CN114443964A

CN114443964A - Information recommendation method and device, electronic equipment and medium

Info

Publication number: CN114443964A
Application number: CN202210119097.2A
Authority: CN
Inventors: 孙倩
Original assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Current assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Priority date: 2022-02-08
Filing date: 2022-02-08
Publication date: 2022-05-06

Abstract

The disclosure provides an information recommendation method, relates to the technical field of artificial intelligence, and particularly relates to the technical field of recommendation systems. The implementation scheme is as follows: acquiring behavior data of a user in a platform, wherein the behavior data comprises first click behavior data of the user on information in the platform and an attention list of the user, and the attention list comprises an author concerned by the user; acquiring second click behavior data of information issued by a user for an author from the first click behavior data, and determining a first association value between the user and the author based on the first click behavior data and the second click behavior data; determining a second association value between the user and the author based on the matching degree of the interest field of the user and the field of the author; determining an interest value of the user for the author based on the first correlation value and the second correlation value; and determining whether to recommend the information published by the author to the user based on the interest value.

Description

Information recommendation method and device, electronic equipment and medium

Technical Field

The present disclosure relates to the field of artificial intelligence technologies, and in particular, to the field of recommendation systems, and in particular, to an information recommendation method, apparatus, electronic device, computer-readable storage medium, and computer program product.

Background

Artificial intelligence is the subject of research that makes computers simulate some human mental processes and intelligent behaviors (such as learning, reasoning, thinking, planning, etc.), both at the hardware level and at the software level. Artificial intelligence hardware technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a voice recognition technology, a natural language processing technology, machine learning/deep learning, a big data processing technology, a knowledge map technology and the like.

With the development of multimedia information platforms, a large number of self-media authors share massive multimedia information, and how to perform personalized recommendation of information to users becomes increasingly important.

The approaches described in this section are not necessarily approaches that have been previously conceived or pursued. Unless otherwise indicated, it should not be assumed that any of the approaches described in this section qualify as prior art merely by virtue of their inclusion in this section. Similarly, unless otherwise indicated, the problems mentioned in this section should not be considered as having been acknowledged in any prior art.

Disclosure of Invention

The disclosure provides an information recommendation method, an information recommendation device, an electronic device, a computer readable storage medium and a computer program product.

According to an aspect of the present disclosure, there is provided an information recommendation method including: acquiring behavior data of a user in a platform, wherein the behavior data comprises first click behavior data of the user on information in the platform and an attention list of the user, and the attention list comprises authors concerned by the user; acquiring second click behavior data of the information issued by the author by the user from the first click behavior data, and determining a first association value between the user and the author on the basis of the first click behavior data and the second click behavior data; determining a field of interest of the user based on the behavior data; determining a second association value between the user and the author based on the matching degree of the interest field and the field of the author; determining an interest value of the user for the author based on the first associated value and the second associated value; and determining whether to recommend the information published by the author to the user based on the interest value.

According to another aspect of the present disclosure, there is provided a training method of a machine learning model, including: obtaining sample behavior data of a sample user, wherein the sample behavior data comprises click behavior data of the sample user on information in a platform and an attention list of the sample user, and the attention list comprises authors concerned by the sample user; marking a real interest value of the sample user, wherein the real interest value represents the interest degree of the sample user to the author; acquiring the information of the domain where the author is located; inputting the sample behavior data and the information of the field where the author is located into a machine learning model, and obtaining an interest prediction value of the sample user, wherein the interest prediction value indicates the prediction interest degree of the sample user for the author; calculating a loss value based on the real interest value and the predicted interest value; and adjusting parameters of the machine learning model based on the loss value.

According to another aspect of the present disclosure, there is provided an information recommendation apparatus including: a first obtaining module configured to obtain behavior data of a user in a platform, wherein the behavior data includes first click behavior data of the user on information in the platform and an attention list of the user, and wherein the attention list includes an author concerned by the user; a second obtaining module configured to obtain second click behavior data of information issued by the user for the author from the first click behavior data, and determine a first association value between the user and the author based on the first click behavior data and the second click behavior data; a first determination module configured to determine a field of interest of the user based on the behavior data; a second determining module configured to determine a second association value between the user and the author based on a matching degree of the interest field and a field in which the author is located; a third determination module configured to determine an interest value of the user in the author based on the first relevance value and the second relevance value; and a fourth determination module configured to determine whether to recommend the information published by the author to the user based on the interest value.

According to another aspect of the present disclosure, there is provided a training apparatus of a machine learning model, including: a third obtaining module configured to obtain sample behavior data of a sample user, wherein the sample behavior data includes click behavior data of the sample user on information in a platform and an attention list of the sample user, and wherein the attention list includes authors interested by the sample user; a marking module configured to mark a real interest value of the sample user, the real interest value characterizing a degree of interest of the sample user in the author; the fourth acquisition module is configured to acquire the domain information of the author; a fifth obtaining module, configured to input the sample behavior data and the domain information of the author into a machine learning model, and obtain an interest prediction value of the sample user, where the interest prediction value indicates a predicted interest degree of the author in the sample user; a calculation module configured to calculate a loss value based on the real value of interest and the predicted value of interest; and an adjustment module configured to adjust parameters of the machine learning model based on the loss values.

According to another aspect of the present disclosure, there is provided an electronic device including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform any of the methods described above.

According to another aspect of the present disclosure, there is provided a non-transitory computer readable storage medium having stored thereon computer instructions for causing the computer to perform any of the methods described above.

According to another aspect of the disclosure, a computer program product is provided, comprising a computer program, wherein the computer program realizes any of the above methods when executed by a processor.

According to one or more embodiments of the present disclosure, a method for recommending information is provided, in which a degree of interest in an author of interest is determined by data of both a degree of matching between a user and a domain of the author of interest and click behavior data of the user on information posted by the author of interest, and whether to recommend the information is determined based on the degree of interest. Therefore, in the information recommendation process, the influence of the author on the information consumption behavior of the user is taken into consideration, the information recommendation accuracy rate can be improved, the fan reading amount of the author can be increased, and the whole click rate of the recommendation system is improved.

It should be understood that the statements in this section do not necessarily identify key or critical features of the embodiments of the present disclosure, nor do they limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the embodiments and, together with the description, serve to explain the exemplary implementations of the embodiments. The illustrated embodiments are for purposes of illustration only and do not limit the scope of the claims. Throughout the drawings, identical reference numbers designate similar, but not necessarily identical, elements.

FIG. 1 shows a flow diagram of an information recommendation method according to an embodiment of the present disclosure;

FIG. 2 shows a flow diagram of a method of training a machine learning model according to an embodiment of the present disclosure;

fig. 3 shows a block diagram of an information recommendation apparatus according to an embodiment of the present disclosure;

FIG. 4 shows a block diagram of a training apparatus of a machine learning model according to an embodiment of the present disclosure; and

FIG. 5 illustrates a block diagram of an exemplary electronic device that can be used to implement embodiments of the present disclosure.

Detailed Description

Exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings, in which various details of the embodiments of the disclosure are included to assist understanding, and which are to be considered as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

In the present disclosure, unless otherwise specified, the use of the terms "first", "second", and the like to describe various elements is not intended to limit the positional relationship, the temporal relationship, or the importance relationship of the elements, and such terms are used only to distinguish one element from another. In some examples, a first element and a second element may refer to the same instance of the element, and in some cases, based on the context, they may also refer to different instances.

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

In the related art, information is recommended to a user only according to the matching degree between the content of the information and the user portrait, and the influence of an author who issues the information on the information consumption behavior of the user is not considered. Especially, with the popularity of the fan economy, the author who the user likes may be interested in the information about whatever content the user publishes, and the traditional recommendation system may omit the recommendation of the information.

To solve the above problem, the present disclosure determines a degree of interest of a user in an author with respect to a degree of matching between the user and a field of an author having been paid attention to and behavior data of the user, and determines whether to recommend information published by the author based on the degree of interest. Therefore, in the information recommendation process, the influence of the author on the information consumption behavior of the user is taken into consideration, the accuracy rate of information recommendation can be improved, the fan reading amount of the author can be increased, and the integral click rate of the recommendation system is improved.

In the technical scheme of the present disclosure, the processes of collecting, storing, using, processing, transmitting, providing, disclosing and the like of the personal information of the related users all conform to the regulations of the related laws and regulations, and do not violate the good custom of the public order.

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

Fig. 1 shows a flowchart of an information recommendation method according to an embodiment of the present disclosure. As shown in fig. 1, the information recommendation method 100 includes: step S101, behavior data of a user in a platform is obtained, wherein the behavior data comprises first click behavior data of the user on information in the platform and an attention list of the user, and the attention list comprises an author concerned by the user; step S102, second click behavior data of the user for the information issued by the author is obtained from the first click behavior data, and a first association value between the user and the author is determined based on the first click behavior data and the second click behavior data; step S103, determining the interest field of the user based on the behavior data; step S104, determining a second association value between the user and the author based on the matching degree of the interest field and the field where the author is located; step S105, determining the interest value of the user to the author based on the first relevance value and the second relevance value; and step S106, determining whether to recommend the information published by the author to the user based on the interest value.

It will be appreciated that a user may be interested in authors of different types, different areas, for various reasons, and only a portion of the interested authors may be interested and have consumer appeal. The method 100 determines the interest level of the user in the author of interest through the data of the matching degree of the user and the domain of the author of interest and the click behavior data of the user on the content published by the author of interest, and determines whether to recommend the information based on the interest level. Therefore, in the information recommendation process, the influence of the author on the information consumption behavior of the user is taken into consideration, the information recommendation accuracy rate can be improved, the fan reading amount of the author can be increased, and the whole click rate of the recommendation system is improved.

According to some embodiments, step S102 comprises: determining a first click rate of the user on information in the platform in a first time period and a second click rate of the user on information in the platform in a second time period based on the first click behavior data, wherein the first time period is longer than the second time period; determining a third click rate of the user on information published by the author in a first time period and a fourth click rate of the user on information published by the author in a second time period based on the second click behavior data; and determining the first correlation value based on the ratio of the third click rate to the first click rate and the ratio of the fourth click rate to the second click rate.

The user may have interest in different authors at different times or lose interest in concerned authors after a period of time, so that behavior data of the user at different periods of time can be extracted to determine behavior changes of the user. Specifically, the click rate of the user on the content published by the author concerned in different time periods may be obtained and compared with the click rate of the user on all information in the platform, for example, the click rate of the user on the content published by the author concerned in the past week and month may be obtained, the click rate of the user on all information in the past week and month may be obtained accordingly, and the first correlation value may be calculated based on the above scheme, so as to determine the interest value of the user on the author in the dimension of the click behavior over a certain time length.

In one example, different thresholds may be set for comparison with the first correlation value to determine the user's interest level in the author in the dimension of click behavior.

According to some embodiments, step S106 comprises: determining a first score of information published by the author based on the interest value, wherein the first score is used for indicating whether the information published by the author is recommended to the user. Thus, a ranking score of information published by an author is calculated based on the user's interest level in the author for determining whether to recommend the information published by the author.

According to some embodiments, step S106 further comprises: acquiring a second score of the information published by the author, wherein the second score is obtained based on the matching degree of the content of the information published by the author and the behavior data of the user; determining a weight value for a weighted calculation of the second score based on the interestingness value to determine the first score.

For example, the content of the information published by the author may be subjected to feature extraction and matching degree calculation based on the behavior picture of the user to determine a second score of the information published by the author. The above features may also be input into a conventional correlation model to directly obtain the second score.

Therefore, on the basis of ranking and scoring based on the content of the information, the method provided by the disclosure fuses the influence of the author on the information consumption behavior of the user, so that the final ranking result of information recommendation is comprehensively obtained based on the factors of both content understanding and the interest degree of the user in the author. Therefore, the information recommendation accuracy is improved, the fan reading amount of an author can be increased, and the whole click rate of the recommendation system is further improved.

According to some embodiments, determining the weight value based on the interest value comprises: in response to the interest value being greater than or equal to a threshold, determining a first sub-weight value based on the interest value, wherein the first sub-weight value is greater than 1; or in response to the interest value being less than the threshold, determining a second sub-weight value, wherein the second sub-weight value is equal to 1. It is to be understood that, when it is determined that the user is interested in the author, a first sub-weight value is used as a weight value for calculating a first score, and a second score of information posted by the author is subjected to weighted calculation with a weight value greater than 1 to obtain a larger first score, so that the information is more likely to be recommended to the user. And when it is determined that the user is not interested in the author, determining whether to recommend the information based on only the content of the information, with a second sub-weight value of 1 as a weight value for calculating the first score.

According to another aspect of the present disclosure, a training method of a machine learning model for information recommendation is provided. As shown in fig. 2, the training method 200 of the machine learning model includes: step S201, obtaining sample behavior data of a sample user, wherein the sample behavior data comprises click behavior data of the sample user on information in a platform and an attention list of the sample user, and the attention list comprises an author concerned by the sample user; step S202, marking the real interest value of the sample user, wherein the real interest value represents the interest degree of the sample user to the author; step S203, obtaining the domain information of the author; step S204, inputting the sample behavior data and the information of the field where the author is located into a machine learning model, and obtaining an interest prediction value of the sample user, wherein the interest prediction value indicates the prediction interest degree of the sample user for the author; step S205, calculating a loss value based on the real interest value and the predicted interest value; and step S206, adjusting parameters of the machine learning model based on the loss value.

Therefore, training of the machine learning model is achieved by constructing a training sample such as sample behavior data of a sample user and domain information of an author, marking a label value of the real interest value of the user, and adjusting parameters of the model based on the marked real interest value and the predicted interest value output by the model, so as to predict the interest degree of the author of the user. This prediction may be used to compute a ranking score for information published by the author to determine whether to recommend this information to the user. By introducing a machine learning technology to predict the interest degree of a user to a concerned author, the efficiency and accuracy of subsequent information recommendation can be improved.

According to another aspect of the present disclosure, an information recommendation apparatus is provided. As shown in fig. 3, the information recommendation apparatus 300 includes: a first obtaining module 301 configured to obtain behavior data of a user in a platform, wherein the behavior data includes first click behavior data of the user on information in the platform and a focus list of the user, and wherein the focus list includes authors focused on by the user; a second obtaining module 302, configured to obtain second click behavior data of information issued by the user for the author from the first click behavior data, and determine a first association value between the user and the author based on the first click behavior data and the second click behavior data; a first determination module 303 configured to determine a field of interest of the user based on the behavior data; a second determining module 304, configured to determine a second association value between the user and the author based on a matching degree between the interest field and a field in which the author is located; a third determination module 305 configured to determine an interest value of the user in the author based on the first relevance value and the second relevance value; and a fourth determination module 306 configured to determine whether to recommend the information published by the author to the user based on the interest value.

The operations of the modules 301-306 of the information recommendation device 300 are similar to the operations of the steps S101-S106 described above, and are not described herein again.

It will be appreciated that a user may be interested in different types of authors in different areas for various reasons, but only in a portion of the interested authors and have appeal for consumption. The information recommendation apparatus 300 determines the degree of interest of the user in the author of interest by using data of both the degree of matching between the fields of the user and the author of interest and the click behavior data of the user on the content distributed by the author of interest, and determines whether to recommend information based on the degree of interest. Therefore, in the information recommendation process, the influence of the author on the information consumption behavior of the user is taken into consideration, the information recommendation accuracy rate can be improved, the fan reading amount of the author can be increased, and the whole click rate of the recommendation system is improved.

According to some embodiments, the second obtaining module 302 comprises: a first determining unit configured to determine a first click rate of the user on information in a platform in a first time period and a second click rate of the user on information in the platform in a second time period based on the first click behavior data, wherein the first time period is longer than the second time period; a second determining unit configured to determine, based on the second click behavior data, a third click rate of the user on information published by the author in a first time period and a fourth click rate of the user on information published by the author in a second time period; and a third determination unit configured to determine the first correlation value based on a ratio of the third click rate to the first click rate and a ratio of the fourth click rate to the second click rate.

The user may have interest in different authors at different times or lose interest in an author concerned after a certain period of time, and therefore, the first determining unit and the second determining unit may respectively extract behavior data of the user at different periods of time to determine the behavior change of the user. Specifically, the click rate of the user on the content published by the author concerned in different time periods may be obtained and compared with the click rate of the user on all information in the platform, for example, the click rate of the user on the content published by the author concerned in the past week and month may be obtained, the click rate of the user on all information in the past week and month may be obtained accordingly, and the third determining unit may calculate the first correlation value to determine the interest value of the user on the author in the dimension of click behavior over a certain time length.

In one example, the second obtaining module 302 may be further configured to set different thresholds to compare with the first correlation value to determine a user's interest level in the author in the dimension of click behavior.

According to some embodiments, the fourth determination module 306 is further configured to: determining a first score of information published by the author based on the interest value, wherein the first score is used for indicating whether the information published by the author is recommended to the user. Thus, the fourth determination module 306 calculates a ranking score for information published by an author based on the user's level of interest in the author for determining whether to recommend the information published by the author.

According to some embodiments, the fourth determination module 306 comprises: an obtaining unit configured to obtain a second score of the information published by the author, wherein the second score is obtained based on a matching degree of the content of the information published by the author and the behavior data of the user; a fourth determination unit configured to determine a weight value for weighted calculation of the second score based on the interest value to determine the first score.

For example, the obtaining unit may extract features of the content of the information published by the author and perform calculation of the degree of matching based on the behavior image of the user to determine the second score of the information published by the author. The obtaining unit may also input the above features into a conventional correlation model to directly obtain the second score.

Thus, on the basis of ranking and scoring based on the content of the information, the information recommendation device 300 integrates the influence of the author on the information consumption behavior of the user, so that the final ranking result of the information recommendation is comprehensively obtained based on the factors of both content understanding and the interest degree of the user in the author. Therefore, the information recommendation accuracy is improved, the fan reading amount of an author can be increased, and the whole click rate of the recommendation system is further improved.

According to some embodiments, the fourth determination unit is configured to: in response to the interest value being greater than or equal to a threshold, determining a first sub-weight value based on the interest value, wherein the first sub-weight value is greater than 1; or in response to the interest value being less than the threshold, determining a second sub-weight value, wherein the second sub-weight value is equal to 1. It is to be understood that, when it is determined that the user is interested in the author, the fourth determination unit performs a weighted calculation in which a second score of the information issued by the author is weighted more than 1 with a first sub-weight value as a weight value for calculating the first score to obtain a larger first score, so that the information is more likely to be recommended to the user. And when it is determined that the user has no interest in the author, the fourth determination unit determines whether to recommend the information based only on the content of the information, with a second sub-weight value of 1 as a weight value for calculating the first score.

According to another aspect of the present disclosure, a training apparatus for a machine learning model for recommending information is provided. As shown in fig. 4, the training apparatus 400 for machine learning model includes: a third obtaining module 401 configured to obtain sample behavior data of a sample user, wherein the sample behavior data includes click behavior data of the sample user on information in a platform and an attention list of the sample user, and wherein the attention list includes an author concerned by the sample user; a marking module 402 configured to mark a real value of interest of the sample user, the real value of interest characterizing a degree of interest of the sample user in the author; a fourth obtaining module 403, configured to obtain domain information where the author is located; a fifth obtaining module 404, configured to input the sample behavior data and the domain information of the author into a machine learning model, and obtain an interest prediction value of the sample user, where the interest prediction value indicates a degree of interest of the sample user in prediction of the author; a calculation module 405 configured to calculate a loss value based on the real value of interest and the predicted value of interest; and an adjustment module 406 configured to adjust parameters of the machine learning model based on the loss value.

The operations of the

modules

401 and 406 of the training apparatus 400 for machine learning model are similar to those of the aforementioned steps S201-S206, and are not repeated herein.

Therefore, the training device 400 for the machine learning model realizes training of the machine learning model by constructing the training sample such as the sample behavior data of the sample user and the domain information of the author, marking the label value of the interest real value of the user, and adjusting the parameters of the model based on the marked interest real value and the interest predicted value output by the model, so as to predict the interest degree of the user on the author. This prediction may be used to compute a ranking score for information published by the author to determine whether to recommend this information to the user. By introducing a machine learning technology to predict the interest degree of a user to a concerned author, the efficiency and accuracy of subsequent information recommendation can be improved.

According to an embodiment of the present disclosure, there is also provided an electronic device, a readable storage medium, and a computer program product.

Referring to fig. 5, a block diagram of a structure of an electronic device 500, which may be a server or a client of the present disclosure, which is an example of a hardware device that may be applied to aspects of the present disclosure, will now be described. Electronic device is intended to represent various forms of digital electronic computer devices, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the disclosure described and/or claimed herein.

As shown in fig. 5, the electronic device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM)502 or a computer program loaded from a storage unit 508 into a Random Access Memory (RAM) 503. In the RAM503, various programs and data required for the operation of the electronic apparatus 500 can also be stored. The calculation unit 501, the ROM 502, and the RAM503 are connected to each other by a bus 504. An input/output (I/O) interface 505 is also connected to bus 504.

A number of components in the electronic device 500 are connected to the I/O interface 505, including: an input unit 506, an output unit 507, a storage unit 508, and a communication unit 509. The input unit 506 may be any type of device capable of inputting information to the electronic device 500, and the input unit 506 may receive input numeric or character information and generate key signal inputs related to user settings and/or function controls of the electronic device, and may include but not be limited toLimited to a mouse, keyboard, touch screen, track pad, track ball, joystick, microphone, and/or remote control. Output unit 507 may be any type of device capable of presenting information and may include, but is not limited to, a display, speakers, a video/audio output terminal, a vibrator, and/or a printer. The storage unit 508 may include, but is not limited to, a magnetic disk, an optical disk. The communication unit 509 allows the electronic device 500 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks, and may include, but is not limited to, a modem, a network card, an infrared communication device, a wireless communication transceiver, and/or a chipset, such as bluetooth^TMDevices, 802.11 devices, WiFi devices, WiMax devices, cellular communication devices, and/or the like.

The computing unit 501 may be a variety of general-purpose and/or special-purpose processing components having processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various 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, and so forth. The calculation unit 501 executes the respective methods and processes described above, such as the information recommendation method. For example, in some embodiments, the information recommendation method may be implemented as a computer software program tangibly embodied in a machine-readable medium, such as storage unit 508. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 500 via the ROM 502 and/or the communication unit 509. When the computer program is loaded into the RAM503 and executed by the computing unit 501, one or more steps of the information recommendation method described above may be performed. Alternatively, in other embodiments, the computing unit 501 may be configured to perform any of the methods described above in any other suitable manner (e.g., by way of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuitry, Field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), system on a chip (SOCs), Complex Programmable Logic Devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, receiving data and instructions from, and transmitting data and instructions to, a storage system, at least one input device, and at least one output device.

Program code for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program codes, when executed by the processor or controller, cause the functions/operations specified in the flowchart and/or block diagram to be performed. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of this disclosure, a machine-readable medium may 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 may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, 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.

To provide for interaction with a user, the systems and techniques described here 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 a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a 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 may be received in any form, including acoustic, speech, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), Wide Area Networks (WANs), and the Internet.

The computer system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server of a distributed system, or a server with a combined blockchain.

It should be understood that various forms of the flows shown above may be used, with steps reordered, added, or deleted. For example, the steps described in the present disclosure may be performed in parallel, sequentially or in different orders, and are not limited herein as long as the desired results of the technical solutions disclosed in the present disclosure can be achieved.

Although embodiments or examples of the present disclosure have been described with reference to the accompanying drawings, it is to be understood that the above-described methods, systems and apparatus are merely exemplary embodiments or examples and that the scope of the present invention is not limited by these embodiments or examples, but only by the claims as issued and their equivalents. Various elements in the embodiments or examples may be omitted or may be replaced with equivalents thereof. Further, the steps may be performed in an order different from that described in the present disclosure. Further, various elements in the embodiments or examples may be combined in various ways. It is important that as technology evolves, many of the elements described herein may be replaced with equivalent elements that appear after the present disclosure.

Claims

1. An information recommendation method, comprising:

acquiring behavior data of a user in a platform, wherein the behavior data comprises first click behavior data of the user on information in the platform and an attention list of the user, and the attention list comprises authors concerned by the user;

acquiring second click behavior data of the information issued by the author by the user from the first click behavior data, and determining a first association value between the user and the author based on the first click behavior data and the second click behavior data;

determining a field of interest of the user based on the behavior data;

determining a second association value between the user and the author based on the matching degree of the interest field and the field of the author;

determining an interest value of the user for the author based on the first associated value and the second associated value; and

and determining whether to recommend the information published by the author to the user based on the interest value.

2. The method of claim 1, wherein the determining a first association value between the user and the author based on the first click behavior data and the second click behavior data comprises:

determining a first click rate of the user on information in the platform in a first time period and a second click rate of the user on information in the platform in a second time period based on the first click behavior data, wherein the first time period is longer than the second time period;

determining a third click rate of the user on information published by the author in a first time period and a fourth click rate of the user on information published by the author in a second time period based on the second click behavior data; and

determining the first correlation value based on a ratio of the third click rate to the first click rate and a ratio of the fourth click rate to the second click rate.

3. The method of claim 1 or 2, wherein determining whether to recommend the information published by the author to the user based on the interest value comprises:

determining a first score of information published by the author based on the interest value, wherein the first score is used for indicating whether the information published by the author is recommended to the user.

4. The method of claim 3, wherein determining, based on the interest value, a first score of information published by the author comprises:

acquiring a second score of the information published by the author, wherein the second score is obtained based on the matching degree of the content of the information published by the author and the behavior data of the user;

determining a weight value for weighted calculation of the second score based on the interest value to determine the first score.

5. The method of claim 4, wherein determining the weight value based on the interest value comprises:

in response to the interest value being greater than or equal to a threshold, determining a first sub-weight value based on the interest value, wherein the first sub-weight value is greater than 1; or

In response to the interest value being less than the threshold, determining a second sub-weight value, wherein the second sub-weight value is equal to 1.

6. A method of training a machine learning model, comprising:

obtaining sample behavior data of a sample user, wherein the sample behavior data comprises click behavior data of the sample user on information in a platform and an attention list of the sample user, and the attention list comprises authors concerned by the sample user;

marking a real interest value of the sample user, wherein the real interest value represents the interest degree of the sample user to the author;

acquiring the information of the domain where the author is located;

inputting the sample behavior data and the information of the field where the author is located into a machine learning model, and obtaining an interest prediction value of the sample user, wherein the interest prediction value indicates the prediction interest degree of the sample user for the author;

calculating a loss value based on the real interest value and the predicted interest value; and

adjusting parameters of the machine learning model based on the loss value.

7. An information recommendation apparatus comprising:

a first obtaining module configured to obtain behavior data of a user in a platform, wherein the behavior data includes first click behavior data of the user on information in the platform and an attention list of the user, and wherein the attention list includes an author concerned by the user;

a second obtaining module configured to obtain second click behavior data of information issued by the user for the author from the first click behavior data, and determine a first association value between the user and the author based on the first click behavior data and the second click behavior data;

a first determination module configured to determine a field of interest of the user based on the behavior data;

a second determining module configured to determine a second association value between the user and the author based on a matching degree of the interest field and a field in which the author is located;

a third determination module configured to determine an interest value of the user in the author based on the first relevance value and the second relevance value; and

a fourth determination module configured to determine whether to recommend the information published by the author to the user based on the interest value.

8. The apparatus of claim 7, wherein the second obtaining means comprises:

a first determining unit configured to determine a first click rate of the user on information in a platform in a first time period and a second click rate of the user on information in the platform in a second time period based on the first click behavior data, wherein the first time period is longer than the second time period;

a second determining unit configured to determine, based on the second click behavior data, a third click rate of the user on information published by the author in a first time period and a fourth click rate of the user on information published by the author in a second time period; and

a third determination unit configured to determine the first correlation value based on a ratio of the third click rate to the first click rate and a ratio of the fourth click rate to the second click rate.

9. The apparatus of claim 7 or 8, wherein the fourth determining module is further configured to:

10. The apparatus of claim 9, wherein the fourth determining means comprises:

an acquisition unit configured to acquire a second score of information published by the author, wherein the second score is obtained based on a matching degree of content of the information published by the author and the behavior data of the user;

a fourth determination unit configured to determine a weight value for weighted calculation of the second score based on the interest value to determine the first score.

11. The apparatus of claim 10, wherein the fourth determination unit is configured to:

12. A training apparatus for a machine learning model, comprising:

a third obtaining module configured to obtain sample behavior data of a sample user, wherein the sample behavior data includes click behavior data of the sample user on information in a platform and an attention list of the sample user, and wherein the attention list includes authors interested by the sample user;

a marking module configured to mark a real interest value of the sample user, the real interest value characterizing a degree of interest of the sample user in the author;

the fourth acquisition module is configured to acquire the domain information of the author;

a fifth obtaining module, configured to input the sample behavior data and the domain information of the author into a machine learning model, and obtain an interest prediction value of the sample user, where the interest prediction value indicates a degree of predicted interest of the sample user in the author;

a calculation module configured to calculate a loss value based on the real value of interest and the predicted value of interest; and

an adjustment module configured to adjust parameters of the machine learning model based on the loss values.

13. An electronic device, comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein

The memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-6.

14. A non-transitory computer readable storage medium having stored thereon computer instructions for causing the computer to perform the method of any one of claims 1-6.

15. A computer program product comprising a computer program, wherein the computer program realizes the method of any one of claims 1-6 when executed by a processor.