A user reading recommendation method, device, equipment and medium

Abstract

The application discloses a user reading recommendation method and device, equipment and storage medium. Including: dividing the obtained original user reading information sample to obtain a plurality of sub-user reading information samples; performing fusion iterative training on the sub-user reading information sample to obtain a final reading recommendation model; receiving the registration information of the user, and determining the target reading according to the registration information and the final reading recommendation model, and recommending the target reading to the user. By dividing the obtained original user reading information sample, and performing fusion iterative training on each sub-user reading information sample, when training with the sub-user reading information sample, the model parameters trained in the last sub-user reading information sample are referred to, thereby avoiding the problem of inaccurate model training caused by the forgetting problem in incremental learning, so that the final reading recommendation model obtained by fusion iteration can accurately recommend the reading to the user.

Description

Technical Field

[0001] This invention relates to the field of computer software technology, and in particular to a method, apparatus, device, and medium for recommending reading materials to users. Background Technology

[0002] With the rapid development of information technology, online reading has gradually become widely promoted. People can use fragmented time to read online, and in order to increase user stickiness, application developers now recommend reading materials to users based on information such as their reading habits.

[0003] However, currently, when recommending books to users, network models are usually used for prediction and recommendation. However, the training of prediction models is usually carried out by incremental learning. Incremental learning usually suffers from the problem of data forgetting. That is, when the model moves from one data period to another, it forgets the knowledge that has been learned, resulting in poor incremental learning performance and the recommended books not matching the user. Summary of the Invention

[0004] This invention provides a method, apparatus, device, and storage medium for recommending reading materials to users, so as to accurately recommend suitable reading materials to users.

[0005] According to one aspect of the present invention, a method for recommending user reading materials is provided, comprising: dividing the acquired original user reading information samples to obtain multiple sub-user reading information samples;

[0006] The final reading recommendation model is obtained by fusing and iteratively training the sub-user reading information samples.

[0007] The system receives the user's registration information, determines the target reading material based on the registration information and the final reading material recommendation model, and recommends the target reading material to the user.

[0008] According to another aspect of the present invention, a user reading recommendation device is provided, comprising: a sub-user reading information sample acquisition module, used to divide the acquired original user reading information sample to acquire multiple sub-user reading information samples;

[0009] The final reading recommendation model acquisition module is used to perform fusion and iterative training on the sub-user reading information samples to obtain the final reading recommendation model.

[0010] The target reading recommendation module is used to receive the user's registration information, determine the target reading material based on the registration information and the final reading material recommendation model, and recommend the target reading material to the user.

[0011] According to another aspect of the present invention, a computer device is provided, the computer device comprising:

[0012] At least one processor; and

[0013] A memory communicatively connected to the at least one processor; wherein,

[0014] The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform a microservice software refactoring method according to any embodiment of the present invention.

[0015] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement a microservice software refactoring method according to any embodiment of the present invention.

[0016] The technical solution of this invention divides the acquired original user reading information samples and performs fusion and iterative training on each acquired sub-user reading information sample. Thus, when training with sub-user reading information samples, the model parameters trained with the previous sub-user reading information sample are referenced, thereby avoiding the problem of inaccurate model training caused by the forgetting problem in incremental learning. Therefore, the final reading recommendation model obtained through fusion and iteration can make accurate reading recommendations to users.

[0017] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart of a user reading recommendation method provided in Embodiment 1 of the present invention;

[0020] Figure 2 This is a flowchart of a user reading recommendation method provided in Embodiment 2 of the present invention;

[0021] Figure 3 This is a schematic diagram of a user reading recommendation device according to Embodiment 3 of the present invention;

[0022] Figure 4 This is a schematic diagram of the structure of an electronic device provided in Embodiment 4 of the present invention. Detailed Implementation

[0023] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0024] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, apparatus, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0025] Example 1

[0026] Figure 1 This is a flowchart of a microservice software refactoring method provided in Embodiment 1 of the present invention. This embodiment is applicable to situations where microservices are refactored at the application level. The method can be executed by a microservice software refactoring device, which can be implemented in hardware and / or software and integrated into an electronic device. Figure 1 As shown, the method includes:

[0027] Step S101: Divide the obtained original user reading information samples into multiple sub-user reading information samples.

[0028] Optionally, the acquired original user reading information samples are divided to obtain multiple sub-user reading information samples, including: determining the collection time range corresponding to the original user reading information samples; dividing the original user reading information according to the order of months within the collection time range to obtain the sub-user reading information samples corresponding to each month.

[0029] Specifically, in this embodiment, after obtaining the original user reading information samples, the samples are preprocessed. This preprocessing may involve one-hot encoding of the samples, deleting all null and outlier values, and finally normalizing all samples to avoid the influence of different units of measurement. After processing, the samples are then partitioned to facilitate subsequent iterative training. The partitioning may be based on a range of data collected from the original user reading information samples. For example, if the reading information of 10 users within a year is obtained, the original user reading information will be partitioned according to the order of months, resulting in 12 sub-user reading information samples. For instance, the first sub-sample contains the reading information of the 10 users in January, and similarly, the twelfth sub-sample contains the reading information of the 10 users in December. Of course, this embodiment is merely illustrative and does not limit the data size of each sub-user reading information sample. Furthermore, in this implementation, the number of iteration cycles of the model will be determined based on the number of information samples read by the divided sub-users.

[0030] Step S102: Perform fusion and iterative training on the sub-user reading information samples to obtain the final reading recommendation model.

[0031] Optionally, the final reading recommendation model is obtained by fusing and iteratively training the sub-user reading information samples, including: obtaining a first reading recommendation model and a first prediction error information matrix by training the sub-user reading information samples of the previous month adjacent to the current traversal month, wherein the first prediction error information matrix includes the user reading prediction error of the next month adjacent to the current traversal month; iteratively training the first reading recommendation model based on the sub-user reading information samples corresponding to the current traversal month to obtain an intermediate reading recommendation model; correcting the intermediate reading recommendation model based on the first prediction error information matrix to obtain a corrected reading recommendation model corresponding to the current traversal month; when the current traversal month is determined to be the last month of the division, the corrected reading recommendation model is used as the final reading recommendation model.

[0032] Optionally, the intermediate reading recommendation model is corrected based on the first prediction error information matrix to obtain the corrected reading recommendation model corresponding to the current traversed month, including: obtaining the user reading prediction value of the intermediate reading recommendation model for the next month adjacent to the current traversed month; refining the user reading prediction value based on the first prediction error information matrix to obtain correction parameters, and correcting the intermediate reading recommendation model based on the correction parameters to obtain the corrected reading recommendation model.

[0033] Specifically, in this embodiment, after dividing the sample data, a fully connected neural network model is used to build a task scenario model. The first subset of user reading information samples is input into this model for training to obtain the basic model Model1. Since the sample data for February is known, the user attributes from the second subset are input into Model1 to obtain prediction data. The difference between the actual reading data and the predicted values ​​for each user in the second subset is calculated, and the prediction error information matrix for January is obtained based on the normalized result of the difference calculation. For example, Matric1 = [0.1 0.1 0.2 0.4 0.6 0.3 0.9 0.5 0.1 0.6]. However, to avoid data forgetting when using the second subset of user reading information samples for model iteration, Matric1 is applied during model training. When the second sample of user reading information from February is input into Model1, an intermediate reading recommendation model Model2' is obtained. Since the sample data for March is known, the user attribute information from the third sample is input into Model2' to obtain predicted values ​​{x1 x2 x3 ... x10}. Because these predicted values ​​are inaccurate, they need to be corrected with reference to Matric1, resulting in corrected predicted values ​​{x1 - 0.1x2 - 0.1x3 - 0.2 ... x10 - 0.6}. This correction process involves adjusting the parameters in Model2' to obtain correction parameters. Based on these correction parameters, Model2' is corrected to obtain a calibrated reading recommendation model Model2, and {x1 - 0.1x2 - 0.1x3 - 0.2 ... x10 - 0.6} is used as the predicted value output by Model2. Then, the difference between the actual data read by each user in the third sample and the predicted value is calculated, and the prediction error matrix for January is updated based on the normalized result of the difference calculation. For example, Similarly, iterative training and fusion are performed in the same way. When the current traversal month is determined to be December, the obtained corrected reading recommendation model Model12 is used as the final reading recommendation model.

[0034] In this embodiment, when using the divided subsamples for iterative training, in order to avoid data forgetting, the training of subsample data in each period will refer to the relevant parameters of the historical training process, rather than being carried out independently, thereby ensuring the accuracy of the final reading recommendation model.

[0035] Step S103: Receive the user's registration information, determine the target reading material based on the registration information and the final reading material recommendation model, and recommend the target reading material to the user.

[0036] Optionally, the target reading material is determined based on the registration information and the final reading material recommendation model, including: extracting the user's attribute information from the registration information, wherein the attribute information includes the user's reading type, reading duration and reading frequency; and obtaining the target reading material by predicting based on the attribute information through the final reading material recommendation model.

[0037] Optionally, recommending the target reading material to the user includes: obtaining the recommendation method pre-selected by the user, wherein the recommendation method includes SMS recommendation or terminal recommendation; and sending the target reading material to the user according to the recommendation method.

[0038] Specifically, in this embodiment, for the user to be recommended, since the user has already filled in relevant attribute information such as reading type, reading duration, and reading frequency during registration according to the recommendation requirements, after obtaining the final reading recommendation model, the target reading material suitable for the user can be directly determined by inputting the above attribute information into the model. In addition, this embodiment can also obtain the historical reading data of the user to be recommended. The above attribute information can also be obtained by analyzing the historical reading data. Therefore, this embodiment does not limit the specific method of obtaining the attribute information; as long as the attribute information of the user to be recommended can be obtained, it is within the scope of protection of this application, and this embodiment does not limit it.

[0039] It should be noted that since users have already selected a recommendation method during registration, such as SMS recommendation or terminal recommendation, once a target book is identified, it can be sent to the user according to their chosen method. For example, if the user is not near a terminal, SMS recommendation can be selected, and a link to the target book will be sent to the user via SMS, allowing them to open the book by clicking the link. Conversely, if the user is near a terminal and can clearly see the screen, terminal recommendation can be selected, displaying the target book's content directly on the screen for more intuitive and efficient reading. This embodiment is merely illustrative and does not limit the specific recommendation method for target books.

[0040] This implementation divides the acquired original user reading information samples and performs fusion and iterative training on each sub-user reading information sample. When training with sub-user reading information samples, the model parameters trained with the previous sub-user reading information sample are referenced, thus avoiding the inaccurate model training caused by the forgetting problem in incremental learning. Therefore, the final reading recommendation model obtained through fusion and iteration can make accurate reading recommendations to users.

[0041] Example 2

[0042] Figure 2 This is a flowchart of a user reading recommendation method provided in Embodiment 2 of the present invention. Based on the above embodiments, after recommending the target reading to the user, the method further includes: receiving recommendation feedback information from the user regarding the target reading; and amplifying the original user reading information sample based on the recommendation feedback information. For example... Figure 2 As shown, the method includes:

[0043] Step S201: Divide the obtained original user reading information samples into multiple sub-user reading information samples.

[0044] Optionally, the acquired original user reading information samples are divided to obtain multiple sub-user reading information samples, including: determining the collection time range corresponding to the original user reading information samples; dividing the original user reading information according to the order of months within the collection time range to obtain the sub-user reading information samples corresponding to each month.

[0045] Step S202: Perform fusion and iterative training on the sub-user reading information samples to obtain the final reading recommendation model.

[0046] Optionally, the final reading recommendation model is obtained by fusing and iteratively training the sub-user reading information samples, including: obtaining a first reading recommendation model and a first prediction error information matrix by training the sub-user reading information samples of the previous month adjacent to the current traversal month, wherein the first prediction error information matrix includes the user reading prediction error of the next month adjacent to the current traversal month; iteratively training the first reading recommendation model based on the sub-user reading information samples corresponding to the current traversal month to obtain an intermediate reading recommendation model; correcting the intermediate reading recommendation model based on the first prediction error information matrix to obtain a corrected reading recommendation model corresponding to the current traversal month; when the current traversal month is determined to be the second to last month of the division, the corrected reading recommendation model is used as the final reading recommendation model.

[0047] Optionally, the intermediate reading recommendation model is corrected based on the first prediction error information matrix to obtain the corrected reading recommendation model corresponding to the current traversed month, including: obtaining the user reading prediction value of the intermediate reading recommendation model for the next month adjacent to the current traversed month; refining the user reading prediction value based on the first prediction error information matrix to obtain correction parameters, and correcting the intermediate reading recommendation model based on the correction parameters to obtain the corrected reading recommendation model.

[0048] Step S203: Receive the user's registration information, determine the target reading material based on the registration information and the final reading material recommendation model, and recommend the target reading material to the user.

[0049] Optionally, the target reading material is determined based on the registration information and the final reading material recommendation model, including: extracting the user's attribute information from the registration information, wherein the attribute information includes the user's reading type, reading duration and reading frequency; and obtaining the target reading material by predicting based on the attribute information through the final reading material recommendation model.

[0050] Optionally, recommending the target reading material to the user includes: obtaining the recommendation method pre-selected by the user, wherein the recommendation method includes SMS recommendation or terminal recommendation; and sending the target reading material to the user according to the recommendation method.

[0051] Step S204: Receive the user's recommended feedback information for the target; amplify the original user reading information sample based on the recommended feedback information.

[0052] Specifically, in this embodiment, after recommending the target reading material to the user, the system will also receive user feedback information regarding the target reading material. The feedback information may include either "satisfied" or "unsatisfied". If the feedback information is determined to be "satisfied", the system will obtain the attribute information of the user to be recommended and the target reading material to be recommended during this recommendation process, and add the above information as new sample data to the original user reading information sample to amplify the sample data.

[0053] It should be noted that, in this embodiment, by augmenting the original user reading information samples, the data capacity of the samples can be increased, which facilitates the subsequent retraining of the model based on the augmented sample data as needed, thereby further improving the accuracy of the recommendation model.

[0054] This implementation divides the acquired original user reading information samples and performs fusion and iterative training on each sub-user reading information sample. When training with sub-user reading information samples, the model parameters trained with the previous sub-user reading information sample are referenced, thus avoiding the inaccurate model training caused by the forgetting problem in incremental learning. Therefore, the final reading recommendation model obtained through fusion and iteration can make accurate reading recommendations to users.

[0055] Example 3

[0056] Figure 3 This is a schematic diagram of a user reading recommendation device provided in Embodiment 3 of the present invention. Figure 3 As shown, the device includes: a sub-user reading information sample acquisition module 310, a final reading recommendation model acquisition module 320, and a target reading recommendation module 330.

[0057] The sub-user reading information sample acquisition module 310 is used to divide the acquired original user reading information sample and acquire multiple sub-user reading information samples.

[0058] The final reading recommendation model acquisition module 320 is used to fuse and iteratively train the sub-user reading information samples to obtain the final reading recommendation model.

[0059] The target reading recommendation module 330 is used to receive the user's registration information, determine the target reading based on the registration information and the final reading recommendation model, and recommend the target reading to the user.

[0060] Optionally, a sub-user reading information sample acquisition module is used to determine the collection time range corresponding to the original user reading information sample;

[0061] Within the time range of data collection, the original user reading information is divided into months in chronological order to obtain sub-user reading information samples for each month.

[0062] Optionally, the final reading recommendation model acquisition module is used to acquire the first reading recommendation model and the first prediction error information matrix obtained by training the sub-user reading information samples of the previous month adjacent to the current traversed month. The first prediction error information matrix includes the user reading prediction error of the next month adjacent to the current traversed month.

[0063] The first reading recommendation model is iteratively trained based on the sub-user reading information samples corresponding to the current month to obtain the intermediate reading recommendation model;

[0064] The intermediate reading recommendation model is corrected based on the first prediction error information matrix to obtain the corrected reading recommendation model corresponding to the current traversed month.

[0065] When the current traversed month is determined to be the second to last month of the division, the corrected reading recommendation model is used as the final reading recommendation model.

[0066] Optionally, the final reading recommendation model acquisition module is used to obtain the user reading prediction values ​​of the intermediate reading recommendation model for the next month adjacent to the current traversed month;

[0067] The predicted values ​​of user reading materials are refined based on the first prediction error information matrix to obtain correction parameters, and the intermediate reading material recommendation model is corrected based on the correction parameters to obtain the corrected reading material recommendation model.

[0068] Optionally, the target reading recommendation module includes a target reading determination unit, used to extract user attribute information from registration information, wherein the attribute information includes the user's reading type, reading duration, and reading frequency;

[0069] The final reading recommendation model uses attribute information to predict and obtain target reading materials.

[0070] Optionally, the target reading recommendation module includes a target reading recommendation unit, used to obtain the recommendation method pre-selected by the user, wherein the recommendation method includes SMS recommendation or terminal recommendation;

[0071] Send the target reading material to the user according to the recommended method.

[0072] Optionally, the device also includes a sample amplification module for receiving recommended feedback information from the user for the target;

[0073] The original user reading information sample was expanded based on the recommendation feedback information.

[0074] The user reading recommendation device provided in this embodiment of the invention can execute a user reading recommendation method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of executing the method.

[0075] Example 4

[0076] Figure 4 A schematic diagram of an electronic device 10, which can be used to implement embodiments of the present invention, is shown. The electronic device is intended to represent various forms of digital computers, 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. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0077] The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the invention described and / or claimed herein.

[0078] like Figure 4As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0079] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0080] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as user reading recommendation methods.

[0081] In some embodiments, the user reading recommendation method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the user reading recommendation method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the user reading recommendation method by any other suitable means (e.g., by means of firmware).

[0082] Various embodiments of the apparatuses and techniques described above herein can be implemented in digital electronic circuit devices, integrated circuit devices, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), device-on-a-chip (SoC) devices, complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable device including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage device, at least one input device, and at least one output device, and transmitting data and instructions to the storage device, the at least one input device, and the at least one output device.

[0083] Computer programs used to implement the methods of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable overhead crane operation alarm device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The computer programs can be executed entirely on the machine, partially on the machine, as a standalone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0084] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution means, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor devices, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0085] To provide interaction with a user, the apparatus and techniques described herein can be implemented on a device having: a display device (e.g., a touchscreen) for displaying information to the user; and buttons through which the user can provide input to the device. Other types of apparatus can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or haptic feedback); and input from the user can be received in any form (including voice input, speech input, or haptic input).

[0086] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0087] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for recommending reading materials to users, characterized in that, include: The original user reading information samples are divided to obtain multiple sub-user reading information samples; The final reading recommendation model is obtained by fusing and iteratively training the sub-user reading information samples. Receive user registration information, determine target reading materials based on the registration information and the final reading material recommendation model, and recommend the target reading materials to the user; The step of fusing and iteratively training the sub-user reading information samples to obtain the final reading recommendation model includes: obtaining the first reading recommendation model and the first prediction error information matrix obtained by training the sub-user reading information samples of the previous month adjacent to the current traversal month, wherein the first prediction error information matrix includes the user reading prediction error of the next month adjacent to the current traversal month. The first reading recommendation model is iteratively trained based on the sub-user reading information samples corresponding to the current month to obtain an intermediate reading recommendation model; The intermediate reading recommendation model is corrected based on the first prediction error information matrix to obtain the corrected reading recommendation model corresponding to the current traversed month. When it is determined that the current traversed month is the last month of the division, the corrected reading recommendation model is used as the final reading recommendation model.

2. The method according to claim 1, characterized in that, The process of dividing the acquired original user reading information samples into multiple sub-user reading information samples includes: Determine the time range corresponding to the original user reading information sample; Within the specified collection time range, the original user reading information is divided according to the order of months to obtain the sub-user reading information sample corresponding to each month.

3. The method according to claim 1, characterized in that, The step of correcting the intermediate reading recommendation model based on the first prediction error information matrix to obtain the corrected reading recommendation model corresponding to the current traversed month includes: Obtain the predicted reading content for the next month adjacent to the current month from the intermediate reading recommendation model; The user's reading material prediction value is corrected based on the first prediction error information matrix to obtain correction parameters, and the intermediate reading material recommendation model is corrected based on the correction parameters to obtain the corrected reading material recommendation model.

4. The method according to claim 1, characterized in that, The step of determining the target reading material based on the registration information and the final reading material recommendation model includes: The user's attribute information is extracted from the registration information, wherein the attribute information includes the user's reading type, reading duration, and reading frequency; The target reading material is obtained by predicting based on the attribute information using the final reading material recommendation model.

5. The method according to claim 1, characterized in that, The step of recommending the target reading material to the user includes: Obtain the recommendation method pre-selected by the user, wherein the recommendation method includes SMS recommendation or terminal recommendation; The target reading material is sent to the user according to the recommended method.

6. The method according to any one of claims 1 to 5, characterized in that, After recommending the target reading material to the user, the method further includes: Receive user recommendation feedback information regarding the target reading material; The original user reading information sample is amplified based on the recommendation feedback information.

7. A user reading recommendation device, characterized in that, include: The sub-user reading information sample acquisition module is used to divide the acquired original user reading information sample and obtain multiple sub-user reading information samples. The final reading recommendation model acquisition module is used to perform fusion and iterative training on the sub-user reading information samples to obtain the final reading recommendation model. The target reading recommendation module is used to receive the user's registration information, determine the target reading material based on the registration information and the final reading material recommendation model, and recommend the target reading material to the user. The final reading recommendation model acquisition module is used to acquire the first reading recommendation model and the first prediction error information matrix obtained by training the sub-user reading information samples of the previous month adjacent to the current traversal month. The first prediction error information matrix includes the user reading prediction error of the next month adjacent to the current traversal month. The first reading recommendation model is iteratively trained based on the sub-user reading information samples corresponding to the current month to obtain an intermediate reading recommendation model; The intermediate reading recommendation model is corrected based on the first prediction error information matrix to obtain the corrected reading recommendation model corresponding to the current traversed month. When it is determined that the current traversed month is the last month of the division, the corrected reading recommendation model is used as the final reading recommendation model.

8. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1-6.

9. A storage medium for computer-executable instructions, wherein a computer program is stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-6.

Images