interest score of the user in an intervened case where only relevant content is provided to the user and a second predicted interest score of the user in a non-intervened case where both relevant content and personalized content are provided to the user; and Determining first recommended content for a relevant page of the first query information, wherein the number of relevant content related to the first query information in the first recommended content is determined based on the predicted interest score, wherein the number of relevant content is determined by adjusting a reference recommended number by an adjustment value determined according to a difference between the first predicted interest score and the second predicted interest score, and the reference recommended number is determined based on the first query information.

2. The method according to claim 1, wherein The gain model is an uplift model.

3. The method according to claim 2, wherein The uplift model is trained by using the following method: Determining a set of sample users; Determining sample user features of the sample users and second query features of second query information input by the sample users; Providing second recommended content in a relevant page of the second query information; Determining a true interest score of the sample user for relevant content related to the second query information based on a browsing time of the sample user for the second recommended content; Processing the sample user features and the second query features by using the uplift model to obtain a predicted interest score of the sample user for the relevant content; And Adjusting parameters of the uplift model to minimize an error between the predicted interest score and the true interest score.

4. The method according to claim 3, wherein, In the intervened case, all recommended content in the second recommended content is related to the second query information, and In the non-intervened case, the second recommended content includes relevant content related to the second query information and personalized content for the sample users.

5. The method according to any one of claims 1 to 4, wherein, The first recommended content further includes personalized content for the user.

6. The method according to claim 1, wherein The reference recommended number is determined based on a category of the first query information.

7. The method according to any one of claims 1 to 4, wherein The first query feature includes at least one of a semantic feature of the first query information and a statistical feature of the first query information.

8. The method according to any one of claims 1 to 4, wherein The user feature includes at least one of a historical access feature of the user for query content, a historical access feature for relevant content, and a historical access feature for personalized content.

9. A data processing apparatus, comprising: A feature extraction unit configured to determine a first query feature corresponding to first query information and a user feature corresponding to a user; A prediction unit, configured to process the first query feature and the user feature by using a gain model to obtain a predicted interest score of the user for relevant content related to the first query information, where the gain model is used to determine a first predicted interest score of the user in an intervened case where only relevant content is provided to the user and a second predicted interest score of the user in a non-intervened case where both relevant content and personalized content are provided to the user; and A recommendation unit, configured to determine first recommended content for a relevant page of the first query information, where the number of relevant content related to the first query information in the first recommended content is determined based on the predicted interest score, where the number of relevant content is determined by adjusting a reference recommended number by an adjustment value determined according to a difference between the first predicted interest score and the second predicted interest score, and the reference recommended number is determined based on the first query information.

10. The device according to claim 9, wherein, The gain model is an uplift model.

11. The apparatus according to claim 10, wherein, The uplift model is trained by using the following method: Determine a set of sample users; Determine sample user features of the sample users and second query features of second query information input by the sample users; Provide second recommended content in a relevant page of the second query information; Determine a true interest score of the sample user for relevant content related to the second query information based on a browsing time of the sample user for the second recommended content; Process the sample user features and the second query features by using the uplift model to obtain a predicted interest score of the sample user for the relevant content; And Adjust parameters of the uplift model to minimize an error between the predicted interest score and the true interest score.

12. The apparatus according to claim 11, where In the intervened case, all recommended content in the second recommended content is related to the second query information, and In the non-intervened case, the second recommended content includes relevant content related to the second query information and personalized content for the sample users.

13. The device according to any one of claims 9 to 12, wherein The first recommended content further includes personalized content for the user.

14. The apparatus according to claim 9, wherein, The reference recommended number is determined based on a category of the first query information.

15. The apparatus according to any one of claims 9 to 12, wherein, The first query feature includes at least one of a semantic feature of the first query information and a statistical feature of the first query information.

16. The device according to any one of claims 9 to 12, wherein, The user feature includes at least one of a historical access feature of the user for query content, a historical access feature of the user for relevant content, and a historical access feature of the user for personalized content.

17. An electronic device, comprising: At least one processor; And A memory communicatively connected to the at least one processor; Wherein The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method according to any one of claims 1-8.

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

19. A computer program product comprising a computer program, wherein, The computer program, when executed by a processor, implements the method according to any one of claims 1-8.

Images