Method for evaluating the application effect of fatty acid balanced fat powder based on big data analysis

Abstract

The invention discloses a method for evaluating the application effect of fatty acid balanced fat powder based on big data analysis, which relates to the technical field of big data analysis. A clustering algorithm is used to divide test persons into K population clusters. For each population cluster, an intake range is set for each intake parameter, and an application effect prediction model is trained to obtain the cluster to which the person to be evaluated belongs as a target cluster. Based on the application effect prediction model of the target cluster, the intake range of each intake parameter and physiological data, an application effect optimization problem is constructed; by solving the application effect optimization problem, recommended fat powder intake parameters are output; and the problems of predictive evaluation and application effect improvement of fatty acid balanced fat powder for different populations are solved in a quantitative manner, thereby providing a reference for nutritionists to provide intake parameter suggestions to users.

Description

Technical Field

[0001] The present invention relates to the technical field of big data analysis, and in particular to a method for evaluating the application effect of fatty acid balanced fat powder based on big data analysis. Background Art

[0002] Fatty acid balance usually refers to the composition ratio of fatty acids reaching a more appropriate level, avoiding too high or too low proportion of a certain type of fatty acid.

[0003] Common major fatty acid types include: saturated fatty acids (SFA), monounsaturated fatty acids (MUFA), polyunsaturated fatty acids (PUFA);

[0004] Among them, MUFA and PUFA are generally considered to be more beneficial to health, while long-term excessive intake of SFA may increase the risk of some diseases.

[0005] The function of fatty acid balanced fat powder is to provide the human body with balanced and appropriate levels of different types of fatty acids.

[0006] At present, the fat powder intake recommendations provided to the target groups are often rough and rely on subjective experience. Generally, the recommendations are provided based on intuitive factors such as the age and gender of the target groups. However, the physiological parameters of different target groups vary greatly. Therefore, in fact, the application effects of fat powder on different target groups also vary greatly, and it is difficult to make an objective evaluation of the application effects. Therefore, a method is needed to evaluate the application effects and generate a method for providing recommended intake parameters that can provide good application effects for different groups of people, so as to provide a reference for nutritionists to provide intake parameter recommendations to users.

[0007] To this end, the present invention proposes a method for evaluating the application effect of fatty acid balanced fat powder based on big data analysis. Summary of the invention

[0008] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes a method for evaluating the application effect of fatty acid balanced fat powder based on big data analysis, which solves the problem of predicting and evaluating the application effect of fatty acid balanced fat powder for different populations and improving the application effect in a quantitative manner.

[0009] To achieve the above purpose, a method for evaluating the application effect of fatty acid balanced fat powder based on big data analysis is proposed, comprising the following steps:

[0010] Step 1: Collect historical application population physiological sample data, historical intake parameter sample data, and historical application effect sample data of N testers in advance; N is the number of selected testers;

[0011] Step 2: Based on the physiological sample data of the historical application population, use the clustering algorithm to divide the test personnel into K population clusters; K is the number of preset population clusters;

[0012] Step 3: For each population cluster, based on historical intake parameter sample data and historical application effect sample data, set an intake range for each intake parameter;

[0013] Step 4: For each population cluster, use the historical application population physiological sample data and historical intake parameter sample data as input, and the historical application effect sample data as output to train the application effect prediction model;

[0014] Step 5: Collect the physiological data of the person to be evaluated, and based on the physiological data and the crowd clusters, obtain the cluster to which the person to be evaluated belongs as the target cluster;

[0015] Step 6: Based on the application effect prediction model of the target cluster, the intake range of each intake parameter and the physiological data, construct an application effect optimization problem; by solving the application effect optimization problem, output the recommended fat powder intake parameters;

[0016] The pre-collection of historical application population physiological sample data, historical intake parameter sample data and historical application effect sample data of N test persons includes the following steps:

[0017] Step 11: Recruit N volunteers as testers in advance;

[0018] Step 12: before the monitoring period starts, different fat powder intake parameters are set for each volunteer, and before the monitoring period starts, a set of physiological parameters of each volunteer is collected, and at the end of the monitoring period, application effect parameters of each volunteer are collected; wherein the duration of the monitoring period is a preset duration parameter;

[0019] The application effect parameter is the change in the effect evaluation index LRS before and after the monitoring period; the effect evaluation index LRS may be the thickness of subcutaneous fat;

[0020] Step 13: The physiological parameter set of each tester is formed into the historical application population physiological sample data, and the application effect parameter of each tester is formed into the historical application effect sample data;

[0021] The method of dividing the test persons into K population clusters comprises the following steps:

[0022] Step 21: for each test person, the set of physiological parameters of each monitoring period of the test person is formed into a set of physiological feature vectors;

[0023] Each physiological feature vector is regarded as an M-dimensional discrete point; wherein M is the dimension of the physiological feature vector, and the coordinate of the M-dimensional discrete point in each dimension corresponds to a physiological parameter in the physiological feature vector;

[0024] Step 22: Randomly select K M-dimensional discrete points as initial cluster centers, and the remaining M-dimensional discrete points as non-initial cluster centers;

[0025] Step 23: For each non-initial cluster center, calculate the Euclidean distance to each initial cluster center, and divide each M-dimensional discrete point into the cluster cluster where the nearest initial cluster center is located;

[0026] Step 24: Calculate the average value of the physiological parameter corresponding to each dimension of the M-dimensional discrete points in each cluster, form a new M-dimensional discrete point with the average value of the physiological parameter corresponding to each dimension, and use the new M-dimensional discrete point as the new cluster center of the cluster;

[0027] Step 25: Recalculate the Euclidean distance from each M-dimensional discrete point to each new cluster center, and re-divide each M-dimensional discrete point into the cluster where the nearest new cluster center is located;

[0028] Step 26: Repeat steps 24-25 until the M-dimensional discrete points in all the divided clusters no longer change, and mark the number of each cluster as k, k = 1, 2, 3, ... K; obtain all the M-dimensional discrete points contained in each cluster, and form a discrete point set from all the M-dimensional discrete points; each cluster corresponds to a crowd cluster, and the k-th crowd cluster includes the M-dimensional space coordinates of the new cluster center of the k-th cluster, and the test person corresponding to each M-dimensional discrete point in the k-th cluster and the physiological feature vector of the test person;

[0029] For each population cluster, based on the historical intake parameter sample data and the historical application effect sample data, the intake range is set for each intake parameter as follows:

[0030] Evaluate the application effect parameters of all testers in the kth population cluster, screen out the testers whose effect evaluation index LRS change does not exceed the range, form a normal tester set, and mark the normal tester set as Ek;

[0031] Collect the parameter values ​​of various intake parameters of each tester in the normal tester set Ek;

[0032] The number of each intake parameter is marked as i. For the i-th intake parameter, the maximum and minimum values ​​of the intake parameter of each tester in the normal tester set Ek are counted, and the maximum value of the i-th intake parameter is marked as ki_max, and the minimum value of the i-th intake parameter is marked as ki_min;

[0033] The training application effect prediction model is performed in the following manner:

[0034] For the kth population cluster:

[0035] The physiological feature vector of each test person and the parameter values ​​of each intake parameter corresponding to the k-th population cluster are combined into a set of application effect feature vectors;

[0036] Each group of application effect feature vectors of the kth population cluster is used as the input of the application effect prediction model, and the application effect prediction model takes the predicted value of the application effect parameter corresponding to each tester as the output; the application effect prediction model takes the parameter value of the application effect parameter of each tester as the prediction target, takes the difference between the predicted value of the application effect parameter and the parameter value of the application effect parameter as the prediction error, and takes minimizing the mean sum of squares of the prediction errors of each tester as the training target; the application effect prediction model is trained until the mean sum of squares of the prediction errors reaches convergence, and the training is stopped; the application effect prediction model is a polynomial regression model;

[0037] The method of collecting the physiological data of the person to be evaluated and obtaining the cluster to which the person to be evaluated belongs as the target cluster based on the physiological data and the crowd cluster is as follows:

[0038] Collecting parameter values ​​of various physiological parameters of the person to be evaluated to form physiological data of the person to be evaluated;

[0039] The physiological data is taken as a set of physiological feature vectors to be evaluated, and the M-dimensional discrete points corresponding to the physiological feature vectors to be evaluated are taken as target points. The Euclidean distance between the target points and the center point of each crowd cluster is calculated, and the crowd cluster with the smallest Euclidean distance is taken as the target cluster.

[0040] The method of constructing the application effect optimization problem is:

[0041] The number of the crowd cluster corresponding to the target cluster is marked as k′;

[0042] The polynomial function corresponding to the trained application effect prediction model is marked as Fk'(H, I), where H is a physiological variable set of physiological parameters, and I is an intake variable set of intake parameters;

[0043] The physiological data of the person to be evaluated is marked as h, and the physiological data h is substituted into the set of physiological variables in the polynomial function Fk'(H, I) to obtain the target optimization function Fk'(h, I);

[0044] Label the variable of the i-th input parameter as xi;

[0045] Set a constraint condition Ui for the i-th intake parameter, where the constraint condition Ui is k′i_min≤xi≤k′i_max; where k′i_min and k′i_max are the minimum and maximum values ​​of the i-th intake parameter in the set of normal testers corresponding to the target cluster, respectively;

[0046] The application effect optimization problem takes the target optimization function as the optimization target and the constraints of all the input parameters as the constraint set;

[0047] The method of outputting the recommended fat powder intake parameters by solving the application effect optimization problem is as follows:

[0048] By using the optimization problem solving algorithm, a variable value set of the intake parameter variables in the application effect optimization problem is obtained, and the parameter value of each intake parameter in the variable value set is the recommended fat powder intake parameter.

[0049] An electronic device is proposed, comprising: a processor and a memory, wherein the memory stores a computer program that can be called by the processor;

[0050] The processor executes the above-mentioned method for evaluating the application effect of fatty acid balanced fat powder based on big data analysis by calling the computer program stored in the memory.

[0051] A computer-readable storage medium is provided, on which a rewritable computer program is stored;

[0052] When the computer program is run on a computer device, the computer device is enabled to execute the above-mentioned method for evaluating the application effect of fatty acid balanced fat powder based on big data analysis.

[0053] Compared with the prior art, the present invention has the following beneficial effects:

[0054] The present invention collects historical application population physiological sample data, historical intake parameter sample data and historical application effect sample data of N test persons in advance, and divides the test persons into K population clusters based on the historical application population physiological sample data and the historical application effect sample data by using a clustering algorithm. For each population cluster, an intake range is set for each intake parameter based on the historical intake parameter sample data and the historical application effect sample data. For each population cluster, the historical application population physiological sample data and the historical intake parameter sample data are used as input, and the historical application effect sample data is used as output to train an application effect prediction model. The physiological data of the person to be evaluated is collected, and based on the physiological data and the person, an application effect prediction model is trained. Cluster the clusters, obtain the cluster to which the person to be evaluated belongs as the target cluster, and construct the application effect optimization problem based on the application effect prediction model of the target cluster, the intake range of each intake parameter and the physiological data; output the recommended fat powder intake parameters by solving the application effect optimization problem; first use the clustering algorithm to classify different audience groups, and then use the application effect prediction model to provide each group with the application effect prediction function of different intake parameters, so as to comprehensively solve the problem of predicting and evaluating the application effect of fatty acid balanced fat powder for different groups and improving the application effect in a quantitative way, so as to provide a reference for nutritionists to provide intake parameter recommendations to users. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 This is a flow chart of a method for evaluating the application effect of fatty acid balanced fat powder based on big data analysis in Example 1 of the present invention;

[0056] Figure 2 This is a schematic diagram of the structure of an electronic device in Embodiment 2 of the present invention;

[0057] Figure 3 This is a schematic diagram of the computer-readable storage medium structure in Example 3 of the present invention. DETAILED DESCRIPTION

[0058] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than 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.

[0059] Example 1

[0060] like Figure 1 As shown, the method for evaluating the application effect of fatty acid balanced fat powder based on big data analysis includes the following steps:

[0061] Step 1: Collect historical application population physiological sample data, historical intake parameter sample data, and historical application effect sample data of N testers in advance; N is the number of selected testers;

[0062] Step 2: Based on the physiological sample data of the historical application population, use the clustering algorithm to divide the test personnel into K population clusters; K is the number of preset population clusters;

[0063] Step 3: For each population cluster, based on historical intake parameter sample data and historical application effect sample data, set an intake range for each intake parameter;

[0064] Step 4: For each population cluster, use the historical application population physiological sample data and historical intake parameter sample data as input, and the historical application effect sample data as output to train the application effect prediction model;

[0065] Step 5: Collect the physiological data of the person to be evaluated, and based on the physiological data and the crowd clusters, obtain the cluster to which the person to be evaluated belongs as the target cluster;

[0066] Step 6: Based on the application effect prediction model of the target cluster, the intake range of each intake parameter and the physiological data, construct an application effect optimization problem; by solving the application effect optimization problem, output the recommended fat powder intake parameters;

[0067] The pre-collection of historical application population physiological sample data, historical intake parameter sample data and historical application effect sample data of N test persons includes the following steps:

[0068] Step 11: Recruit N volunteers as testers in advance;

[0069] Step 12: before the monitoring period starts, different fat powder intake parameters are set for each volunteer, and before the monitoring period starts, a set of physiological parameters of each volunteer is collected, and at the end of the monitoring period, application effect parameters of each volunteer are collected; wherein the duration of the monitoring period is a preset duration parameter;

[0070] The intake parameters may include the intake frequency and the dosage of fat powder ingested each time. The physiological parameters in the physiological parameter set include but are not limited to age, gender, height, weight, blood lipids, blood sugar, liver and kidney function, body fat rate, muscle mass, blood pressure, basal metabolic rate, etc.; the application effect parameter is the change in the effect evaluation index LRS before and after the monitoring period; the effect evaluation index LRS may be the thickness of subcutaneous fat. It is understood that the thickness of subcutaneous fat can be measured in a non-invasive manner, such as using a skin caliper or an ultrasonic measuring instrument. It can reflect the state of fat deposition in the human body and indirectly reflect the lipid level in the blood;

[0071] Step 13: The physiological parameter set of each tester is formed into the historical application population physiological sample data, and the application effect parameter of each tester is formed into the historical application effect sample data;

[0072] Furthermore, the method of dividing the test persons into K population clusters based on the historical application population physiological sample data by using a clustering algorithm comprises the following steps:

[0073] Step 21: for each test person, the set of physiological parameters of each monitoring period of the test person is formed into a set of physiological feature vectors;

[0074] Each physiological feature vector is regarded as an M-dimensional discrete point; wherein M is the dimension of the physiological feature vector, and the coordinates of the M-dimensional discrete point in each dimension correspond to a physiological parameter in the physiological feature vector; as an example, the coordinates of an M-dimensional discrete point may be (28, 1, 175, 60...), wherein 28, 1, 175, 60 respectively represent the age of 28 years old, the gender of male, the height of 175 cm, and the weight of 60 kg;

[0075] Step 22: Randomly select K M-dimensional discrete points as initial cluster centers, and the remaining M-dimensional discrete points as non-initial cluster centers;

[0076] Step 23: For each non-initial cluster center, calculate the Euclidean distance to each initial cluster center, and divide each M-dimensional discrete point into the cluster cluster where the nearest initial cluster center is located;

[0077] Step 24: Calculate the average value of the physiological parameter corresponding to each dimension of the M-dimensional discrete points in each cluster, form a new M-dimensional discrete point with the average value of the physiological parameter corresponding to each dimension, and use the new M-dimensional discrete point as the new cluster center of the cluster;

[0078] Step 25: Recalculate the Euclidean distance from each M-dimensional discrete point to each new cluster center, and re-divide each M-dimensional discrete point into the cluster where the nearest new cluster center is located;

[0079] Step 26: Repeat steps 24-25 until the M-dimensional discrete points in all the divided clusters no longer change, and mark the number of each cluster as k, k = 1, 2, 3, ... K; obtain all the M-dimensional discrete points contained in each cluster, and form a discrete point set from all the M-dimensional discrete points; each cluster corresponds to a crowd cluster, and the k-th crowd cluster includes the M-dimensional space coordinates of the new cluster center of the k-th cluster, and the test person corresponding to each M-dimensional discrete point in the k-th cluster and the physiological feature vector of the test person;

[0080] It can be understood that by clustering the test persons, people with similar physical fitness can be divided into a cluster, so that when generating recommended values ​​of intake parameters for the persons to be evaluated, the historical application effect data of people with similar physical fitness can be referred to, so as to improve the application effect for each person to be evaluated in a targeted manner;

[0081] Furthermore, for each population cluster, based on the historical intake parameter sample data and the historical application effect sample data, the intake range is set for each intake parameter in the following manner:

[0082] The application effect parameters of all testers in the kth population cluster are evaluated, and the testers whose effect evaluation index LRS change amount does not exceed the range are screened out to form a normal tester set, and the normal tester set is marked as Ek; it should be noted that the method of screening out the testers whose effect evaluation index LRS change amount does not exceed the range can be manually screened by professional medical personnel;

[0083] Collect the parameter values ​​of various intake parameters of each tester in the normal tester set Ek;

[0084] The number of each intake parameter is marked as i. For the i-th intake parameter, the maximum and minimum values ​​of the intake parameter of each tester in the normal tester set Ek are counted, and the maximum value of the i-th intake parameter is marked as ki_max, and the minimum value of the i-th intake parameter is marked as ki_min; it can be understood that ki_max and ki_min provide a set of guidance ranges for the intake parameters of the people in the k-th cluster when taking fat powder, ensuring that when the recommended values ​​of the intake parameters are provided to the person to be evaluated later, they do not exceed the guidance range, so as to reduce the occurrence rate of abnormal situations after the person to be evaluated takes fat powder;

[0085] Furthermore, for each population cluster, the historical application population physiological sample data and historical intake parameter sample data are used as input, and the historical application effect sample data is used as output, and the method of training the application effect prediction model is as follows:

[0086] For the kth population cluster:

[0087] The physiological feature vector of each test person and the parameter values ​​of each intake parameter corresponding to the k-th population cluster are combined into a set of application effect feature vectors;

[0088] Each group of application effect feature vectors of the kth population cluster is used as the input of the application effect prediction model, and the application effect prediction model takes the predicted value of the application effect parameter corresponding to each tester as the output; the application effect prediction model takes the parameter value of the application effect parameter of each tester as the prediction target, takes the difference between the predicted value of the application effect parameter and the parameter value of the application effect parameter as the prediction error, and takes minimizing the mean sum of squares of the prediction errors of each tester as the training target; the application effect prediction model is trained until the mean sum of squares of the prediction errors reaches convergence, and the training is stopped; the application effect prediction model is a polynomial regression model;

[0089] Furthermore, the method of collecting the physiological data of the person to be evaluated and obtaining the cluster to which the person to be evaluated belongs as the target cluster based on the physiological data and the crowd cluster is as follows:

[0090] Collecting parameter values ​​of various physiological parameters of the person to be evaluated to form physiological data of the person to be evaluated;

[0091] The physiological data is taken as a set of physiological feature vectors to be evaluated, and the M-dimensional discrete points corresponding to the physiological feature vectors to be evaluated are taken as target points. The Euclidean distance between the target points and the center point of each crowd cluster is calculated, and the crowd cluster with the smallest Euclidean distance is taken as the target cluster.

[0092] Furthermore, the application effect prediction model based on the target cluster, the intake range of each intake parameter and the physiological data, is constructed in the following manner to optimize the application effect:

[0093] The number of the crowd cluster corresponding to the target cluster is marked as k′;

[0094] The polynomial function corresponding to the trained application effect prediction model is marked as Fk'(H, I), where H is a physiological variable set of physiological parameters, and I is an intake variable set of intake parameters; it can be understood that by substituting the parameter values ​​of the specific physiological parameters and the parameter values ​​of the intake parameters of each person into the polynomial function Fk'(H, I), the predicted values ​​of the application effect parameters output by the application effect prediction model can be obtained;

[0095] The physiological data of the person to be evaluated is marked as h, and the physiological data h is substituted into the set of physiological variables in the polynomial function Fk'(H, I) to obtain the target optimization function Fk'(h, I);

[0096] Label the variable of the i-th input parameter as xi;

[0097] Set a constraint condition Ui for the i-th intake parameter, where the constraint condition Ui is k′i_min ≤xi≤k′i_max; where k′i_min and k′i_max are the minimum and maximum values ​​of the i-th intake parameter in the set of normal testers corresponding to the target cluster, respectively;

[0098] The application effect optimization problem takes the target optimization function as the optimization target and the constraints of all the input parameters as the constraint set;

[0099] Furthermore, the method of outputting the recommended fat powder intake parameters by solving the application effect optimization problem is:

[0100] By using the optimization problem solving algorithm, a variable value set of the intake parameter variables in the application effect optimization problem is obtained, and the parameter value of each intake parameter in the variable value set is the recommended fat powder intake parameter.

[0101] Example 2

[0102] Figure 2 1 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present application. Figure 2 As shown, according to another aspect of the present application, an electronic device 100 is also provided. The electronic device 100 may include one or more processors and one or more memories. The memories store computer readable codes, and when the computer readable codes are executed by one or more processors, the method for evaluating the application effect of fatty acid balanced fat powder based on big data analysis as described above can be executed.

[0103] The method or device according to the embodiment of the present application can also be used by Figure 2 The electronic device architecture shown in FIG. Figure 2 As shown, the electronic device 100 may include a bus 101, one or more CPUs 102, a ROM 103, a RAM 104, a communication port 105 connected to a network, an input / output component 106, a hard disk 107, etc. The storage device in the electronic device 100, such as the ROM 103 or the hard disk 107, may store the method for evaluating the application effect of the fatty acid balanced fat powder based on big data analysis provided in the present application.

[0104] Furthermore, the electronic device 100 may further include a user interface 108. Figure 2 The architecture shown is only exemplary and can be omitted according to actual needs when implementing different devices. Figure 2 One or more components of an electronic device are shown.

[0105] Example 3

[0106] Figure 3Schematic diagram of a computer-readable storage medium structure provided by an embodiment of the present application. Figure 3 As shown, it is a computer-readable storage medium 200 according to an embodiment of the present application. Computer-readable instructions are stored on the computer-readable storage medium 200. When the computer-readable instructions are executed by the processor, the method for evaluating the application effect of fatty acid balance fat powder based on big data analysis according to the embodiment of the present application described with reference to the above drawings can be executed. The computer-readable storage medium 200 includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may, for example, include random access memory (RAM) and cache memory (cache), etc. Non-volatile memory may, for example, include read-only memory (ROM), hard disk, flash memory, etc.

[0107] In addition, according to the implementation of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the present application provides a non-transitory machine-readable storage medium, the non-transitory machine-readable storage medium stores machine-readable instructions, the machine-readable instructions can be run by a processor to execute instructions corresponding to the method steps provided by the present application, and when the computer program is executed by a central processing unit (CPU), the above functions defined in the method of the present application are executed.

[0108] The methods, apparatuses, and devices of the present application may be implemented in many ways. For example, the methods, apparatuses, and devices of the present application may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present application are not limited to the order specifically described above unless otherwise specified. In addition, in some embodiments, the present application may also be implemented as programs recorded in a recording medium, which include machine-readable instructions for implementing the method according to the present application. Therefore, the present application also covers recording media storing programs for executing the method according to the present application.

[0109] In addition, the parts of the above-mentioned technical solutions provided in the embodiments of the present application that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive redundancy.

[0110] The specific implementation modes as described above further describe the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation mode of the present invention and is not intended to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0111] The above preset parameters or preset thresholds are all set by technicians in this field according to actual conditions or obtained through large-scale data simulation.

[0112] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A method for evaluating the application effect of fatty acid balanced fat powder based on big data analysis, characterized in that: The following steps are involved: Step 1: Collect historical application population physiological sample data, historical intake parameter sample data, and historical application effect sample data of N testers in advance; N is the number of selected testers; Step 2: Based on the physiological sample data of the historical application population, use the clustering algorithm to divide the test personnel into K population clusters; K is the number of preset population clusters; Step 3: For each population cluster, based on historical intake parameter sample data and historical application effect sample data, set an intake range for each intake parameter; Step 4: For each population cluster, use the historical application population physiological sample data and historical intake parameter sample data as input, and the historical application effect sample data as output to train the application effect prediction model; Step 5: Collect the physiological data of the person to be evaluated, and based on the physiological data and the crowd clusters, obtain the cluster to which the person to be evaluated belongs as the target cluster; Step 6: Based on the application effect prediction model of the target cluster, the intake range of each intake parameter and the physiological data, construct an application effect optimization problem; by solving the application effect optimization problem, output the recommended fat powder intake parameters; The pre-collection of historical application population physiological sample data, historical intake parameter sample data and historical application effect sample data of N test persons comprises the following steps: Step 11: Recruit N volunteers as testers in advance; Step 12: setting different fat powder intake parameters for each volunteer before the monitoring period starts, collecting a set of physiological parameters of each volunteer before the monitoring period starts, and collecting application effect parameters of each volunteer at the end of the monitoring period; wherein the duration of the monitoring period is a preset duration parameter; Step 13: The physiological parameter set of each tester is formed into the historical application population physiological sample data, and the application effect parameter of each tester is formed into the historical application effect sample data; The application effect parameter is the change in the effect evaluation index LRS before and after the monitoring period; The method of dividing the test persons into K population clusters comprises the following steps: Step 21: for each test person, the set of physiological parameters of each monitoring period of the test person is formed into a set of physiological feature vectors; Each physiological feature vector is regarded as an M-dimensional discrete point; wherein M is the dimension of the physiological feature vector, and the coordinate of the M-dimensional discrete point in each dimension corresponds to a physiological parameter in the physiological feature vector; Step 22: Randomly select K M-dimensional discrete points as initial cluster centers, and the remaining M-dimensional discrete points as non-initial cluster centers; Step 23: For each non-initial cluster center, calculate the Euclidean distance to each initial cluster center, and divide each M-dimensional discrete point into the cluster cluster where the nearest initial cluster center is located; Step 24: Calculate the average value of the physiological parameter corresponding to each dimension of the M-dimensional discrete points in each cluster, form a new M-dimensional discrete point with the average value of the physiological parameter corresponding to each dimension, and use the new M-dimensional discrete point as the new cluster center of the cluster; Step 25: Recalculate the Euclidean distance from each M-dimensional discrete point to each new cluster center, and re-divide each M-dimensional discrete point into the cluster where the nearest new cluster center is located; Step 26: Repeat steps 24-25 until the M-dimensional discrete points in all the divided clusters no longer change, and mark the number of each cluster as k, k = 1, 2, 3, ... K; obtain all the M-dimensional discrete points contained in each cluster, and form a discrete point set from all the M-dimensional discrete points; each cluster corresponds to a crowd cluster, and the k-th crowd cluster includes the M-dimensional space coordinates of the new cluster center of the k-th cluster, and the test person corresponding to each M-dimensional discrete point in the k-th cluster and the physiological feature vector of the test person; For each population cluster, based on the historical intake parameter sample data and the historical application effect sample data, the intake range is set for each intake parameter as follows: Evaluate the application effect parameters of all testers in the kth population cluster, screen out the testers whose effect evaluation index LRS change does not exceed the range, form a normal tester set, and mark the normal tester set as Ek; Collect the parameter values ​​of various intake parameters of each tester in the normal tester set Ek; The number of each intake parameter is marked as i. For the i-th intake parameter, the maximum and minimum values ​​of the intake parameter of each tester in the normal tester set Ek are counted, and the maximum value of the i-th intake parameter is marked as ki_max, and the minimum value of the i-th intake parameter is marked as ki_min; The training application effect prediction model is performed in the following manner: For the kth population cluster: The physiological feature vector of each test person and the parameter values ​​of each intake parameter corresponding to the k-th population cluster are combined into a set of application effect feature vectors; Each group of application effect feature vectors of the kth population cluster is used as the input of the application effect prediction model, and the application effect prediction model takes the predicted value of the application effect parameter corresponding to each tester as the output; the application effect prediction model takes the parameter value of the application effect parameter of each tester as the prediction target, takes the difference between the predicted value of the application effect parameter and the parameter value of the application effect parameter as the prediction error, and takes minimizing the mean sum of squares of the prediction errors of each tester as the training target; the application effect prediction model is trained until the mean sum of squares of the prediction errors reaches convergence, and the training is stopped; the application effect prediction model is a polynomial regression model; The method of collecting the physiological data of the person to be evaluated and obtaining the cluster to which the person to be evaluated belongs as the target cluster based on the physiological data and the crowd cluster is as follows: Collecting parameter values ​​of various physiological parameters of the person to be evaluated to form physiological data of the person to be evaluated; The physiological data is taken as a set of physiological feature vectors to be evaluated, and the M-dimensional discrete points corresponding to the physiological feature vectors to be evaluated are taken as target points. The Euclidean distance between the target points and the center point of each crowd cluster is calculated, and the crowd cluster with the smallest Euclidean distance is taken as the target cluster. It is characterized in that the method of constructing the application effect optimization problem is: The number of the crowd cluster corresponding to the target cluster is marked as k'; The polynomial function corresponding to the trained application effect prediction model is marked as Fk'(H,I), where H is a physiological variable set of physiological parameters, and I is an intake variable set of intake parameters; The physiological data of the person to be evaluated is marked as h, and the physiological data h is substituted into the set of physiological variables in the polynomial function Fk'(H,I) to obtain the target optimization function Fk'(h,I); Label the variable of the i-th input parameter as xi; Set a constraint condition Ui for the i-th intake parameter, where the constraint condition Ui is k'i_min≤xi≤k'i_max; where k'i_min and k'i_max are the minimum and maximum values ​​of the i-th intake parameter in the set of normal testers corresponding to the target cluster, respectively.

2. The method for evaluating the application effect of fatty acid balanced fat powder based on big data analysis according to claim 1, characterized in that: The application effect optimization problem takes the target optimization function as the optimization target and the constraints of all the input parameters as the constraint set; The method of outputting the recommended fat powder intake parameters by solving the application effect optimization problem is as follows: By using the optimization problem solving algorithm, a variable value set of the intake parameter variables in the application effect optimization problem is obtained, and the parameter value of each intake parameter in the variable value set is the recommended fat powder intake parameter.

3. An electronic device, characterized in that: include: A processor and a memory, wherein: The memory stores a computer program that can be called by the processor; The processor executes the method for evaluating the application effect of fatty acid balanced fat powder based on big data analysis as described in claim 1 in the background by calling the computer program stored in the memory.

4. A computer-readable storage medium, characterized in that: A rewritable computer program is stored thereon; When the computer program is executed on a computer device, the computer device executes the method for evaluating the application effect of fatty acid balanced fat powder based on big data analysis as claimed in claim 1 in the background.

Images