An intelligent fitness recommendation method and system based on big data analysis

Abstract

The present invention discloses an intelligent fitness recommendation method and system based on big data analysis, and specifically relates to the technical field of fitness recommendation. The method and system obtain the user's training progress index, training load difference index and body feedback index to obtain the applicability coefficient of the current fitness plan, and judge whether to change the user's current fitness plan according to the applicability coefficient. When the user's current fitness plan needs to be changed, a training goal questionnaire and current training index data are sent to the user, and the user's training goal is received. According to the received user training data, an action feature set corresponding to each training goal is determined and sent to the user for selection, and a final fitness plan is generated according to the user's selection result. In this way, the applicability of the user's current fitness plan can be judged in time, and the user's fitness plan can be recommended or changed in time, so that the user's fitness effect is more ideal and the impact on the user's physical health is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of fitness recommendation, and in particular to an intelligent fitness recommendation method and system based on big data analysis. Background Art

[0002] With the popularity of fitness, more and more people choose to make fitness a part of their lives. In order to exercise more scientifically and systematically, many people choose to use fitness platforms to develop scientific fitness plans. Fitness platforms will develop scientific fitness plans based on users' fitness goals, thereby maximizing the fitness effect on users.

[0003] However, in the actual fitness of users, as the user's fitness progresses, the fitness plan originally formulated by the fitness platform may no longer be suitable for the user's current fitness status. Failure to recommend or change the user's fitness plan in a timely manner may lead to unsatisfactory fitness results for the user and even affect the user's physical health. Summary of the invention

[0004] The purpose of the present invention is to solve the problem and propose an intelligent fitness recommendation method and system based on big data analysis.

[0005] In a first aspect of the present invention, a smart fitness recommendation method based on big data analysis is first proposed, the method comprising:

[0006] Obtaining the user's training progress data, training load data, and physical feedback data, and extracting a training progress target index, a training load difference index, and a physical feedback index from the training progress data, the training load data, and the physical feedback data;

[0007] The training progress target index, the training load difference index and the body feedback index are used in a fuzzy inference system to obtain the applicability coefficient of the current fitness program, and whether to change the user's current fitness program is determined according to the applicability coefficient;

[0008] When the user's current fitness program needs to be changed, a training goal questionnaire and current training index data are sent to the user, and the user's training goal is received;

[0009] According to the received user training data, the action feature set corresponding to each training target is determined, and the action feature set corresponding to each training target is sent to the user for selection, and the final fitness plan is generated according to the user's selection result.

[0010] Optionally, the method for obtaining the training progress target index is:

[0011] Get the actual value of each specific training goal of the user and the initial value corresponding to each specific training goal from the beginning of the fitness plan to the current time period, and mark the actual value of each specific training goal and the initial value corresponding to each specific training goal as and , The number indicating the number of specific training targets, =1, 2, 3, 4, ..., , is a positive integer, Indicates the total number of specific training targets;

[0012] The progress ratio of each specific training target is calculated based on the actual value of each specific training target and the initial value corresponding to each specific training target. The calculation formula is: , For the The progress ratio of a specific training goal;

[0013] Compare the progress ratio of each specific training target with the expected progress ratio range, and re-mark the specific training targets that are not within the expected progress ratio range as , The number of specific training targets that are not within the expected progress ratio range. =1, 2, 3, 4, ..., , is a positive integer, Indicates the total number of specific training targets that are not within the expected progression ratio range;

[0014] Calculate the training progress index. The specific calculation expression is: , where It is the training progress index.

[0015] Optionally, the method for obtaining the training load difference index is:

[0016] Get the load percentage of each load indicator trained by the user from the beginning of the fitness program to the current time period. The calculation formula is: , Indicates The current load of the load indicator, Indicates The base load of a load index, For the The load percentage of each load indicator; =1, 2, 3, 4, ..., , is a positive integer, Indicates the total number of load indicators;

[0017] Calculate the training load index using the formula: , where is the training load index;

[0018] Calculate the training load difference index using the formula: , where is the training load difference index, is the preset training load threshold.

[0019] Optionally, the method for obtaining the body feedback index is:

[0020] Get the score of each physical feeling indicator after each training session from the start of the fitness program to the current time period, and get the feedback index of each physical feeling indicator based on the score of each physical feeling indicator and the corresponding minimum and maximum scores. The calculation formula is: , where Indicates The scores of the physical feeling indicators, Indicates The preset minimum scores for each physical feeling indicator, Indicates The preset maximum score of each physical feeling indicator. For the Feedback index of physical feeling indicators; =1, 2, 3, 4, ..., , is a positive integer, a represents the total number of body feeling indicators;

[0021] Calculate the feedback index of the body feeling index after each training session , the calculation formula is: ; In the formula, Indicates The feedback index of the body feeling index after the training is completed. =1, 2, 3, 4, ..., , is a positive integer, Indicates the total number of trainings;

[0022] Calculate the body feedback index using the formula: ; In the formula, It is the body feedback index.

[0023] Optionally, the training progress target index, the training load difference index and the body feedback index are used in a fuzzy inference system to obtain a suitability coefficient of the current fitness program, including:

[0024] The training progress index, training load difference index and body feedback index are used as input items of fuzzy rules, and the applicability coefficient of the current fitness program is used as output item;

[0025] Fuzzify the input items and convert the exact input values ​​into fuzzy sets;

[0026] Define fuzzy rules based on the fuzzy sets of input items to map input items to output items;

[0027] According to the fuzzy rules, the input items are inferred to determine the fuzzy value of the output items;

[0028] For each rule, calculate the membership of its predecessor and take the minimum value as the activation of the rule;

[0029] Synthesize the results of all rules and calculate the fuzzy set of output items;

[0030] The fuzzy value of the output item is converted into an exact value, and the applicability coefficient is output according to the defuzzification result.

[0031] Optionally, judging whether to change the user's current fitness program according to the suitability coefficient includes:

[0032] The suitability coefficient is compared with the preset suitability coefficient threshold. If the suitability coefficient is less than the preset suitability coefficient threshold, the user's current fitness plan is changed. If the suitability coefficient is not less than the preset suitability coefficient threshold, there is no need to change the user's current fitness plan.

[0033] Optionally, determining an action feature set corresponding to each training target based on the received user training data, and sending the action feature set corresponding to each training target to the user for selection, and generating a final fitness plan based on the user selection result includes:

[0034] The user training data is a number of user training targets;

[0035] According to each training target, the action corresponding to the training target is matched from the database, and the action feature corresponding to the training target is used as the action feature set, and each action in the action feature set is sorted according to a preset weight value to obtain the action feature corresponding to the training target as the action feature set;

[0036] The user selects actions based on the action feature set corresponding to each training goal and obtains the final fitness plan.

[0037] In addition, the present invention also provides an intelligent fitness recommendation system based on big data analysis, the system comprising

[0038] Data extraction module: obtains the user's training progress data, training load data and physical feedback data, and extracts the training progress target index, training load difference index and physical feedback index from the training progress data, training load data and physical feedback data;

[0039] Applicability module: The applicability coefficient of the current fitness program is obtained by using the training progress index, training load difference index and body feedback index through the fuzzy inference system, and whether to change the user's current fitness program is determined according to the applicability coefficient;

[0040] Data receiving module: when the user's current fitness plan needs to be changed, the module sends the user a training goal questionnaire and current training index data, and receives the user's training goal;

[0041] Fitness plan recommendation module: Determine the action feature set corresponding to each training goal based on the received user training data, and send the action feature set corresponding to each training goal to the user for selection, and generate the final fitness plan based on the user's selection results.

[0042] Beneficial effects of the present invention:

[0043] The present invention proposes an intelligent fitness recommendation method and system based on big data analysis, which obtains the user's training progress target index, training load difference index and physical feedback index, and obtains the applicability coefficient of the current fitness plan through a fuzzy reasoning system. It also determines whether to change the user's current fitness plan according to the applicability coefficient. When the user's current fitness plan needs to be changed, a training goal questionnaire and current training index data are sent to the user, and the user's training goal is received. According to the received user training data, an action feature set corresponding to each training goal is determined, and the action feature set corresponding to each training goal is sent to the user for selection. The final fitness plan is generated according to the user's selection result. In this way, the applicability of the user's current fitness plan can be judged in time, and the user's fitness plan can be recommended or changed in time, so that the user's fitness effect is more ideal and the impact on the user's physical health is reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The present invention will be further described below in conjunction with the accompanying drawings.

[0045] Figure 1 A flowchart of an intelligent fitness recommendation method based on big data analysis provided by an embodiment of the present invention;

[0046] Figure 2 A framework diagram of an intelligent fitness recommendation system based on big data analysis provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making any creative work shall fall within the scope of protection of the present invention.

[0049] The embodiment of the present invention provides an intelligent fitness recommendation method based on big data analysis. Figure 1 , Figure 1 A flowchart of an intelligent fitness recommendation method based on big data analysis provided by an embodiment of the present invention. The method acts on a target fitness platform, wherein:

[0050] Obtaining the user's training progress data, training load data, and physical feedback data, and extracting a training progress target index, a training load difference index, and a physical feedback index from the training progress data, the training load data, and the physical feedback data;

[0051] The training progress target index, the training load difference index and the body feedback index are used in a fuzzy inference system to obtain the applicability coefficient of the current fitness program, and whether to change the user's current fitness program is determined according to the applicability coefficient;

[0052] When the user's current fitness program needs to be changed, a training goal questionnaire and current training index data are sent to the user, and the user's training goal is received;

[0053] According to the received user training data, the action feature set corresponding to each training target is determined, and the action feature set corresponding to each training target is sent to the user for selection, and the final fitness plan is generated according to the user's selection result.

[0054] An intelligent fitness recommendation method based on big data analysis provided by an embodiment of the present invention can timely judge the applicability of the user's current fitness plan through the above method, and can timely recommend or change the user's fitness plan, so that the user's fitness effect is more ideal and the impact on the user's physical health is reduced.

[0055] In one embodiment, the method for obtaining the training progress target index is:

[0056] Get the actual value of each specific training goal of the user and the initial value corresponding to each specific training goal from the beginning of the fitness plan to the current time period, and mark the actual value of each specific training goal and the initial value corresponding to each specific training goal as and , The number indicating the number of specific training targets, =1, 2, 3, 4, ..., , is a positive integer, Indicates the total number of specific training targets;

[0057] The progress ratio of each specific training target is calculated based on the actual value of each specific training target and the initial value corresponding to each specific training target. The calculation formula is: , For the The progress ratio of a specific training goal;

[0058] Compare the progress ratio of each specific training target with the expected progress ratio range, and re-mark the specific training targets that are not within the expected progress ratio range as , The number of specific training targets that are not within the expected progress ratio range. =1, 2, 3, 4, ..., , is a positive integer, Indicates the total number of specific training targets that are not within the expected progression ratio range;

[0059] Calculate the training progress index. The specific calculation expression is: , where It is the training progress index.

[0060] It should be noted that specific training goals are some training goals selected and determined by the user himself, such as strength, endurance, speed, etc. The specific specific training goals are determined by the user himself and are not limited.

[0061] It should be noted that the user's training progress data, training load data and body feedback data can be obtained directly through the fitness platform or in other ways, without any specific limitation.

[0062] It should be noted that the expected progress ratio range of each specific training goal is set by the user himself or generated by a professional fitness platform, or it can be other ways, which are not limited or elaborated in detail.

[0063] It should be noted that the training progress index refers to the degree to which the progress of each specific training target compared to the initial corresponding initial time value after the user has executed the current fitness plan for a period of time is not in line with expectations; the greater the degree of non-conformity with expectations, the more it is necessary to change the current fitness plan. The reason is that when the training progress index shows that the user's training progress is significantly different from expectations, this may mean that the user's body has adapted to the current training intensity and movement combination, resulting in stagnation of progress in muscle growth, strength improvement or endurance enhancement. This stagnation phenomenon is often called a "training plateau". At this time, if the training plan is not adjusted, the user may stay at the same level for a long time and cannot improve further, resulting in unsatisfactory fitness results for the user and even affecting the user's physical health.

[0064] In one embodiment, the method for obtaining the training load difference index is:

[0065] Get the load percentage of each load indicator trained by the user from the beginning of the fitness program to the current time period. The calculation formula is: , Indicates The current load of the load indicator, Indicates The base load of a load index, For the The load percentage of each load indicator; =1, 2, 3, 4, ..., , is a positive integer, Indicates the total number of load indicators;

[0066] Calculate the training load index using the formula: , where is the training load index;

[0067] Calculate the training load difference index using the formula: , where is the training load difference index, is the preset training load threshold.

[0068] It should be noted that the load indicators are some load indicators selected and determined by the users themselves, such as training time, training frequency, training intensity, training distance, etc. The specific load indicators are determined by the users themselves and are not limited; in addition, the preset training load threshold is set by professionals based on actual conditions and is not limited or elaborated on.

[0069] It should be noted that baseline load refers to the initial training intensity or total amount of the user before starting a new training plan or fitness program. It usually reflects the user's training level, physical condition or daily activity load. In fitness and exercise planning, baseline load is an important reference point used to measure the user's progress and adaptation in training; the specific baseline load can be obtained through the user's fitness records without limitation.

[0070] It should be noted that the training load difference index refers to the degree of difference between the user's actual training load and the preset training load threshold; if the difference between the user's actual training load and the preset training load threshold is greater, it means that the current fitness plan needs to be changed more, because the training load difference index directly measures the degree of deviation between the user's actual training load and the preset ideal load, and this difference reflects the fit between the current fitness plan and the user's goals and physical needs. If the difference is large, it means that the current fitness plan may have lost its effectiveness for the user. Too high a training load may cause excessive fatigue of the body, increase the risk of injury, and may even cause the training effect to stagnate or regress, hindering the realization of long-term goals. Conversely, too low a training load means that the user's body may not get enough stimulation, and it is impossible to effectively improve physical fitness or achieve fitness goals, which will lead to inefficient training and slow progress. Therefore, the greater the training load difference index, the more it is necessary to adjust the current fitness plan in a timely manner to ensure that the training intensity and total amount can reasonably match the user's ability level and target needs, help users continue to make progress within a safe and effective range, and avoid wasting time or increasing health risks due to improper training plans.

[0071] In one embodiment, the method for obtaining the body feedback index is:

[0072] Get the score of each physical feeling indicator after each training session from the start of the fitness program to the current time period, and get the feedback index of each physical feeling indicator based on the score of each physical feeling indicator and the corresponding minimum and maximum scores. The calculation formula is: , where Indicates The scores of the physical feeling indicators, Indicates The preset minimum scores for each physical feeling indicator, Indicates The preset maximum score of each physical feeling indicator. For the Feedback index of physical feeling indicators; =1, 2, 3, 4, ..., , is a positive integer, a represents the total number of body feeling indicators;

[0073] Calculate the feedback index of the body feeling index after each training session , the calculation formula is: ; In the formula, Indicates The feedback index of the body feeling index after the training is completed. =1, 2, 3, 4, ..., , is a positive integer, Indicates the total number of trainings;

[0074] Calculate the body feedback index using the formula: ; In the formula, It is the body feedback index.

[0075] It should be noted that the preset minimum score of the physical feeling index and the preset maximum score of the physical feeling index are set by professionals based on actual conditions, and no specific restrictions or elaborations are made.

[0076] It should be noted that the body feedback index refers to the degree of rating of the user's physical feelings after training; the lower the rating of the user's physical feelings after training, the more it means that the current fitness plan needs to be changed, because if the score is low, it means that the user may feel overly tired, painful, or dissatisfied with the current training results. This indicates that the current fitness plan may not be suitable for the user's physical fitness level or does not meet their training goals, which may lead to overtraining, physical discomfort, increased risk of injury, and even make the user lose the motivation to continue training. Therefore, when the body feedback index is low, it means that the current fitness plan may need to be adjusted to improve the user's physical condition, avoid negative impacts on health, and ensure that the training effect better meets the user's expectations and needs.

[0077] In one embodiment, the training progress target index, the training load difference index and the body feedback index are used in a fuzzy inference system to obtain the applicability coefficient of the current fitness program, and judging whether to change the current fitness program of the user according to the applicability coefficient includes:

[0078] The training progress index, training load difference index and body feedback index are used as input items of fuzzy rules, and the applicability coefficient of the current fitness program is used as output item;

[0079] Fuzzify the input items and convert the exact input values ​​into fuzzy sets;

[0080] Define fuzzy rules based on the fuzzy sets of input items to map input items to output items;

[0081] According to the fuzzy rules, the input items are inferred to determine the fuzzy value of the output items;

[0082] For each rule, calculate the membership of its predecessor and take the minimum value as the activation of the rule;

[0083] Synthesize the results of all rules and calculate the fuzzy set of output items;

[0084] Convert the fuzzy value of the output item into an exact value, and output the applicability coefficient according to the defuzzification result;

[0085] The suitability coefficient is compared with the preset suitability coefficient threshold. If the suitability coefficient is less than the preset suitability coefficient threshold, the user's current fitness plan is changed. If the suitability coefficient is not less than the preset suitability coefficient threshold, there is no need to change the user's current fitness plan.

[0086] In one implementation method, through the above method, the fuzzy reasoning system can comprehensively consider the three key factors of training progress achievement index, training load difference index and body feedback index, so that the calculated applicability coefficient of the current fitness plan is more accurate, which can fully reflect the user's training status and ensure that the comprehensive evaluation results are more accurate and comprehensive.

[0087] In one embodiment, according to the received user training data, the action feature set corresponding to each training target is determined, and the action feature set corresponding to each training target is sent to the user for selection. The final fitness plan is generated according to the user selection result, including:

[0088] The user training data is a number of user training targets;

[0089] According to each training target, the action corresponding to the training target is matched from the database, and the action feature corresponding to the training target is used as the action feature set, and each action in the action feature set is sorted according to a preset weight value to obtain the action feature corresponding to the training target as the action feature set;

[0090] The user selects actions based on the action feature set corresponding to each training goal and obtains the final fitness plan.

[0091] It should be noted that the preset weight values ​​are set by professionals according to actual situations and are not specifically limited or elaborated.

[0092] In one implementation, the above method can accurately match and sort the action feature set from the database according to the user's specific training goals to ensure that the recommended actions best meet the user's needs. The final fitness plan is generated based on the user's selection results, making the training plan more personalized and user-oriented, which can effectively improve the pertinence and effectiveness of the training. In addition, the preset weight values ​​are used to ensure that the sorting of action features meets actual professional standards, thereby optimizing the scientificity and practicality of the fitness plan. This method not only improves user satisfaction, but also enhances the adaptability and effectiveness of the training plan.

[0093] Based on the same inventive concept, the embodiment of the present invention also provides an intelligent fitness recommendation system based on big data analysis. Figure 2 , Figure 2 A framework diagram of an intelligent fitness recommendation system based on big data analysis provided by an embodiment of the present invention, the system includes:

[0094] Data extraction module: obtains the user's training progress data, training load data and physical feedback data, and extracts the training progress target index, training load difference index and physical feedback index from the training progress data, training load data and physical feedback data;

[0095] Applicability module: The applicability coefficient of the current fitness program is obtained by using the training progress index, training load difference index and body feedback index through the fuzzy inference system, and whether to change the user's current fitness program is determined according to the applicability coefficient;

[0096] Data receiving module: when the user's current fitness plan needs to be changed, the module sends the user a training goal questionnaire and current training index data, and receives the user's training goal;

[0097] Fitness plan recommendation module: Determine the action feature set corresponding to each training goal based on the received user training data, and send the action feature set corresponding to each training goal to the user for selection, and generate the final fitness plan based on the user's selection results.

[0098] An intelligent fitness recommendation system based on big data analysis provided by an embodiment of the present invention can timely judge the applicability of the user's current fitness plan through the above method, and can timely recommend or change the user's fitness plan, so that the user's fitness effect is more ideal and the impact on the user's physical health is reduced.

[0099] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. An intelligent fitness recommendation method based on big data analysis, characterized in that: The following steps are involved: Obtaining the user's training progress data, training load data, and physical feedback data, and extracting a training progress target index, a training load difference index, and a physical feedback index from the training progress data, the training load data, and the physical feedback data; The training progress target index, the training load difference index and the body feedback index are used in a fuzzy inference system to obtain the applicability coefficient of the current fitness program, and whether to change the user's current fitness program is determined according to the applicability coefficient; When the user's current fitness program needs to be changed, a training goal questionnaire and current training index data are sent to the user, and the user's training goal is received; Determine the action feature set corresponding to each training target based on the received user training data, send the action feature set corresponding to each training target to the user for selection, and generate a final fitness plan based on the user selection result; The method for obtaining the training progress target index is: Get the actual value of each specific training goal of the user and the initial value corresponding to each specific training goal from the beginning of the fitness plan to the current time period, and mark the actual value of each specific training goal and the initial value corresponding to each specific training goal as and , The number indicating the number of specific training targets, =1, 2, 3, 4, ..., , is a positive integer, Indicates the total number of specific training targets; The progress ratio of each specific training target is calculated based on the actual value of each specific training target and the initial value corresponding to each specific training target. The calculation formula is: , For the The progress ratio of a specific training goal; Compare the progress ratio of each specific training target with the expected progress ratio range, and re-mark the specific training targets that are not within the expected progress ratio range as , The number of specific training targets that are not within the expected progress ratio range. =1, 2, 3, 4, ..., , is a positive integer, Indicates the total number of specific training targets that are not within the expected progression ratio range; Calculate the training progress index. The specific calculation expression is: , where It is the training progress achievement index; The method for obtaining the training load difference index is: Get the load percentage of each load indicator trained by the user from the beginning of the fitness program to the current time period. The calculation formula is: , Indicates The current load of the load indicator, Indicates The base load of a load index, For the The load percentage of each load indicator; =1, 2, 3, 4, ..., , is a positive integer, Indicates the total number of load indicators; Calculate the training load index using the formula: , where is the training load index; Calculate the training load difference index using the formula: , where is the training load difference index, To preset the training load threshold; The method for obtaining the body feedback index is: Get the score of each physical feeling indicator after each training session from the start of the fitness program to the current time period, and get the feedback index of each physical feeling indicator based on the score of each physical feeling indicator and the corresponding minimum and maximum scores. The calculation formula is: , where Indicates The scores of the physical feeling indicators, Indicates The preset minimum scores for each physical feeling indicator, Indicates The preset maximum score of each physical feeling indicator. For the Feedback index of physical feeling indicators; =1, 2, 3, 4, ..., , is a positive integer, Indicates the total number of body feeling indicators; Calculate the feedback index of the body feeling index after each training session , the calculation formula is: ; In the formula, Indicates The feedback index of the body feeling index after the training is completed. =1, 2, 3, 4, ..., , is a positive integer, Indicates the total number of trainings; Calculate the body feedback index using the formula: ; In the formula, is the body feedback index; The applicability coefficient of the current fitness program is obtained by using the fuzzy inference system to obtain the training progress index, training load difference index and body feedback index, including: The training progress index, training load difference index and body feedback index are used as input items of fuzzy rules, and the applicability coefficient of the current fitness program is used as output item; Fuzzify the input items and convert the exact input values ​​into fuzzy sets; Define fuzzy rules based on the fuzzy sets of input items to map input items to output items; According to the fuzzy rules, the input items are inferred to determine the fuzzy value of the output items; For each rule, calculate the membership of its predecessor and take the minimum value as the activation of the rule; Synthesize the results of all rules and calculate the fuzzy set of output items; Convert the fuzzy value of the output item into an exact value, and output the applicability coefficient according to the defuzzification result; Judging whether to change the user's current fitness plan based on the suitability coefficient includes: The suitability coefficient is compared with the preset suitability coefficient threshold. If the suitability coefficient is less than the preset suitability coefficient threshold, the user's current fitness plan is changed. If the suitability coefficient is not less than the preset suitability coefficient threshold, there is no need to change the user's current fitness plan.

2. The intelligent fitness recommendation method based on big data analysis according to claim 1, characterized in that: According to the received user training data, the action feature set corresponding to each training target is determined, and the action feature set corresponding to each training target is sent to the user for selection. The final fitness plan is generated according to the user selection result, including: The user training data is a number of user training targets; According to each training target, the action corresponding to the training target is matched from the database, and the action feature corresponding to the training target is used as the action feature set, and each action in the action feature set is sorted according to a preset weight value to obtain the action feature corresponding to the training target as the action feature set; The user selects actions based on the action feature set corresponding to each training goal and obtains the final fitness plan.

3. An intelligent fitness recommendation system based on big data analysis, used to implement the intelligent fitness recommendation method based on big data analysis described in any one of claims 1-2, characterized in that: The system comprises: Data extraction module: obtains the user's training progress data, training load data and physical feedback data, and extracts the training progress target index, training load difference index and physical feedback index from the training progress data, training load data and physical feedback data; Applicability module: The applicability coefficient of the current fitness program is obtained by using the training progress index, training load difference index and body feedback index through the fuzzy inference system, and whether to change the user's current fitness program is determined according to the applicability coefficient; Data receiving module: when the user's current fitness plan needs to be changed, the training goal questionnaire and current training index data are sent to the user, and the user's training goal is received; Fitness program recommendation module: Determine the action feature set corresponding to each training target based on the received user training data, send the action feature set corresponding to each training target to the user for selection, and generate the final fitness program based on the user's selection result; The method for obtaining the training progress target index is: Get the actual value of each specific training goal of the user and the initial value corresponding to each specific training goal from the beginning of the fitness plan to the current time period, and mark the actual value of each specific training goal and the initial value corresponding to each specific training goal as and , The number indicating the number of specific training targets, =1, 2, 3, 4, ..., , is a positive integer, Indicates the total number of specific training targets; The progress ratio of each specific training target is calculated based on the actual value of each specific training target and the initial value corresponding to each specific training target. The calculation formula is: , For the The progress ratio of a specific training goal; Compare the progress ratio of each specific training target with the expected progress ratio range, and re-mark the specific training targets that are not within the expected progress ratio range as , The number of specific training targets that are not within the expected progress ratio range. =1, 2, 3, 4, ..., , is a positive integer, Indicates the total number of specific training targets that are not within the expected progression ratio range; Calculate the training progress index. The specific calculation expression is: , where It is the training progress achievement index; The method for obtaining the training load difference index is: Get the load percentage of each load indicator trained by the user from the beginning of the fitness program to the current time period. The calculation formula is: , Indicates The current load of the load indicator, Indicates The base load of a load index, For the The load percentage of each load indicator; =1, 2, 3, 4, ..., , is a positive integer, Indicates the total number of load indicators; Calculate the training load index using the formula: , where is the training load index; Calculate the training load difference index using the formula: , where is the training load difference index, To preset the training load threshold; The method for obtaining the body feedback index is: Get the score of each physical feeling indicator after each training session from the start of the fitness program to the current time period, and get the feedback index of each physical feeling indicator based on the score of each physical feeling indicator and the corresponding minimum and maximum scores. The calculation formula is: , where Indicates The scores of the physical feeling indicators, Indicates The preset minimum scores for each physical feeling indicator, Indicates The preset maximum score of each physical feeling indicator. For the Feedback index of physical feeling indicators; =1, 2, 3, 4, ..., , is a positive integer, Indicates the total number of body feeling indicators; Calculate the feedback index of the body feeling index after each training session , the calculation formula is: ; In the formula, Indicates The feedback index of the body feeling index after the training is completed. =1, 2, 3, 4, ..., , is a positive integer, Indicates the total number of trainings; Calculate the body feedback index using the formula: ; In the formula, is the body feedback index; The applicability coefficient of the current fitness program is obtained by using the fuzzy inference system to obtain the training progress index, training load difference index and body feedback index, including: The training progress index, training load difference index and body feedback index are used as input items of fuzzy rules, and the applicability coefficient of the current fitness program is used as output item; Fuzzify the input items and convert the exact input values ​​into fuzzy sets; Define fuzzy rules based on the fuzzy sets of input items to map input items to output items; According to the fuzzy rules, the input items are inferred to determine the fuzzy value of the output items; For each rule, calculate the membership of its predecessor and take the minimum value as the activation of the rule; Synthesize the results of all rules and calculate the fuzzy set of output items; Convert the fuzzy value of the output item into an exact value, and output the applicability coefficient according to the defuzzification result; Judging whether to change the user's current fitness plan based on the suitability coefficient includes: The suitability coefficient is compared with the preset suitability coefficient threshold. If the suitability coefficient is less than the preset suitability coefficient threshold, the user's current fitness plan is changed. If the suitability coefficient is not less than the preset suitability coefficient threshold, there is no need to change the user's current fitness plan.

Images