Fitness course recommendation method and device, storage medium and electronic device

Abstract

This application discloses a method, apparatus, storage medium, and electronic device for recommending fitness courses. The method includes: obtaining a first preset number of fitness courses from a database based on the user characteristics of a target user; obtaining the course characteristics of each fitness course from the first preset number of fitness courses, and inputting the course characteristics of each fitness course and the user characteristics of the target user into a target model for processing to obtain a preference score for each fitness course by the target user; and determining a target fitness course from the first preset number of fitness courses based on the preference score, wherein the target fitness course is the fitness course to be recommended to the target user. This application solves the problem of high computational cost in related technologies for determining fitness courses to be recommended to users from a large number of fitness courses.

Description

Technical Field

[0001] This application relates to the field of fitness class recommendation technology, and more specifically, to a method, apparatus, storage medium, and electronic device for recommending fitness classes. Background Technology

[0002] Before users engage in fitness training, in order to facilitate their selection of preferred courses or to compile a list of courses that match their fitness needs, it is necessary to determine the courses to be recommended to them. Specifically, this involves selecting courses from a massive database and then identifying the selected courses as those to be recommended to the user. This process is computationally intensive, uses complex calculation methods, and the resulting courses may not align with user preferences.

[0003] There is currently no effective solution to the problem of the large amount of computation required to determine the fitness courses to be recommended to users from a large number of fitness courses in related technologies. Summary of the Invention

[0004] This application provides a method, apparatus, storage medium, and electronic device for recommending fitness courses, in order to solve the problem of high computational load in related technologies for determining fitness courses to be recommended to users from a large number of fitness courses.

[0005] According to one aspect of this application, a method for recommending fitness courses is provided. The method includes: obtaining a first preset number of fitness courses from a database based on user characteristics of a target user; obtaining course features of each fitness course from the first preset number of fitness courses, and inputting the course features of each fitness course and the user characteristics of the target user into a target model for processing to obtain a preference score for each fitness course by the target user, wherein the target model is trained from multiple sets of sample data, each set of sample data including course features, user features, and user preference tags for fitness courses; and determining a target fitness course from the first preset number of fitness courses based on the preference score, wherein the target fitness course is the fitness course to be recommended to the target user.

[0006] Optionally, before inputting the course features of each fitness course and the user features of the target user into the target model for processing to obtain the target user's preference score for each fitness course, the method further includes: obtaining the user features of multiple sample users and the course features of multiple sample fitness courses, and obtaining the cross features between the sample users and the sample fitness courses, wherein the cross features are used to characterize the training behavior features of the sample users towards the sample fitness courses; determining the preference labels of the sample users towards the sample fitness courses based on the cross features between the sample users and the sample fitness courses, wherein the preference labels include positive labels and negative labels; concatenating the user features of the sample users and the course features of the sample fitness courses based on each preference label to obtain multiple sets of sample data; and training a preset classification model based on the multiple sets of sample data to obtain the target model.

[0007] Optionally, determining the sample user's preference label for the sample fitness course based on the cross-features between the sample user and the sample fitness course includes: determining the sample user's preference label for the sample fitness course as a positive label if the cross-features include at least one of the following features: the sample user's behavior of repeatedly practicing the sample fitness course, or the sample user's behavior of giving a positive evaluation of the sample fitness course; determining the sample user's preference label for the sample fitness course as a negative label if the cross-features include at least one of the following features: the sample user's behavior of quitting the sample fitness course midway, or the sample user's behavior of giving a negative evaluation of the sample fitness course.

[0008] Optionally, obtaining a first preset number of fitness courses from multiple fitness courses in a fitness course database based on the user characteristics of the target user includes: determining the group to which the target user belongs based on the user characteristics of the target user, wherein the user characteristics include the user's attribute characteristics and exercise characteristics; determining the group's completion rate of each fitness course in the multiple fitness courses; and obtaining fitness courses from the multiple fitness courses whose completion rate of the group is greater than a preset value, thereby obtaining the first preset number of fitness courses.

[0009] Optionally, determining the target fitness course from the first preset number of fitness courses based on preference scores includes: sorting the first preset number of fitness courses according to the preference scores of each fitness course to obtain a sorting result; obtaining the top second preset number of fitness courses from the sorting result; and determining the second preset number of fitness courses as the target fitness courses.

[0010] Optionally, after determining the target fitness courses to be recommended to the target user from a first preset number of fitness courses based on preference scores, the method further includes: obtaining the target user's fitness goals, wherein the fitness goals are characterized by at least the following features: training type, training body part, and training duration; selecting a third preset number of fitness courses that match the fitness goals from the target fitness courses; matching the third preset number of fitness courses with multiple preset segments within a preset time interval to obtain a training course schedule; and recommending the training course schedule to the target user.

[0011] According to another aspect of this application, a fitness course recommendation device is provided. The device includes: a first acquisition unit, configured to acquire a first preset number of fitness courses from a database based on user characteristics of a target user; a second acquisition unit, configured to acquire course characteristics of each fitness course from the first preset number of fitness courses, and input the course characteristics of each fitness course and the user characteristics of the target user into a target model for processing to obtain a preference score for each fitness course by the target user, wherein the target model is trained from multiple sets of sample data, each set of sample data including course characteristics, user characteristics, and user preference tags for fitness courses; and a first determination unit, configured to determine a target fitness course from the first preset number of fitness courses based on the preference score, wherein the target fitness course is the fitness course to be recommended to the target user.

[0012] Optionally, the device further includes: a third acquisition unit, configured to acquire user features of multiple sample users and course features of multiple sample fitness courses before inputting the course features of each fitness course and the user features of the target user into the target model for processing to obtain the target user's preference score for each fitness course, and to acquire cross features between sample users and sample fitness courses, wherein the cross features are used to characterize the training behavior features of sample users towards sample fitness courses; a second determination unit, configured to determine the preference label of the sample user towards the sample fitness course based on the cross features between the sample user and the sample fitness course, wherein the preference label includes positive and negative labels; to concatenate the user features of the sample user and the course features of the sample fitness course based on each preference label to obtain multiple sets of sample data; and to train a preset classification model based on the multiple sets of sample data to obtain the target model.

[0013] According to another aspect of the present invention, a non-volatile storage medium is also provided, the non-volatile storage medium including a stored program, wherein the program, when running, controls the device where the non-volatile storage medium is located to perform a recommended method for a fitness course.

[0014] According to another aspect of the present invention, an electronic device is also provided, comprising a processor and a memory; the memory stores computer-readable instructions, and the processor is configured to execute the computer-readable instructions, wherein the computer-readable instructions, when executed, perform a recommended method for a fitness course.

[0015] This application employs the following steps: First, based on the user characteristics of the target user, a first preset number of fitness courses are obtained from multiple fitness courses in a database; then, the course characteristics of each fitness course within the first preset number of fitness courses are obtained, and the course characteristics of each fitness course, along with the user characteristics of the target user, are input into a target model for processing to obtain the target user's preference score for each fitness course. The target model is trained using multiple sets of sample data, each set including course characteristics, user characteristics, and user preference tags for the fitness courses; finally, based on the preference scores, a target fitness course is determined from the first preset number of fitness courses. This target fitness course is the fitness course to be recommended to the target user, thus solving the problem of high computational cost in determining the fitness course to be recommended to the user from a large number of fitness courses in related technologies. This improves the efficiency of determining the fitness course to be recommended to the user. Attached Figure Description

[0016] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0017] Figure 1 This is a flowchart of a method for recommending fitness courses according to an embodiment of this application;

[0018] Figure 2 This is a flowchart illustrating another method for recommending fitness classes according to embodiments of this application; and

[0019] Figure 3 This is a schematic diagram of a fitness course recommendation device provided according to an embodiment of this application. Detailed Implementation

[0020] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0021] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0022] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application 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 for the embodiments of this application described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises 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 apparatus.

[0023] According to an embodiment of this application, a method for recommending fitness classes is provided.

[0024] Figure 1 This is a flowchart of a recommended method for a fitness course according to an embodiment of this application. For example... Figure 1 As shown, the method includes the following steps:

[0025] Step S102: Obtain a first preset number of fitness courses from multiple fitness courses in the database based on the user characteristics of the target user.

[0026] Specifically, user characteristics can include attribute characteristics and exercise characteristics. Attribute characteristics can include gender, age, height, weight, etc. Exercise characteristics can be the user's training characteristics over a historical period, such as the characteristics of the courses attended and the completion rate of the courses. The database stores multiple fitness courses of various types suitable for different users. Based on the user characteristics of the target user, a first preset number of fitness courses with high relevance to the target user are selected.

[0027] Fitness courses highly relevant to the target user can be characterized by fitness courses highly relevant to the group to which the target user belongs. Optionally, in the fitness course recommendation method provided in this application embodiment, obtaining a first preset number of fitness courses from multiple fitness courses in the fitness course database based on the user characteristics of the target user includes: determining the group to which the target user belongs based on the user characteristics of the target user, wherein the user characteristics include the user's attribute characteristics and exercise characteristics; determining the group's completion rate of each fitness course in the multiple fitness courses; and obtaining fitness courses from the multiple fitness courses where the group's completion rate is greater than a preset value, thereby obtaining the first preset number of fitness courses.

[0028] In one optional implementation, if the target user is 20 years old, male, 170cm tall, and weighs 75kg, and the target user belongs to a group of young men with a slightly overweight physique, and the preset value is 80% exercise completion rate, then 600 fitness courses with an exercise completion rate greater than 80% are selected from 2,000 fitness courses, which is the first preset number of fitness courses.

[0029] For example, if the target user is 55 years old, female, 160cm tall, and weighs 44kg, and the target user belongs to the elderly female group with a slender build, and the preset value is 90% exercise completion rate, then from 2,000 fitness courses, fitness courses with an exercise completion rate of more than 90% for young males are selected, resulting in 200 fitness courses, which is the first preset number of fitness courses.

[0030] Step S104: Obtain the course features of each fitness course in the first preset number of fitness courses, and input the course features of each fitness course and the user features of the target user into the target model for processing to obtain the target user's preference score for each fitness course. The target model is trained by multiple sets of sample data, and each set of sample data includes course features, user features and user preference tags for fitness courses.

[0031] It should be noted that the characteristics of a fitness course can include course duration, course difficulty, types and number of movements included in the course, etc. The user's preference label for the fitness course can be "0" or "1", where "0" represents the lowest level of preference and "1" represents the highest level of preference. The target model obtains the target user's preference score for the fitness course based on the course characteristics of a fitness course and the user characteristics of the target user. The preference score is distributed between 0 and 1.

[0032] Step S106: Determine the target fitness course from the first preset number of fitness courses based on the preference score, wherein the target fitness course is the fitness course to be recommended to the target user.

[0033] In one optional implementation, the target user belongs to a group of young men who are overweight. From 2,000 fitness courses, 600 fitness courses are selected based on the completion rate of young men who are over 80%. The target user's preference score for these 600 fitness courses is obtained through target model processing. From these 600 fitness courses, fitness courses with a preference score greater than 0.7 are selected to obtain 50 fitness courses to be recommended to the target user.

[0034] The fitness course recommendation method provided in this application embodiment obtains a first preset number of fitness courses from a database based on the user characteristics of the target user; obtains the course characteristics of each fitness course in the first preset number of fitness courses, and inputs the course characteristics of each fitness course and the user characteristics of the target user into a target model for processing to obtain the target user's preference score for each fitness course. The target model is trained from multiple sets of sample data, each set of sample data including course characteristics, user characteristics, and user preference tags for fitness courses; and determines the target fitness course from the first preset number of fitness courses based on the preference score. The target fitness course is the fitness course to be recommended to the target user. This solves the problem of high computational cost in determining the fitness course to be recommended to the user from a large number of fitness courses in related technologies. Therefore, it improves the efficiency of determining the fitness course to be recommended to the user.

[0035] Before predicting the target user's preference score for a fitness course using the target model, the target model needs to be trained. Optionally, in the fitness course recommendation method provided in this application embodiment, before inputting the course features of each fitness course and the user features of the target user into the target model for processing to obtain the target user's preference score for each fitness course, the method further includes: obtaining the user features of multiple sample users and the course features of multiple sample fitness courses, and obtaining the cross features between the sample users and the sample fitness courses, wherein the cross features are used to characterize the training behavior features of the sample users for the sample fitness courses; determining the preference label of the sample user for the sample fitness courses based on the cross features between the sample users and the sample fitness courses, wherein the preference label includes positive labels and negative labels; concatenating the user features of the sample users and the course features of the sample fitness courses based on each preference label to obtain multiple sets of sample data; and training a preset classification model based on the multiple sets of sample data to obtain the target model.

[0036] Specifically, the sample data consists of two parts: features and labels. The feature part includes user features and cross features. User features include attribute features and activity features. Cross features are the user's training behavior characteristics in relation to fitness courses, which may include behaviors such as clicking on a fitness course, starting a fitness course, ending a fitness course, and exiting a fitness course. The label part consists of preference labels, indicating the degree of preference of the sample users for the sample fitness courses. Positive labels indicate preference, and negative labels indicate no preference.

[0037] In one optional implementation, user ID represents user characteristics, course ID represents course characteristics, and "0" or "1" represents preference tags. User ID, course ID, and tag information within a preset time period are concatenated to obtain multiple sets of sample data. The preset classification model is DeepFM. DeepFM is trained based on the multiple sets of sample data to obtain the target model.

[0038] Preference labels are determined by a user's training behavior in a fitness course over a period of time. Optionally, in the fitness course recommendation method provided in this application embodiment, determining the sample user's preference label for the sample fitness course based on the cross-features between the sample user and the sample fitness course includes: determining the sample user's preference label for the sample fitness course as a positive label if the cross-features include at least one of the following features: the sample user's behavior of repeatedly practicing the sample fitness course, or the sample user's behavior of giving a positive evaluation of the sample fitness course; determining the sample user's preference label for the sample fitness course as a negative label if the cross-features include at least one of the following features: the sample user's behavior of quitting the sample fitness course midway, or the sample user's behavior of giving a negative evaluation of the sample fitness course.

[0039] In one alternative implementation, fitness courses that are repeatedly practiced or have positive post-training evaluations are selected based on the user's training behavior over the past 30 days as positive samples, and a positive label "1" is assigned to the positive samples. Fitness courses with negative post-training evaluations and abnormal exits are selected as negative samples, and a negative label "0" is assigned to the negative samples.

[0040] Optionally, in the fitness course recommendation method provided in this application embodiment, determining the target fitness course from a first preset number of fitness courses based on preference scores includes: sorting the first preset number of fitness courses according to the preference scores of each fitness course to obtain a sorting result; obtaining the top second preset number of fitness courses from the sorting result; and determining the second preset number of fitness courses as the target fitness courses.

[0041] In one optional implementation, a first preset number of fitness courses are determined from 2,000 fitness courses, resulting in 600 fitness courses. The target user's preference score for the 600 fitness courses is obtained through target model processing, resulting in 600 preference scores. The 600 preference scores are distributed between 0 and 1. The fitness courses are sorted in order from 1 to 0, and the top 50 courses are selected as the target fitness courses to be recommended to the target user.

[0042] After obtaining the target fitness course, a training schedule to be recommended to the target user can be determined based on the target fitness course. Optionally, in the fitness course recommendation method provided in this application embodiment, after determining the target fitness course to be recommended to the target user from a first preset number of fitness courses based on preference scores, the method further includes: obtaining the target user's fitness goal, wherein the fitness goal is characterized by at least the following features: training type, training body part, and training duration; selecting a third preset number of fitness courses that match the fitness goal from the target fitness courses; matching the third preset number of fitness courses with multiple preset segments within a preset time interval to obtain a training schedule; and recommending the training schedule to the target user.

[0043] For example, if there are 50 fitness courses to recommend to the target user with the fitness goal of leg fat reduction, 20 courses matching the leg fat reduction goal can be selected from the 50 courses and arranged over two weeks to obtain a two-week leg training course list. This course list is selected from multiple courses in the database based on the user's user characteristics and training behavior. The fitness courses in this course list match the user's preferences, which helps the user stick to the training and thus achieve the corresponding training results.

[0044] Figure 2 This is a flowchart illustrating another method for recommending a fitness course according to an embodiment of this application. Figure 2 As shown, the method includes:

[0045] The system consists of an offline processing component and an online prediction component. The offline processing component includes model training and fitness course recall.

[0046] Specifically, model training involves concatenating user features, course features, and cross-features into a single sample, and then training a classification model based on this concatenated sample. This results in a target model used to predict users' preference scores for fitness courses, which is then stored in an HDFS database. It should be noted that since course features and cross-features are difficult to obtain online in real time, the course features and cross-features obtained during model training can be stored in a MongoDB database.

[0047] User features include attribute features (such as gender, age, height, weight, etc.) and activity features (such as the duration of recent running or training). Course features can include features such as the body parts being trained, course difficulty, and course duration. Cross-feature features are training behavior features, which can include behaviors such as clicking on a course, starting training, ending training, and exiting training. Sample concatenation can be performed by determining whether a user prefers a particular course based on cross-features over a period of time, obtaining preference labels, and then concatenating user features and course features using preference labels. For example, user ID represents user features, course ID represents course features, and "0" or "1" represents preference labels. By concatenating user IDs, course IDs, and label information within a preset time period, multiple sets of sample data are obtained.

[0048] The recall of fitness courses can be achieved by segmenting users based on their attribute characteristics, and recalling some features from multiple courses based on the completion rate of the target user's group. For example, the top 600 courses with the highest completion rate can be recalled from 2,000 courses, and the statistical results can be stored in the MongoDB database.

[0049] Furthermore, the system reads the recalled courses from the MongoDB database, retrieves the course features of the recalled courses and the cross-features of the target users when they participate in the recalled courses, and obtains the user features of the target users online. The user features, course features, and cross-features are input into the target model, which predicts the target users' preference scores for each recalled course. Specifically, the obtained data is between 0 and 1. The recalled courses are then filtered according to the preference scores. For example, the recalled courses are sorted according to the preference scores, and the top 20% of the features in the sorted list are extracted to obtain the courses preferred by the target users.

[0050] Through the embodiments of this application, a target model is trained offline and fitness courses related to the target user's group are recalled. The target model is then used online to predict the target user's preference score for the recalled courses, thereby determining the courses preferred by the user. This solves the problem of high computational cost in related technologies for determining fitness courses to be recommended to users from a large number of fitness courses, thus improving the efficiency of determining fitness courses to be recommended to users.

[0051] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0052] This application also provides a fitness course recommendation device. It should be noted that the fitness course recommendation device of this application embodiment can be used to execute the fitness course recommendation method provided in this application embodiment. The fitness course recommendation device provided in this application embodiment will be described below.

[0053] Figure 3 This is a schematic diagram of a recommended device for a fitness course according to an embodiment of this application. Figure 3 As shown, the device includes: a first acquisition unit 10, a second acquisition unit 20, and a first determination unit 30.

[0054] Specifically, the first acquisition unit 10 is used to acquire a first preset number of fitness courses from multiple fitness courses in the database based on the user characteristics of the target user.

[0055] The second acquisition unit 20 is used to acquire the course features of each fitness course in the first preset number of fitness courses, and input the course features of each fitness course and the user features of the target user into the target model for processing to obtain the target user's preference score for each fitness course. The target model is trained by multiple sets of sample data, and each set of sample data includes course features, user features and user preference tags for fitness courses.

[0056] The first determining unit 30 is used to determine a target fitness course from a first preset number of fitness courses based on a preference score, wherein the target fitness course is a fitness course to be recommended to the target user.

[0057] Optionally, in the fitness course recommendation device provided in this application embodiment, the device further includes: a third acquisition unit, configured to acquire user features of multiple sample users, course features of multiple sample fitness courses, and cross features between sample users and sample fitness courses before inputting the course features of each fitness course and the user features of the target user into the target model for processing to obtain the preference score of the target user for each fitness course; wherein the cross features are used to characterize the training behavior features of the sample user for the sample fitness course; a second determination unit, configured to determine the preference label of the sample user for the sample fitness course based on the cross features between the sample user and the sample fitness course, wherein the preference label includes positive labels and negative labels; concatenate the user features of the sample user and the course features of the sample fitness course based on each preference label to obtain multiple sets of sample data; and train a preset classification model based on the multiple sets of sample data to obtain the target model.

[0058] The fitness course recommendation device provided in this application embodiment obtains a first preset number of fitness courses from a database based on the user characteristics of the target user through a first acquisition unit 10; a second acquisition unit 20 obtains the course characteristics of each fitness course in the first preset number of fitness courses, and inputs the course characteristics of each fitness course and the user characteristics of the target user into a target model for processing to obtain the target user's preference score for each fitness course. The target model is trained by multiple sets of sample data, and each set of sample data includes course characteristics, user characteristics, and user preference tags for fitness courses; a first determination unit 30 determines the target fitness course from the first preset number of fitness courses based on the preference score. The target fitness course is the fitness course to be recommended to the target user. This solves the problem of large computational load in related technologies for determining the fitness course to be recommended to the user from a large number of fitness courses, thereby improving the efficiency of determining the fitness course to be recommended to the user.

[0059] The fitness course recommendation device includes a processor and a memory. The first acquisition unit 10, the second acquisition unit 20, and the first determination unit 30 are all stored in the memory as program units. The processor executes the program units stored in the memory to achieve the corresponding functions.

[0060] The processor contains a kernel, which retrieves the corresponding program unit from memory. One or more kernels can be configured, and adjusting kernel parameters can address the computationally intensive problem of determining which fitness courses to recommend to users from a large number of available courses in related technologies.

[0061] The memory may include non-permanent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0062] This application also provides a non-volatile storage medium, which includes a stored program, wherein the program, when running, controls the device where the non-volatile storage medium is located to execute a recommended method for a fitness course.

[0063] This application also provides an electronic device comprising a processor and a memory; the memory stores computer-readable instructions, and the processor executes the computer-readable instructions, wherein the computer-readable instructions, when executed, perform a recommended method for a fitness course. The electronic device described herein may be a server, PC, PAD, mobile phone, etc.

[0064] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0065] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0066] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0067] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0068] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0069] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0070] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0071] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0072] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0073] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for recommending fitness classes, characterized in that, include: Based on the user characteristics of the target user, a first preset number of fitness courses are obtained from multiple fitness courses in the database; The course features of each fitness course in the first preset number of fitness courses are obtained, and the course features of each fitness course and the user features of the target user are input into a target model for processing to obtain the target user's preference score for each fitness course. The target model is obtained by training a preset classification model with multiple sets of sample data. Each set of sample data includes course features, user features, and user preference labels for the fitness courses. The preference labels are determined based on the cross-features between the sample user and the sample fitness course. The cross-features characterize the sample user's training behavior characteristics for the sample fitness course. If the cross-features include at least one of the following features, the sample user's preference label for the sample fitness course is determined to be a positive label: the sample user repeatedly practices the sample fitness course; the sample user gives a positive evaluation of the sample fitness course. If the cross-features include at least one of the following features, the sample user's preference label for the sample fitness course is determined to be a negative label: the sample user quits the sample fitness course midway; the sample user gives a negative evaluation of the sample fitness course. Based on the preference score, a target fitness course is determined from the first preset number of fitness courses, wherein the target fitness course is a fitness course to be recommended to the target user; The process of retrieving a first preset number of fitness courses from multiple fitness courses in the database based on the target user's user characteristics includes: The target user's group is determined based on the user characteristics of the target user, wherein the user characteristics include the user's attribute characteristics and movement characteristics; Determine the group's completion rate for each of the multiple fitness classes; The fitness courses for which the group's exercise completion rate is greater than a preset value are obtained from the plurality of fitness courses, thus obtaining the first preset number of fitness courses.

2. The method according to claim 1, characterized in that, Before inputting the course features of each fitness course and the user features of the target user into the target model for processing to obtain the preference score of the target user for each fitness course, the method further includes: Obtain user characteristics of multiple sample users, course characteristics of multiple sample fitness courses, and obtain cross-features between the sample users and the sample fitness courses; The preference labels of the sample users for the sample fitness courses are determined based on the cross-features between the sample users and the sample fitness courses, wherein the preference labels include positive labels and negative labels; The multiple sets of sample data are obtained by concatenating the user characteristics of the sample users and the course characteristics of the sample fitness courses based on each of the preference tags. The target model is obtained by training a preset classification model based on the multiple sets of sample data.

3. The method according to claim 1, characterized in that, The step of determining the target fitness course from the first preset number of fitness courses based on the preference score includes: The first preset number of fitness courses are sorted according to the preference score of each fitness course to obtain the sorting result; Obtain the first second preset number of fitness courses from the sorted results; The second preset number of fitness courses is determined as the target fitness courses.

4. The method according to claim 1, characterized in that, After determining the target fitness course to be recommended to the target user from the first preset number of fitness courses based on the preference score, the method further includes: Obtain the fitness goals of the target user, wherein the fitness goals are characterized by at least the following features: training type, training body part, and training duration; Select a third preset number of fitness courses from the target fitness courses that match the fitness goal; The third preset number of fitness courses is matched with multiple preset segments within a preset time interval to obtain a training course schedule; The training course schedule is recommended to the target users.

5. A device for recommending fitness classes, characterized in that, include: The first acquisition unit is used to acquire a first preset number of fitness courses from multiple fitness courses in the database based on the user characteristics of the target user. The second acquisition unit is used to acquire the course features of each fitness course in the first preset number of fitness courses, and input the course features of each fitness course and the user features of the target user into a target model for processing to obtain the target user's preference score for each fitness course. The target model is obtained by training a preset classification model with multiple sets of sample data. Each set of sample data includes course features, user features, and user preference tags for the fitness courses. The preference tags are determined based on the cross features between the sample user and the sample fitness course. The cross features characterize the sample user's training behavior characteristics for the sample fitness course. If the cross features include at least one of the following characteristics, the sample user's preference tag for the sample fitness course is determined to be a positive tag: the sample user repeatedly practices the sample fitness course; the sample user gives a positive evaluation of the sample fitness course. If the cross features include at least one of the following characteristics, the sample user's preference tag for the sample fitness course is determined to be a negative tag: the sample user quits the sample fitness course midway; the sample user gives a negative evaluation of the sample fitness course. The first determining unit is configured to determine a target fitness course from the first preset number of fitness courses based on the preference score, wherein the target fitness course is a fitness course to be recommended to the target user. The device is further configured to: determine the group to which the target user belongs based on the user characteristics of the target user, wherein the user characteristics include the user's attribute characteristics and exercise characteristics; determine the group's completion rate of practice for each of the plurality of fitness courses; and obtain fitness courses from the plurality of fitness courses for which the group's completion rate of practice is greater than a preset value, thereby obtaining the first preset number of fitness courses.

6. The apparatus according to claim 5, characterized in that, The device further includes: The third acquisition unit is used to acquire user features of multiple sample users, course features of multiple sample fitness courses, and cross features between the sample users and the sample fitness courses before inputting the course features of each fitness course and the user features of the target user into the target model for processing to obtain the preference score of the target user for each fitness course. The second determining unit is used to determine the preference label of the sample user for the sample fitness course based on the cross features between the sample user and the sample fitness course, wherein the preference label includes positive label and negative label; The multiple sets of sample data are obtained by concatenating the user characteristics of the sample users and the course characteristics of the sample fitness courses based on each of the preference tags. The target model is obtained by training a preset classification model based on the multiple sets of sample data.

7. A non-volatile storage medium, characterized in that, The non-volatile storage medium includes a stored program, wherein, when the program is executed, it controls the device on which the non-volatile storage medium resides to perform the recommended method of the fitness course as described in any one of claims 1 to 4.

8. An electronic device, characterized in that, The device includes a processor and a memory, the memory storing computer-readable instructions, and the processor being configured to execute the computer-readable instructions, wherein the computer-readable instructions, when executed, perform the recommended method of the fitness course as described in any one of claims 1 to 4.

Images