A fat loss training management method, device and electronic device based on big data

Abstract

The present application relates to the field of fat loss training management, and specifically to a fat loss training management method, device and electronic device based on big data. It uses artificial intelligence technology based on deep learning to perform data analysis on the body composition data and fat loss training program of the monitored object, and dig out the temporal dynamic change pattern of the body composition of the monitored object, as well as the deep semantic information of the fat loss training program, and then based on the semantic offset between the two, intelligently judge whether the fat loss training program needs to be adjusted. In this way, the blindness and inefficiency of traditional fat loss methods can be effectively avoided, and tailored fat loss training suggestions can be provided to the monitored object to ensure that the fat loss process is more scientific, accurate and efficient.

Description

Technical Field

[0001] The present application relates to the field of fat loss training management, and specifically to a fat loss training management method, device and electronic equipment based on big data. Background Art

[0002] In today's society, with the rapid changes in lifestyle and the diversification of dietary structure, obesity has become an increasingly serious health challenge worldwide.

[0003] In response to this challenge, fat reduction has become a health focus of the general public. However, traditional fat reduction methods often rely on individual self-management and simple dietary adjustments, lack scientificity and personalization, and are difficult to effectively respond to complex physiological changes and psychological needs.

[0004] Therefore, in order to provide more scientific, personalized and efficient fat loss training guidance, a fat loss training management method based on big data is expected. Summary of the invention

[0005] This application is made in consideration of the above problems. One purpose of this application is to provide a fat loss training management method, device and electronic device based on big data.

[0006] The embodiment of the present application provides a fat loss training management device based on big data, which includes:

[0007] A body fat scale data acquisition module, used to obtain a time series of the body composition data of the monitored subject, wherein the body composition data includes body weight, body fat percentage, muscle mass, bone mass, water content, basal metabolic rate, visceral fat level, body age, protein content, body mass index and lean body weight;

[0008] A fat-reducing training program setting module is used to receive a fat-reducing training program input by a user;

[0009] A big data module, used for performing feature encoding on the time series of the body composition data and the fat loss training program to obtain body composition time series associated aggregation features and fat loss training program semantic encoding features;

[0010] The training management result generation module is used to determine the training management result based on the body composition time series association aggregation feature and the semantic coding feature of the fat reduction training program, and the training management result is used to indicate whether the fat reduction training program needs to be adjusted.

[0011] For example, according to the fat loss training management device based on big data in an embodiment of the present application, the big data module includes:

[0012] A body composition data time series correlation analysis unit, used for performing time series correlation analysis on the time series of the body composition data to obtain a time series of body composition time series correlation implicit feature vectors;

[0013] A body component feature time series propagation aggregation unit is used to perform feature time series propagation aggregation on the time series of the body component time series associated implicit feature vector to obtain a body component time series propagation aggregation representation vector as the body component time series associated aggregation feature;

[0014] The fat reduction training program semantic coding unit is used to perform global semantic coding on the fat reduction training program to obtain a fat reduction training program semantic coding feature vector as the fat reduction training program semantic coding feature.

[0015] For example, according to the fat loss training management device based on big data in an embodiment of the present application, the body composition data time series correlation analysis unit includes:

[0016] an embedding coding subunit, configured to respectively embed code each body component data in the time series of the body component data using a body component embedding matrix to obtain a time series of body component fully connected embedding coding vectors;

[0017] The time series correlation coding subunit is used to perform time series correlation coding on the time series of the body component fully connected embedded coding vector to obtain the time series of the body component time series correlation implicit feature vector.

[0018] For example, according to the fat loss training management device based on big data in an embodiment of the present application, the time-series association coding subunit is used to:

[0019] The time series of the body component fully connected embedded coding vector is input into the body component temporal association encoder based on the Bi-LSTM model to obtain the time series of the body component temporal association implicit feature vector.

[0020] For example, according to the fat loss training management device based on big data in an embodiment of the present application, the body composition feature time series propagation aggregation unit is used to:

[0021] The time series of the implicit feature vector associated with the body component time series is input into a feature forward propagation network based on significant time series attenuation of node energy to obtain the body component time series propagation aggregation representation vector.

[0022] For example, according to the fat loss training management device based on big data in an embodiment of the present application, the body composition feature time series propagation aggregation unit includes:

[0023] An energy significant descriptor calculation subunit, configured to calculate the energy significant descriptors of the respective body component time series associated implicit feature vectors in the time series of the body component time series associated implicit feature vectors based on the mean and variance of the respective body component time series associated implicit feature vectors to obtain a time series of energy significant descriptors;

[0024] A timestamp marking subunit, used for extracting the timestamp of each body component time series associated implicit feature vector in the time series of the body component time series associated implicit feature vector to obtain a time series of timestamps;

[0025] A historical node feature division subunit is used to use the body component time series association implicit feature vector corresponding to the current time point in the time series of the body component time series association implicit feature vector as the current body component time series association implicit feature vector, and use the body component time series association implicit feature vectors corresponding to other time points as the historical body component time series association implicit feature vectors, so as to obtain a time series of the current body component time series association implicit feature vector and the historical body component time series association implicit feature vector;

[0026] The attenuation factor calculation subunit is used to calculate the energy time attenuation factor of each historical body component time series associated implicit feature vector in the time series of the historical body component time series associated implicit feature vector relative to the current body component time series associated implicit feature vector to obtain a time series of energy time attenuation factors;

[0027] A historical node propagation aggregation subunit is used to perform time series propagation aggregation on the time series of the historical body component time series associated implicit feature vectors using the time series of the energy significant descriptor as a positive adjustment factor and the time series of the energy time decay factor as a negative adjustment factor to obtain a historical node body component time series propagation aggregation representation feature vector;

[0028] The global time series aggregation subunit is used to fuse the historical node body component time series propagation aggregation representation feature vector and the current body component time series associated implicit feature vector to obtain the body component time series propagation aggregation representation vector.

[0029] For example, according to the fat loss training management device based on big data in an embodiment of the present application, the attenuation factor calculation subunit is used to:

[0030] Calculating a timestamp difference between the historical body composition time series associated implicit feature vector and the current body composition time series associated implicit feature vector;

[0031] The timestamp difference is rounded down and divided by a preset time decay period and then squared, and the square result is multiplied by a preset decay rate to obtain an initial energy decay coefficient;

[0032] The energy time decay factor is obtained by calculating an exponential function value with the natural constant as the base and the initial energy decay coefficient as the exponent.

[0033] For example, according to the fat loss training management device based on big data in an embodiment of the present application, the training management result generation module includes:

[0034] A position difference unit, used for calculating the position difference between the semantic encoding feature vector of the fat loss training program and the body composition temporal propagation aggregation representation vector to obtain a semantic offset difference representation vector;

[0035] The training management result generating unit is used to input the semantic shift differential representation vector into the classifier-based training management result generator to obtain the training management result.

[0036] The embodiment of the present application also provides a fat loss training management method based on big data, which includes:

[0037] Obtaining a time series of body composition data of the monitored subject, wherein the body composition data includes weight, body fat percentage, muscle mass, bone mass, water content, basal metabolic rate, visceral fat level, body age, protein content, body mass index and lean body mass;

[0038] receiving a fat loss training program input by a user;

[0039] Performing feature encoding on the time series of the body composition data and the fat loss training program to obtain body composition time series associated aggregation features and fat loss training program semantic encoding features;

[0040] Based on the body composition temporal association aggregation features and the semantic coding features of the fat loss training program, a training management result is determined, and the training management result is used to indicate whether the fat loss training program needs to be adjusted.

[0041] An embodiment of the present application further provides an electronic device, comprising:

[0042] A memory for storing instructions;

[0043] A processor is coupled to the memory, and the processor is configured to execute instructions stored in the memory to implement the fat loss training management device based on big data as described in any of the preceding items.

[0044] According to the fat-reduction training management method, device and electronic device based on big data of the embodiment of the present application, it uses artificial intelligence technology based on deep learning to analyze the body composition data and fat-reduction training program of the monitored object, dig out the time series dynamic change pattern of the body composition of the monitored object, and the deep semantic information of the fat-reduction training program, and then based on the semantic offset between the two, intelligently judge whether the fat-reduction training program needs to be adjusted. In this way, the blindness and inefficiency of traditional fat-reduction methods can be effectively avoided, and tailored fat-reduction training suggestions can be provided for the monitored object, ensuring that the fat-reduction process is more scientific, accurate and efficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application are briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application, and are not intended to limit the present application.

[0046] Figure 1 A schematic diagram of the structure of a fat loss training management device based on big data in an embodiment of the present application is shown;

[0047] Figure 2 A schematic diagram showing the structure of a big data module of a fat loss training management device based on big data in an embodiment of the present application is shown;

[0048] Figure 3 A flowchart showing a fat loss training management method based on big data in an embodiment of the present application is shown; and

[0049] Figure 4 An application scenario diagram of a fat loss training management device based on big data in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0050] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0051] The terms used in this specification are those common terms currently widely used in the art in consideration of the functions of the present application, but these terms may vary according to the intentions of those skilled in the art, precedents, or new technologies in the art. In addition, specific terms may be selected, and in this case, their detailed meanings will be described in the detailed description of the present application. Therefore, the terms used in the specification should not be understood as simple names, but rather as a general description based on the meaning of the terms and the present application.

[0052] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0053] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. At the same time, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0054] In addition, the application architecture diagram in the embodiment of the present application is intended to more clearly illustrate the technical solution in the embodiment of the present application, and does not constitute a limitation on the technical solution provided in the embodiment of the present application. Of course, for other application architectures and business applications, the technical solution provided in the embodiment of the present application is also applicable to similar problems.

[0055] The following is a non-restrictive description of the fat loss training management device based on big data provided according to at least one embodiment of the present application through several examples or embodiments. As described below, different features in these specific examples or embodiments can be combined with each other without conflicting with each other to obtain new examples or embodiments, and these new examples or embodiments also fall within the scope of protection of the present application.

[0056] In response to the above technical problems, the technical concept of this application is to use artificial intelligence technology based on deep learning to analyze the body composition data and fat-reducing training program of the monitored object, dig out the time-series dynamic change pattern of the body composition of the monitored object, and the deep semantic information of the fat-reducing training program, and then based on the semantic offset between the two, intelligently judge whether the fat-reducing training program needs to be adjusted. In this way, the blindness and inefficiency of traditional fat-reducing methods can be effectively avoided, and tailored fat-reducing training suggestions can be provided to the monitored object to ensure that the fat-reducing process is more scientific, accurate and efficient.

[0057] Based on this, this application proposes a fat loss training management device based on big data. Figure 1The structural schematic diagram of the fat-reducing training management device 800 based on big data in the embodiment of the present application is shown. The fat-reducing training management device 800 based on big data includes: a body fat scale data acquisition module 810, which is used to obtain the time series of body composition data of the monitored object, wherein the body composition data includes weight, body fat percentage, muscle mass, bone mass, water content, basal metabolic rate, visceral fat level, body age, protein content, body mass index and fat-free body weight; a fat-reducing training program setting module 820, which is used to receive the fat-reducing training program input by the user; a big data module 830, which is used to feature encode the time series of the body composition data and the fat-reducing training program to obtain the body composition time series associated aggregation feature and the fat-reducing training program semantic coding feature; a training management result generation module 840, which is used to determine the training management result based on the body composition time series associated aggregation feature and the fat-reducing training program semantic coding feature, and the training management result is used to indicate whether the fat-reducing training program needs to be adjusted.

[0058] Among them, Figure 2 As shown, the big data module 830 includes: a body composition data time series association analysis unit 831, which is used to perform time series association analysis on the time series of the body composition data to obtain a time series of body composition time series association implicit feature vectors; a body composition feature time series propagation aggregation unit 832, which is used to perform feature time series propagation aggregation on the time series of the body composition time series association implicit feature vectors to obtain a body composition time series propagation aggregation representation vector as the body composition time series association aggregation feature; a fat reduction training program semantic encoding unit 833, which is used to perform global semantic encoding on the fat reduction training program to obtain a fat reduction training program semantic encoding feature vector as the fat reduction training program semantic encoding feature.

[0059] Specifically, in the feature mining process of the time series of the body composition data and the fat loss training program, first of all, considering that the body composition data contains a variety of physiological indicators, each indicator reflects different aspects of the human body's physical state, and there is a certain correlation between different body composition parameters. Therefore, in order to fully understand the body composition status of the monitored object at each time point, the present application uses a body composition embedding matrix to embed and encode each body composition data in the time series of the body composition data, map the body composition data at each time point into a high-dimensional feature space, reveal the potential correlation between different body composition parameters, and more comprehensively understand the physical condition of the monitored object, so as to generate a time series of body composition fully connected embedded coding vectors.

[0060] Next, in order to capture the dynamic trend of body composition data changing over time, the present application introduces a body composition time series association encoder based on the Bi-LSTM model to perform time series association analysis on the time series of the body composition fully connected embedded coding vector, so as to utilize the bidirectional information transmission characteristics of the Bi-LSTM model to capture the forward and backward dependencies of the body composition data in time, accurately reveal the inherent laws and trends of changes in the human body state, and use the outputs of each hidden layer in the Bi-LSTM model as the time series of the body composition time series association implicit feature vector, so as to provide strong data support for the subsequent adjustment of the fat reduction training program.

[0061] Correspondingly, the body component data time series association analysis unit 831 includes: an embedding coding subunit, which uses a body component embedding matrix to perform embedding coding on each body component data in the time series of the body component data to obtain a time series of body component fully connected embedded coding vectors; and a time series association coding subunit, which performs time series association coding on the time series of the body component fully connected embedded coding vectors to obtain a time series of the body component time series association implicit feature vectors.

[0062] Among them, the temporal association encoding subunit is used to: input the time series of the body component fully connected embedded coding vector into the body component temporal association encoder based on the Bi-LSTM model to obtain the time series of the body component temporal association implicit feature vector.

[0063] Furthermore, in order to integrate the body composition information of the monitored object in the global time domain and obtain the global change characteristics of its body state in the time dimension, the present application further introduces a feature forward propagation network based on the significant time series decay of node energy to aggregate the time series of the implicit feature vector associated with the body composition time series. Specifically, the feature forward propagation network based on the significant time series decay of node energy first calculates its energy significant descriptor according to the mean and variance of each implicit feature vector associated with the body composition time series, and quantifies its feature importance. Subsequently, the time information is assigned to the implicit feature vector associated with the body composition time series of each time node, and the time series decay mechanism is used to generate the corresponding energy time decay factor based on the time span between the implicit feature vector associated with the body composition time series of each historical time node and the implicit feature vector associated with the current (i.e., the end of the time series) body composition time series, so that when the weighted feature is aggregated, the features can be aggregated based on the feature significance of each time node and the time series decay characteristics of the information to generate a body composition time series propagation aggregation representation vector. In this way, not only can comprehensive information on changes in body composition of the monitored subjects during the fat loss process be unearthed, but the importance of recent changes is also highlighted through the time-series decay mechanism, thereby enhancing the timeliness and accuracy of fat loss training management.

[0064] At the same time, in order to understand the deep meaning of the fat-reducing training program set by the user, the fat-reducing training program is further semantically encoded, and the pre-trained natural language processing model is used to deeply analyze the various exercise types, intensities, frequencies, durations and other information in the fat-reducing training program, and extract the contextual semantic feature representation of the fat-reducing training program to obtain the semantic encoding feature vector of the fat-reducing training program, which is used for the subsequent correlation analysis between the training program and the changes in body composition. In an embodiment of the present application, a semantic encoder based on a Transformer structure can be used to implement this semantic encoding process.

[0065] Correspondingly, the body component feature time series propagation aggregation unit 832 is used to: input the time series of the body component time series associated implicit feature vector into the feature forward propagation network based on the significant time series attenuation of node energy to obtain the body component time series propagation aggregation representation vector.

[0066] Specifically, the body component feature time series propagation aggregation unit 832 includes: an energy significant descriptor calculation subunit, which is used to calculate the energy significant descriptors of each body component time series associated implicit feature vector based on the mean and variance of each body component time series associated implicit feature vector in the time series of the body component time series associated implicit feature vector to obtain a time series of energy significant descriptors; a timestamp marking subunit, which is used to extract the timestamps of each body component time series associated implicit feature vector in the time series of the body component time series associated implicit feature vector to obtain a time series of timestamps; a historical node feature division subunit, which is used to use the body component time series associated implicit feature vector corresponding to the current time point in the time series of the body component time series associated implicit feature vector as the current body component time series associated implicit feature vector, and use the body component time series associated implicit feature vectors corresponding to other time points as the historical body component time series associated implicit feature vectors to obtain the current body component time series associated implicit feature vector. A time series of time-series-associated implicit feature vectors and historical body component time-series-associated implicit feature vectors; a decay factor calculation subunit, used to calculate the energy time decay factor of each historical body component time-series-associated implicit feature vector in the time series of the historical body component time-series-associated implicit feature vectors relative to the current body component time-series-associated implicit feature vector to obtain a time series of energy time decay factors; a historical node propagation aggregation subunit, used to perform time propagation aggregation on the time series of the historical body component time-series-associated implicit feature vectors using the time series of the energy-significant descriptors as a positive adjustment factor and the time series of the energy time decay factors as a negative adjustment factor to obtain a historical node body component time-series propagation aggregation representation feature vector; a global time-series aggregation subunit, used to fuse the historical node body component time-series propagation aggregation representation feature vector and the current body component time-series-associated implicit feature vector to obtain the body component time-series propagation aggregation representation vector.

[0067] Among them, the attenuation factor calculation subunit is used to: calculate the timestamp difference between the historical body composition time series associated implicit feature vector and the current body composition time series associated implicit feature vector; round down the timestamp difference and divide it by the preset time attenuation period and then square it, and multiply the square result by the preset attenuation rate to obtain the initial energy attenuation coefficient; calculate the exponential function value with the natural constant as the base and the initial energy attenuation coefficient as the exponent to obtain the energy time attenuation factor.

[0068] In a specific example, the body component feature time series propagation aggregation unit 832 is used to process the time series of the body component time series associated implicit feature vector using the following time series propagation fusion formula to obtain the body component time series propagation aggregation representation vector, wherein the time series propagation fusion formula is:

[0069]

[0070]

[0071]

[0072]

[0073]

[0074] in, represents the time series of the implicit feature vectors associated with the temporal association of the body components, The value of is the number of feature vectors in the time series of the implicit feature vector associated with the time series of the body composition, , , , and Respectively represent the first, second, and third , The implicit feature vector associated with the current body composition time series, Indicates the The temporal association latent feature vector of the body components eigenvalues, The value of The length of the implicit feature vector of the temporal association of body components, For the said The feature mean of the implicit feature vector of the temporal association of body components, For the said The characteristic variance of the implicit characteristic vector of the temporal association of body components, Indicates the The fourth-order central moment of the implicit eigenvector of the temporal association of body components, that is, The expected value of the fourth power of the difference between each eigenvalue and its mean in the implicit eigenvector of the temporal association of body components, For the said The energy-significant descriptor of the implicit feature vector of temporal association of body components, is the energy significant descriptor of the implicit feature vector associated with the current body component time series, The timestamp representing the implicit feature vector associated with the current body composition time series, Indicates the The timestamp of the implicit feature vector associated with the temporal series of body components, Indicates rounding down. Indicates the preset decay rate, Indicates the preset time decay period, represents the exponential function operation, is the weight hyperparameter, A representation vector representing the temporal propagation aggregation of the body components.

[0075] Furthermore, by calculating the semantic offset difference representation vector between the semantic encoding feature vector of the fat loss training program and the body composition temporal propagation aggregation representation vector, the semantic deviation between the fat loss training program and the body composition changes of the monitored object can be quantified, thereby revealing whether the execution effect of the training program is consistent with the physical change trend of the monitored object, and using this as a key basis for evaluating whether the training program needs to be adjusted.

[0076] Then, the semantic shift differential representation vector is input into the classifier-based training management result generator to obtain the training management result. In the technical solution of the present application, the classifier-based training management result generator adopts a fully connected neural network structure to classify and predict the adaptability of the fat reduction training program according to the characteristic distribution of the semantic shift differential representation vector to determine whether the current fat reduction training program needs to be adjusted. In this way, the fat reduction training program can be dynamically evaluated and optimized to ensure that it always maintains a high degree of fit with the physical condition of the monitored subject, thereby improving the scientificity and efficiency of the fat reduction process.

[0077] Correspondingly, the training management result generation module 840 includes: a position difference unit, used to calculate the position difference between the semantic encoding feature vector of the fat loss training program and the body composition temporal propagation aggregation representation vector to obtain a semantic offset difference representation vector; a training management result generation unit, used to input the semantic offset difference representation vector into a classifier-based training management result generator to obtain the training management result.

[0078] It should be understood that the role of the classifier is to use the given categories and known training data to learn classification rules and classifiers, and then classify (or predict) unknown data. Logistic regression, SVM, etc. are often used to solve binary classification problems. For multi-class classification problems, logistic regression or SVM can also be used, but multiple binary classifications are required to form a multi-classification, but this is prone to errors and inefficient. Commonly used multi-classification methods include the Softmax classification function.

[0079] In a preferred example, considering that the semantic encoding feature vector of the fat loss training program and the body composition temporal propagation aggregate representation vector respectively represent the textual semantic features of the fat loss training program and the body composition temporal associated implicit aggregate propagation features, after calculating the semantic shift differential representation vector between the semantic encoding feature vector of the fat loss training program and the body composition temporal propagation aggregate representation vector, the obtained semantic shift differential representation vector will also cause decoding semantic regression offset due to the feature distribution mode and timing differences between the semantic encoding feature vector of the fat loss training program and the body composition temporal propagation aggregate representation vector, thereby causing outlier class regression inference mapping deviation, and further affecting the accuracy of the training management result obtained by inputting the semantic shift differential representation vector into the classifier-based training management result generator.

[0080] Based on this, when the semantic shift differential representation vector is input into the training management result generator based on the classifier to obtain the training management result, the semantic shift differential representation vector is optimized. Accordingly, the training management result generating unit includes: an optimization subunit, which is used to optimize the semantic shift differential representation vector to obtain an optimized semantic shift differential representation vector; and a classification subunit, which is used to input the optimized semantic shift differential representation vector into the training management result generator based on the classifier to obtain the training management result.

[0081] Wherein, the optimization subunit is used to: calculate the feature mean of the semantic shift differential representation vector, and divide the feature mean by the difference between the maximum eigenvalue and the minimum eigenvalue of the semantic shift differential representation vector to obtain a distribution representation value of the semantic shift differential representation vector; subtract the distribution representation value of the semantic shift differential representation vector and divide it by the distribution representation value of the semantic shift differential representation vector to obtain a distribution modulation value of the semantic shift differential representation vector; activate the semantic shift differential representation vector through a probabilistic function to obtain a probabilistic semantic shift differential representation vector; perform point subtraction between the probabilistic semantic shift differential representation vector and the distribution modulation value of the semantic shift differential representation vector, take the absolute value and calculate The negative of the logarithmic value with base 2 is used to obtain a probabilistic semantic shift differential representation vector distribution modulation information vector; after dividing the semantic shift differential representation vector distribution representation value by one minus the difference of each eigenvalue of the probabilistic semantic shift differential representation vector, all eigenvalues ​​of the probabilistic semantic shift differential representation vector are summed and divided by the length of the semantic shift differential representation vector to obtain a probabilistic semantic shift differential representation vector distribution modulation bias value; and the product of the probabilistic semantic shift differential representation vector distribution modulation information vector and the probabilistic semantic shift differential representation vector distribution modulation bias value and the weight as a hyperparameter is dotted to obtain an optimized semantic shift differential representation vector.

[0082] In a specific example, the optimization subunit is used to optimize the semantic shift differential representation vector using the following optimization formula to obtain the optimized semantic shift differential representation vector; wherein the optimization formula is:

[0083]

[0084] And among them:

[0085]

[0086] in, represents the feature mean of the semantic shift differential representation vector, and They represent the maximum eigenvalue and the minimum eigenvalue in the semantic shift differential representation vector respectively, 𝑝 represents the distribution representation value of the semantic shift differential representation vector, represents the probabilistic semantic shift differential representation vector obtained after the semantic shift differential representation vector is activated by the probabilistic function, represents the first eigenvalues, represents the logarithmic function value with base 2, 𝜀 is the weight as a hyperparameter, The value of is the length of the probabilistic semantic offset differential representation vector, Represents the optimized semantic offset differential representation vector.

[0087] That is, in the above preferred example, the eigenvalue-based probability information distribution planning of the semantic shift differential representation vector is performed by using the Bernoulli probability modulation distribution of the semantic shift differential representation vector relative to the eigenvalue distribution, and the probability reverse mapping of the probability features of the semantic shift differential representation vector as a whole is used as an extended coverage of the set mapping space of the semantic shift differential representation vector to independently understand the interaction path between the intuitive probability information distribution and the abstract probability space mapping of the semantic shift differential representation vector, so as to improve the accuracy of the training management results obtained by inputting the optimized semantic shift differential representation vector into the classifier-based training management result generator by avoiding the counterfactual reasoning mapping of the outlier feature distribution of the semantic shift differential representation vector to the class regression probability.

[0088] Accordingly, the present application also provides an electronic device, comprising: a memory for storing instructions; a processor coupled to the memory, the processor being configured to execute the fat loss training management device based on big data as described in any of the preceding items based on the instructions stored in the memory.

[0089] Based on the above embodiments, see Figure 3 As shown in FIG. 1 , it is a flow chart of a fat loss training management method based on big data in an embodiment of the present application. Figure 3 As shown, according to the fat loss training management method based on big data according to the embodiment of the present application, the steps include: S510, obtaining the time series of body composition data of the monitored object, wherein the body composition data includes weight, body fat percentage, muscle mass, bone mass, water content, basal metabolic rate, visceral fat level, body age, protein content, body mass index and lean body weight; S520, receiving a fat loss training plan input by a user; S530, feature encoding the time series of the body composition data and the fat loss training plan to obtain body composition time series associated aggregation features and fat loss training plan semantic coding features; S540, determining the training management result based on the body composition time series associated aggregation features and the fat loss training plan semantic coding features, and the training management result is used to indicate whether the fat loss training plan needs to be adjusted.

[0090] Here, those skilled in the art can understand that the specific operations of each step in the above-mentioned fat loss training management method based on big data have been referred to above. Figure 1 to Figure 2 It has been introduced in detail in the description of the big data-based fat loss training management device 800, and therefore, its repeated description will be omitted.

[0091] Figure 4 FIG. 1 is an application scenario diagram of a fat loss training management device based on big data according to an embodiment of the present application. Figure 4 As shown, in this application scenario, first, the time series of the body composition data of the monitored object is obtained (for example, Figure 4 D1 shown in FIG. 1 ) and receiving a fat loss training program input by a user (eg, Figure 4 Then, the time series of the body composition data and the fat loss training program are input to a server (for example, Figure 4 S) as shown in , wherein the server is able to use the big data-based fat loss training management algorithm to process the time series of the body composition data and the fat loss training program to obtain a training management result indicating whether the fat loss training program needs to be adjusted.

[0092] Based on the above embodiment, another exemplary embodiment of an electronic device is also provided in the embodiment of the present application. In some possible implementations, the electronic device in the embodiment of the present application may include a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor can implement the steps of the fat loss training management method based on big data in the above embodiment when executing the program.

[0093] The embodiment of the present application also provides a computer-readable storage medium, on which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, the fat-reducing training management method based on big data according to the embodiment of the present application described with reference to the above figures can be executed. The computer-readable storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. The volatile memory may, for example, include a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may, for example, include a read-only memory (ROM), a hard disk, a flash memory, etc.

[0094] The embodiment of the present application also provides a computer program product or computer program, which includes computer executable instructions, and the computer executable instructions are stored in a computer readable storage medium. The processor of the computer device reads the computer executable instructions from the computer readable storage medium, and the processor executes the computer executable instructions, so that the computer device executes the fat loss training management method based on big data according to the embodiment of the present application.

[0095] Those skilled in the art will appreciate that the contents disclosed in this application may be subject to various variations and improvements. For example, the various devices or components described above may be implemented by hardware, or by software, firmware, or a combination of some or all of the three.

[0096] In addition, although the present application makes various references to certain units in the system according to embodiments of the present application, any number of different units can be used and run on the client and / or server. The units are only illustrative, and different aspects of the system and method can use different units.

[0097] Those skilled in the art will appreciate that all or part of the steps in the above method can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a disk or an optical disk. Optionally, all or part of the steps in the above embodiment can also be implemented using one or more integrated circuits. Accordingly, each module / unit in the above embodiment can be implemented in the form of hardware or in the form of software function modules. The present application is not limited to any particular form of combination of hardware and software.

[0098] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.

[0099] The above is an explanation of the present application and should not be considered as limiting thereof. Although several exemplary embodiments of the present application have been described, those skilled in the art will readily appreciate that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present application.

Claims

1. A fat loss training management device based on big data, characterized in that: include: A body fat scale data acquisition module, used to obtain a time series of the body composition data of the monitored subject, wherein the body composition data includes body weight, body fat percentage, muscle mass, bone mass, water content, basal metabolic rate, visceral fat level, body age, protein content, body mass index and lean body weight; A fat-reducing training program setting module is used to receive a fat-reducing training program input by a user; A big data module, used for performing feature encoding on the time series of the body composition data and the fat loss training program to obtain body composition time series associated aggregation features and fat loss training program semantic encoding features; A training management result generating module, used to determine the training management result based on the temporal association aggregation feature of the body composition and the semantic coding feature of the fat-reducing training program, wherein the training management result is used to indicate whether the fat-reducing training program needs to be adjusted; Wherein, the big data module includes: A body composition data time series correlation analysis unit, used for performing time series correlation analysis on the time series of the body composition data to obtain a time series of body composition time series correlation implicit feature vectors; A body component feature time series propagation aggregation unit is used to perform feature time series propagation aggregation on the time series of the body component time series associated implicit feature vector to obtain a body component time series propagation aggregation representation vector as the body component time series associated aggregation feature; A fat-reduction training program semantic coding unit, used for performing global semantic coding on the fat-reduction training program to obtain a fat-reduction training program semantic coding feature vector as the fat-reduction training program semantic coding feature; Wherein, the body composition data time series correlation analysis unit includes: an embedding coding subunit, configured to respectively embed code each body component data in the time series of the body component data using a body component embedding matrix to obtain a time series of body component fully connected embedding coding vectors; A time series correlation coding subunit, used for performing time series correlation coding on the time series of the body component fully connected embedded coding vector to obtain the time series of the body component time series correlation implicit feature vector; Wherein, the timing association coding subunit is used for: The time series of the body component fully connected embedded coding vector is input into the body component temporal association encoder based on the Bi-LSTM model to obtain the time series of the body component temporal association implicit feature vector.

2. The fat loss training management device based on big data according to claim 1, characterized in that: The body composition feature time series propagation aggregation unit is used to: The time series of the implicit feature vector associated with the body component time series is input into a feature forward propagation network based on significant time series attenuation of node energy to obtain the body component time series propagation aggregation representation vector.

3. The fat loss training management device based on big data according to claim 2 is characterized in that: The body composition feature time series propagation aggregation unit comprises: An energy significant descriptor calculation subunit, configured to calculate the energy significant descriptors of the respective body component time series associated implicit feature vectors in the time series of the body component time series associated implicit feature vectors based on the mean and variance of the respective body component time series associated implicit feature vectors to obtain a time series of energy significant descriptors; A timestamp marking subunit, used for extracting the timestamp of each body component time series associated implicit feature vector in the time series of the body component time series associated implicit feature vector to obtain a time series of timestamps; A historical node feature division subunit is used to use the body component time series association implicit feature vector corresponding to the current time point in the time series of the body component time series association implicit feature vector as the current body component time series association implicit feature vector, and use the body component time series association implicit feature vectors corresponding to other time points as the historical body component time series association implicit feature vectors, so as to obtain a time series of the current body component time series association implicit feature vector and the historical body component time series association implicit feature vector; The attenuation factor calculation subunit is used to calculate the energy time attenuation factor of each historical body component time series associated implicit feature vector in the time series of the historical body component time series associated implicit feature vector relative to the current body component time series associated implicit feature vector to obtain a time series of energy time attenuation factors; A historical node propagation aggregation subunit is used to perform time series propagation aggregation on the time series of the historical body component time series associated implicit feature vectors using the time series of the energy significant descriptor as a positive adjustment factor and the time series of the energy time decay factor as a negative adjustment factor to obtain a historical node body component time series propagation aggregation representation feature vector; A global time series aggregation subunit, used for fusing the historical node body component time series propagation aggregation representation feature vector and the current body component time series associated implicit feature vector to obtain the body component time series propagation aggregation representation vector; Wherein, the attenuation factor calculation subunit is used for: Calculating a timestamp difference between the historical body composition time series associated implicit feature vector and the current body composition time series associated implicit feature vector; The timestamp difference is rounded down and divided by a preset time decay period and then squared, and the square result is multiplied by a preset decay rate to obtain an initial energy decay coefficient; The energy time decay factor is obtained by calculating an exponential function value with the natural constant as the base and the initial energy decay coefficient as the exponent.

4. The fat loss training management device based on big data according to claim 3 is characterized in that: The training management result generating module comprises: A position difference unit, used for calculating the position difference between the semantic encoding feature vector of the fat loss training program and the body composition temporal propagation aggregation representation vector to obtain a semantic offset difference representation vector; The training management result generating unit is used to input the semantic shift differential representation vector into the classifier-based training management result generator to obtain the training management result.

5. A fat loss training management method based on big data, characterized in that: include: Obtaining a time series of body composition data of the monitored subject, wherein the body composition data includes weight, body fat percentage, muscle mass, bone mass, water content, basal metabolic rate, visceral fat level, body age, protein content, body mass index and lean body mass; receiving a fat loss training program input by a user; Performing feature encoding on the time series of the body composition data and the fat loss training program to obtain body composition time series associated aggregation features and fat loss training program semantic encoding features; Determine a training management result based on the body composition temporal association aggregation feature and the semantic coding feature of the fat loss training program, wherein the training management result is used to indicate whether the fat loss training program needs to be adjusted; The time series of the body composition data and the fat loss training program are feature encoded to obtain the body composition time series associated aggregation features and the fat loss training program semantic encoding features, including: Performing time series correlation analysis on the time series of the body composition data to obtain a time series of body composition time series correlation implicit feature vectors; Performing feature time series propagation aggregation on the time series of the implicit feature vector of the body component time series association to obtain a body component time series propagation aggregation representation vector as the body component time series association aggregation feature; Performing global semantic coding on the fat-reducing training program to obtain a semantic coding feature vector of the fat-reducing training program as a semantic coding feature of the fat-reducing training program; The step of performing time series correlation analysis on the time series of the body composition data to obtain a time series of body composition time series correlation implicit feature vectors includes: Using a body component embedding matrix, each body component data in the time series of the body component data is respectively embedded and encoded to obtain a time series of body component fully connected embedded coding vectors; Performing temporal association coding on the time series of the body component fully connected embedded coding vector to obtain the time series of the body component temporal association implicit feature vector; The step of performing temporal association coding on the time series of the fully connected embedded coding vector of the body components to obtain the time series of the temporal association implicit feature vector of the body components includes: The time series of the body component fully connected embedded coding vector is input into the body component temporal association encoder based on the Bi-LSTM model to obtain the time series of the body component temporal association implicit feature vector.

6. An electronic device, characterized in that: include: A memory for storing instructions; A processor is coupled to the memory, and the processor is configured to execute the fat loss training management method based on big data as described in claim 5 based on the instructions stored in the memory.

Images