A personalized exercise recommendation method for adolescent psychological crisis intervention

Abstract

The application discloses a personalized exercise recommendation method for psychological crisis intervention of teenagers, and comprises the following steps: inputting physical sample data and psychological sample data into a first model to obtain sample physical and mental characteristics; inputting an exercise scheme sample into a second model to obtain sample exercise characteristics; determining positive samples and negative samples from the sample exercise characteristics based on the sample physical and mental characteristics; obtaining a loss value according to the distance between the positive samples and the sample physical and mental characteristics and the distance between the negative samples and the sample physical and mental characteristics; training the first model and the second model by using the loss value; inputting to-be-processed physical data and psychological data into the trained first model to obtain predicted physical and mental characteristics; and determining a target exercise scheme as an exercise recommendation result according to the distance between the predicted physical and mental characteristics and exercise scheme characteristics. The application can provide effective exercises for improving psychological problems of teenagers and realize personalized and accurate psychological crisis intervention.

Description

Technical Field

[0001] This invention relates to the field of medical and health sciences, specifically to a personalized exercise recommendation method for intervention in adolescent psychological crises. Background Technology

[0002] Mental health is an important component of an individual's physical and mental well-being and a foundation for the comprehensive development of personality, especially important among adolescents. In recent years, the detection rate of mental health problems among Chinese adolescents has been rising continuously. Therefore, effective psychological crisis intervention for adolescents has become particularly urgent.

[0003] In recent years, an increasing number of researchers have begun to utilize advanced artificial intelligence technology to provide personalized psychological crisis intervention programs for adolescents. However, most existing psychological crisis intervention programs are centered on "cognitive strategies," which aim to alleviate negative emotions by guiding adolescents to engage in self-reflection. These programs are relatively effective for adolescents who are adept at rational analysis, but not for those who are not. For example, adolescents who tend to dwell on negative events are prone to repeatedly agonizing over problems, leading to increased anxiety or depression. For these adolescents, non-cognitive psychological crisis intervention programs are needed to avoid further exacerbating their emotional burden.

[0004] Existing research and educational practices have provided ample evidence that exercise not only helps improve physical fitness but also significantly promotes mood regulation, stress release, and self-efficacy. Therefore, exercise is increasingly recognized as an important component of psychological crisis intervention systems.

[0005] However, research on using exercise for psychological crisis intervention in adolescents is still in its early stages. Existing methods often recommend intervention exercises based solely on adolescents' physical fitness and interests, neglecting the differentiated effects of different exercises on psychological improvement. Numerous studies both domestically and internationally have shown that different types and intensities of exercise have clearly differentiated effects on improving various psychological problems such as depression, anxiety, and stress. If the appropriate type and intensity of exercise are not matched to the specific psychological problems of adolescents, the effectiveness of exercise intervention will be greatly reduced, or even fail to achieve the desired psychological improvement. Furthermore, there are significant differences in the physical development level, psychological characteristics, and responses to exercise stimuli among adolescents of different ages, and the psychological benefits of the same exercise will not be consistent across different age groups. Simply applying a one-size-fits-all exercise program to all age groups without considering age factors can easily lead to a significant decrease in the effectiveness of exercise intervention. Summary of the Invention

[0006] This invention provides a personalized exercise recommendation method for adolescent psychological crisis intervention, in order to solve the above-mentioned technical problems.

[0007] In a first aspect, the present invention provides a personalized exercise recommendation method for intervention in adolescent psychological crises, the method comprising:

[0008] Obtain physical fitness data, psychological data, and exercise program samples from adolescents;

[0009] The physical fitness sample data and psychological sample data are input into the first model to obtain the physical and mental representation of the sample; the exercise program sample is input into the second model to obtain the exercise representation of the sample.

[0010] Based on the physical and mental representations of the samples, positive and negative samples are determined from the motion representations of the samples;

[0011] The loss value is obtained based on the distance between the positive sample and the sample's mental and physical representation, and the distance between the negative sample and the sample's mental and physical representation.

[0012] The first model and the second model are trained using the loss value to obtain the trained first model and the trained second model;

[0013] The physical and psychological data to be processed are input into the trained first model to obtain the predicted physical and mental representation; and the target exercise program is determined as the exercise recommendation result based on the distance between the predicted physical and mental representation and the exercise program representation.

[0014] In some embodiments of the present invention, the positive sample includes a first effective motion representation, and the negative sample includes a first invalid motion representation, a second invalid motion representation, and a second effective motion representation. The first effective motion representation and the first invalid motion representation are sample motion representations corresponding to the same age group and the same psychological problem, and the second effective motion representation and the second invalid motion representation are sample motion representations of the same psychological problem across age groups. The loss value includes a first loss value and a second loss value.

[0015] The first loss value is calculated based on the first distance between the first effective motion representation and the sample physical and mental representation, and the second distance between the first ineffective motion representation and the sample physical and mental representation.

[0016] The second loss value is calculated based on the first loss value, according to the third distance between the sample's physical and mental representation and the second invalid movement representation, the second valid movement representation, and the first distance.

[0017] In some embodiments of the present invention, the first loss value It is obtained by calculation using the following formula:

[0018] ;

[0019] In the formula, It is the first effective representation of motion. It is the first ineffective motion characteristic. It is a physical and mental representation of the sample. It is the first distance. This is the second distance, where M is the total number of mental and physical representations in the sample. It's a hyperparameter;

[0020] Second loss value It is obtained by calculation using the following formula:

[0021] ;

[0022] In the formula, It's a hyperparameter. This includes the second ineffective motion representation and the second effective motion representation. It is the third distance.

[0023] In some embodiments of the present invention, the motion scheme sample is obtained in the following manner:

[0024] Based on the physical and psychological sample data, an exercise plan was developed.

[0025] Obtain psychological sample data after intervention training according to the formulated exercise plan;

[0026] Based on the psychological sample data after the intervention training, the effectiveness of the formulated exercise program is screened to obtain candidate exercise programs.

[0027] In some embodiments of the present invention, obtaining the exercise plan based on the physical fitness sample data and psychological sample data includes:

[0028] Identify psychological problems from the aforementioned psychological sample data;

[0029] The exercise plan is obtained based on the physical condition sample data and the psychological issues.

[0030] In some embodiments of the present invention, both the first model and the second model are based on deep neural networks.

[0031] Secondly, the present invention also provides a personalized exercise recommendation device for adolescent psychological crisis intervention, the device comprising:

[0032] The sample acquisition module is used to acquire physical fitness sample data, psychological sample data, and exercise program samples of adolescents.

[0033] The representation acquisition module is used to input the physical fitness sample data and psychological sample data into the first model to obtain the physical and mental representation of the sample; and to input the exercise program sample into the second model to obtain the exercise representation of the sample.

[0034] The sample acquisition module is used to determine positive and negative samples from the motion representation of the sample based on the sample's physical and mental representation;

[0035] The loss calculation module is used to obtain the loss value based on the distance between the positive sample and the sample's mental and physical representation, and the distance between the negative sample and the sample's mental and physical representation;

[0036] The model training module is used to train the first model and the second model using the loss value to obtain the trained first model and the trained second model;

[0037] The exercise recommendation module is used to input the physical and psychological data to be processed into the trained first model to obtain the predicted physical and mental representation; and to determine the target exercise program as the exercise recommendation result based on the distance between the predicted physical and mental representation and the exercise program representation.

[0038] Thirdly, the present invention also provides an electronic device, including a memory and a processor; the memory stores a computer program, and the processor is used to run the computer program in the memory to perform operations in the personalized exercise recommendation method for adolescent psychological crisis intervention provided in the first aspect.

[0039] Fourthly, the present invention also provides a storage medium storing a plurality of instructions adapted for loading by a processor to execute the steps in the personalized exercise recommendation method for adolescent psychological crisis intervention provided in the first aspect.

[0040] The personalized exercise recommendation method for adolescent psychological crisis intervention provided by this invention involves inputting physical fitness sample data and psychological sample data into a first model to obtain sample physical and mental representations; inputting exercise program samples into a second model to obtain sample exercise representations; determining positive and negative samples from the sample exercise representations based on the sample physical and mental representations; obtaining loss values ​​based on the distances between positive samples and the sample physical and mental representations, and the distances between negative samples and the sample physical and mental representations; training the first and second models using the loss values ​​to obtain trained first and second models; inputting the physical fitness data and psychological data to be processed into the trained first model to obtain predicted physical and mental representations; and determining the target exercise program as the exercise recommendation result based on the distance between the predicted physical and mental representations and the exercise program representations. This invention, by exploring the correlation between adolescents' physical and mental conditions and different exercises, matches adolescents with the most suitable and effective exercises for improving their psychological problems, thereby achieving personalized and precise psychological crisis intervention. Attached Figure Description

[0041] 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.

[0042] Figure 1 This is a flowchart illustrating the personalized exercise recommendation method for adolescent psychological crisis intervention provided in this embodiment of the invention. Detailed Implementation

[0043] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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 are within the scope of protection of the present invention.

[0044] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0045] "A and / or B" includes the following three combinations: A only, B only, and a combination of A and B.

[0046] The use of "applies to" or "configured to" in this invention implies an open and inclusive language, which does not exclude the applicability to or configuration to devices performing additional tasks or steps. Additionally, the use of "based on" implies openness and inclusivity, because processes, steps, calculations, or other actions "based on" one or more conditions or values ​​may in practice be based on additional conditions or values ​​beyond those conditions.

[0047] In this invention, the term "exemplary" is used to mean "serving as an example, illustration, or description." Any embodiment described as "exemplary" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0048] The following describes, with reference to the accompanying drawings, a personalized exercise recommendation method for adolescent psychological crisis intervention provided by an embodiment of the present invention.

[0049] like Figure 1 As shown in the figure, this invention provides a personalized exercise recommendation method for adolescent psychological crisis intervention, which includes the following steps:

[0050] S101, to obtain physical fitness sample data, psychological sample data and exercise program samples of teenagers.

[0051] In some examples, data from 90 students across three age groups (elementary, middle, and high school) are used as training samples, forming a student set S = {s1, s2, ..., s90}. This includes 30 elementary school students, 30 middle school students, and 30 high school students. For each student s... m (1≦m≦90), collect their physical and psychological sample data.

[0052] Student S m The physical fitness sample data is denoted as P. phy (s m )={p m1 , p m2 , ..., p mu}, where p mj (1≦j≦u) represents the assessment score of the j-th physical dimension. Physical dimensions may include height, weight, heart rate, and lung capacity. For example, student s1's physical data Pphy (s1)={145, 40, 82, 2100}.

[0053] Student S m The psychological sample data is denoted as P. psy (s m )={q m1 , q m2 , ..., q mv}, where q mk (1≦k≦v) represents the assessment score of the k-th psychological dimension. Psychological dimensions may include anxiety index, depression index, and stress index. The psychological dimension q with the highest assessment score from the psychological data is selected. m_max As a student m The main psychological problems. For example, the psychological data P of student s1. psy (s1)={65, 40, 55}, where the main psychological problem is anxiety.

[0054] S102, input the physical fitness sample data and psychological sample data into the first model to obtain the physical and mental representation of the sample; input the exercise program sample into the second model to obtain the exercise representation of the sample.

[0055] In some examples, both the first model and the second model are based on deep neural networks (DNNs).

[0056] For each student m (1≦m≦M), students s m Physical fitness sample data P phy (s m ) and psychological sample data P psy (s m The data is fed into the first model, DNN1, to obtain student s. m The sample's physical and mental representation H(s) m = [0.12, 0.55, 0.33, 0.80].

[0057] For each motion scheme sample e r (1≦r≦R), the motion type is encoded using one-heat encoding, and the motion intensity is encoded using hierarchical encoding. These are then fed into the second model, DNN2, to obtain the sample motion representation H(e). r For example, after uniquely encoding the motion type and hierarchically encoding the motion intensity of motion scheme sample e1, it becomes [1, 0, 0, 1, 20], resulting in the sample motion representation H(e). r = [0.10, 0.60, 0.35, 0.75].

[0058] S103, Based on the physical and mental representation of the sample, determine the positive and negative samples from the motion representation of the sample.

[0059] S104, obtain the loss value based on the distance between the positive sample and the sample's mental and physical representation, and the distance between the negative sample and the sample's mental and physical representation.

[0060] S105, the first model and the second model are trained using the loss value to obtain the trained first model and the trained second model.

[0061] S106, the physical and psychological data to be processed are input into the trained first model to obtain the predicted physical and mental representation; and based on the distance between the predicted physical and mental representation and the exercise plan representation, the target exercise plan is determined as the exercise recommendation result. Here, the exercise plan representation is the encoding result of all exercise plan samples by the trained second model.

[0062] In some examples, for new student s x to represent its physical and mental characteristics H(s) x The cosine similarity between the motion scheme representation and the candidate motion library E is calculated, and the three best motions with the highest cosine similarity are finally selected as the recommendation results.

[0063] This invention provides a personalized exercise recommendation method for adolescent psychological crisis intervention. The method involves inputting physical and psychological sample data into a first model to obtain sample physical and mental representations; inputting exercise program samples into a second model to obtain sample exercise representations; identifying positive and negative samples from the sample exercise representations based on these representations; obtaining loss values ​​based on the distances between positive and negative samples and their respective physical and mental representations; training the first and second models using these loss values ​​to obtain trained first and second models; inputting the physical and psychological data to be processed into the trained first model to obtain predicted physical and mental representations; and determining a target exercise program as the exercise recommendation result based on the distance between the predicted physical and mental representations and the exercise program representations. This invention explores the correlation between adolescents' physical and mental conditions and different exercises, thereby matching adolescents with the most suitable and effective exercises for improving their psychological problems, achieving personalized and precise psychological crisis intervention.

[0064] In some embodiments of the present invention, the positive sample includes a first effective motion representation, and the negative sample includes a first invalid motion representation, a second invalid motion representation, and a second effective motion representation. The first effective motion representation and the first invalid motion representation are sample motion representations corresponding to the same age group and the same psychological problem, and the second effective motion representation and the second invalid motion representation are sample motion representations of the same psychological problem across age groups. The loss value includes a first loss value and a second loss value.

[0065] The first loss value is calculated based on a first distance between the first effective motion representation and the sample physical and mental representation, and a second distance between the first ineffective motion representation and the sample physical and mental representation.

[0066] The second loss value is calculated based on the first loss value, according to the third distance between the sample's physical and mental representation and the second invalid movement representation, the second valid movement representation, and the first distance.

[0067] It should be noted that after obtaining the first loss value, the first model and the second model are trained using the first loss value to obtain the initially trained first model and the second model. The initially trained first model and the second model can be compared within the same age group.

[0068] After obtaining the second loss value, the first and second models that were initially trained are further trained using the second loss value to obtain the final trained first and second models. Based on learning "how to choose among peers", the final trained first and second models can further achieve cross-age group comparison.

[0069] In some embodiments of the present invention, the first loss value It is obtained by calculation using the following formula:

[0070] ;

[0071] In the formula, It is the first effective representation of motion. It is the first ineffective motion characteristic. It is a physical and mental representation of the sample. It is the first distance. This is the second distance, where M is the total number of mental and physical representations in the sample. It's a hyperparameter. =0.3. This is the function for calculating cosine similarity.

[0072] In some examples, for each student s m (1≦m≦M), using its sample physical and mental characteristics H(s) m) are used as anchor samples, and corresponding positive and negative sample sets are constructed based on them: positive sample set The first effective motor representation for the same age group and the same psychological problems; negative sample set This is the first ineffective movement representation for people of the same age and with the same psychological problems.

[0073] Understandably, the positive sample set and negative sample set The following method is used to determine the students: Select other students of the same age (physical data) and with the same psychological problems (psychological data) as the current students. If the exercise program for these students is effective, the exercise program representation is used as a positive sample; otherwise, it is used as a negative sample.

[0074] Based on the aforementioned first loss value, optimize the parameters of the first model DNN1 and the second model DNN2 to ensure the generated sample mental and physical representation H(s) m It is closer to the first effective motor representation of the same psychological problem in the same age group, and further away from the first ineffective motor representation of the same psychological problem in the same age group.

[0075] Second loss value It is obtained by calculation using the following formula:

[0076] ;

[0077] In the formula, It's a hyperparameter. =0.5, This includes the second ineffective motion representation and the second effective motion representation. It is the third distance.

[0078] In some examples, for each student s m (1≦m≦M), using its sample physical and mental characteristics H(s) m ) are used as anchor samples, and corresponding positive and negative sample sets are constructed based on them: positive sample set The negative sample set remains unchanged, except for ,Increase The second effective motor representation and the second ineffective motor representation for the same psychological problem across age groups.

[0079] Understandably, the negative sample set This involves using the exercise programs of other students from different age groups (physical data) and with the same psychological problems (psychological data) as negative samples. The positive sample set... and negative sample set Focusing on effective differentiation within the same age group, negative sample set The screening focuses on overall exclusion constraints across age groups.

[0080] Based on the aforementioned second loss value, optimize the parameters of the first model DNN1 and the second model DNN2 to ensure the generated sample mental and physical representation H(s) m It is closer to the first effective motor representation of the same psychological problem in the same age group, and further away from the second effective motor representation and the second ineffective motor representation of the same psychological problem across age groups.

[0081] The personalized exercise recommendation method for adolescent psychological crisis intervention provided in this invention is based on representation optimization of supervised comparative learning within the same age group; representation optimization of supervised comparative learning across age groups; selection of sports from a candidate sports library based on representation similarity; mining the correlation between adolescents' physical and mental conditions and different sports; and recommending the most suitable sports for adolescents to improve their psychological problems. This significantly improves the effectiveness of exercise recommendations in improving psychological problems and is applicable to any school psychological crisis intervention scenario.

[0082] In some embodiments of the present invention, the motion scheme sample is obtained in the following manner:

[0083] Based on the physical and psychological sample data, an exercise plan is developed.

[0084] Specifically, psychological problems are identified from the psychological sample data; and the exercise plan is obtained based on the physical fitness sample data and the psychological problems.

[0085] Obtain psychological sample data after intervention training according to the formulated exercise plan.

[0086] Based on the psychological sample data after the intervention training, the effectiveness of the formulated exercise program is screened to obtain candidate exercise programs.

[0087] In some examples, domain experts based on students' s m Physical fitness data P phy (s m ) and major psychological problems q m_max Customized exercise programs for them r ={t r i r , d r}, where t r For sports type, i r For exercise intensity, d r This represents the duration of the exercise. For example, student s1's exercise e1 = {jogging, light intensity, 20 minutes}.

[0088] Student S m According to movement e r Intervention training was conducted, and student data was collected again after the intervention training was completed. m Psychological data P postpsy (s m )={q post m1 , q post m2 , ..., q post mv}, where q post mk (1≦k≦v) represents the assessment score of the k-th psychological dimension after the intervention training is completed. For example, the psychological data P of student s1 post psy (s1)={59, 39, 50}.

[0089] Calculate motion scheme e r The effectiveness ∆q=q m_max -q post m_max If ∆q is greater than the specified threshold of 10, then motion scheme e is considered... r If a motion scheme is valid, it is included in the candidate motion library E; otherwise, the motion scheme e is considered invalid. r Invalid. For example, if the validity of motion e1 is ∆q=65-50=15>10, it means that motion e1 is valid and is included in the candidate motion library E.

[0090] This invention also provides a personalized exercise recommendation device for adolescent psychological crisis intervention, the device comprising:

[0091] The sample acquisition module is used to acquire physical fitness sample data, psychological sample data, and exercise program samples of adolescents.

[0092] The representation acquisition module is used to input the physical fitness sample data and psychological sample data into the first model to obtain the physical and mental representation of the sample; and to input the exercise program sample into the second model to obtain the exercise representation of the sample.

[0093] The sample acquisition module is used to determine positive and negative samples from the motion representation of the sample based on the sample's physical and mental representation;

[0094] The loss calculation module is used to obtain the loss value based on the distance between the positive sample and the sample's mental and physical representation, and the distance between the negative sample and the sample's mental and physical representation;

[0095] The model training module is used to train the first model and the second model using the loss value to obtain the trained first model and the trained second model;

[0096] The exercise recommendation module is used to input the physical and psychological data to be processed into the trained first model to obtain the predicted physical and mental representation; and to determine the target exercise program as the exercise recommendation result based on the distance between the predicted physical and mental representation and the exercise program representation.

[0097] The personalized exercise recommendation device for adolescent psychological crisis intervention provided in this embodiment of the invention corresponds to the personalized exercise recommendation method for adolescent psychological crisis intervention provided in the above embodiments, and will not be described again here.

[0098] Based on any of the above embodiments, another embodiment of the present invention provides an electronic device, as shown in the figure. This electronic device may include a processor, a communications interface, a memory, and a communication bus, wherein the processor, communications interface, and memory communicate with each other via the communication bus. The processor can call logical instructions in the memory to execute the aforementioned personalized exercise recommendation method for adolescent psychological crisis intervention.

[0099] Furthermore, the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, and can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0100] On the other hand, embodiments of the present invention also provide a storage medium storing a plurality of instructions adapted for loading by a processor to execute the personalized exercise recommendation method for adolescent psychological crisis intervention as provided in the above embodiments.

[0101] On the other hand, embodiments of the present invention also provide a computer program product, including a computer program that, when executed by a processor, implements the aforementioned personalized exercise recommendation method for adolescent psychological crisis intervention.

[0102] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0103] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.

[0104] The above provides a detailed description of a personalized exercise recommendation method for adolescent psychological crisis intervention provided by the embodiments of the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A personalized exercise recommendation method for adolescent psychological crisis intervention, characterized in that, The method includes: Obtain physical fitness data, psychological data, and exercise program samples from adolescents; The physical fitness sample data and psychological sample data are input into the first model to obtain the physical and mental representation of the sample; the exercise program sample is input into the second model to obtain the exercise representation of the sample. Based on the physical and mental representations of the samples, positive and negative samples are determined from the motion representations of the samples; The loss value is obtained based on the distance between the positive sample and the sample's mental and physical representation, and the distance between the negative sample and the sample's mental and physical representation. The first model and the second model are trained using the loss value to obtain the trained first model and the trained second model; The physical and psychological data to be processed are input into the trained first model to obtain the predicted physical and mental representation; and the target exercise program is determined as the exercise recommendation result based on the distance between the predicted physical and mental representation and the exercise program representation.

2. The personalized exercise recommendation method for adolescent psychological crisis intervention according to claim 1, characterized in that, The positive samples include a first effective motion representation, and the negative samples include a first ineffective motion representation, a second ineffective motion representation, and a second effective motion representation. The first effective motion representation and the first ineffective motion representation are sample motion representations corresponding to the same age group and the same psychological problem, and the second effective motion representation and the second ineffective motion representation are sample motion representations of the same psychological problem across age groups. The loss value includes a first loss value and a second loss value. The first loss value is calculated based on the first distance between the first effective motion representation and the sample physical and mental representation, and the second distance between the first ineffective motion representation and the sample physical and mental representation. The second loss value is calculated based on the first loss value, according to the third distance between the sample's physical and mental representation and the second invalid movement representation, the second valid movement representation, and the first distance.

3. The personalized exercise recommendation method for adolescent psychological crisis intervention according to claim 2, characterized in that, First loss value It is obtained by calculation using the following formula: ； In the formula, It is the first effective representation of motion. It is the first ineffective motion characteristic. It is a physical and mental representation of the sample. It is the first distance. This is the second distance, where M is the total number of mental and physical representations in the sample. It's a hyperparameter; Second loss value It is obtained by calculation using the following formula: ； In the formula, It's a hyperparameter. This includes the second ineffective motion representation and the second effective motion representation. It is the third distance.

4. The personalized exercise recommendation method for adolescent psychological crisis intervention according to claim 1, characterized in that, The motion scheme sample was obtained in the following way: Based on the physical and psychological sample data, an exercise plan was developed. Obtain psychological sample data after intervention training according to the formulated exercise plan; Based on the psychological sample data after the intervention training, the effectiveness of the formulated exercise program is screened to obtain candidate exercise programs.

5. The personalized exercise recommendation method for adolescent psychological crisis intervention according to claim 4, characterized in that, The step of obtaining and formulating an exercise plan based on the physical and psychological sample data includes: Identify psychological problems from the aforementioned psychological sample data; The exercise plan is obtained based on the physical condition sample data and the psychological issues.

6. The personalized exercise recommendation method for adolescent psychological crisis intervention according to claim 1, characterized in that, Both the first model and the second model are based on deep neural networks.

7. A personalized exercise recommendation device for adolescent psychological crisis intervention, characterized in that, The device includes: The sample acquisition module is used to acquire physical fitness sample data, psychological sample data, and exercise program samples of adolescents. The representation acquisition module is used to input the physical fitness sample data and psychological sample data into the first model to obtain the physical and mental representation of the sample; and to input the exercise program sample into the second model to obtain the exercise representation of the sample. The sample acquisition module is used to determine positive and negative samples from the motion representation of the sample based on the sample's physical and mental representation; The loss calculation module is used to obtain the loss value based on the distance between the positive sample and the sample's mental and physical representation, and the distance between the negative sample and the sample's mental and physical representation; The model training module is used to train the first model and the second model using the loss value to obtain the trained first model and the trained second model; The exercise recommendation module is used to input the physical and psychological data to be processed into the trained first model to obtain the predicted physical and mental representation; and to determine the target exercise program as the exercise recommendation result based on the distance between the predicted physical and mental representation and the exercise program representation.

8. An electronic device, characterized in that, It includes a memory and a processor; the memory stores a computer program, and the processor is used to run the computer program in the memory to perform the steps in the personalized exercise recommendation method for adolescent psychological crisis intervention as described in any one of claims 1 to 6.

9. A storage medium, characterized in that, The storage medium stores multiple instructions that are applicable to a processor for loading to execute the steps in the personalized exercise recommendation method for adolescent psychological crisis intervention as described in any one of claims 1 to 6.

Images