CN117133407B - Nutritional balance assessment method and system for multi-label neural network for children - Google Patents
Nutritional balance assessment method and system for multi-label neural network for children Download PDFInfo
- Publication number
- CN117133407B CN117133407B CN202311396009.4A CN202311396009A CN117133407B CN 117133407 B CN117133407 B CN 117133407B CN 202311396009 A CN202311396009 A CN 202311396009A CN 117133407 B CN117133407 B CN 117133407B
- Authority
- CN
- China
- Prior art keywords
- nutritional
- information
- diet
- nutrition
- target object
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 235000016709 nutrition Nutrition 0.000 title claims abstract description 229
- 238000000034 method Methods 0.000 title claims abstract description 30
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 16
- 235000005911 diet Nutrition 0.000 claims abstract description 119
- 230000037213 diet Effects 0.000 claims abstract description 106
- 230000035764 nutrition Effects 0.000 claims abstract description 102
- 230000033001 locomotion Effects 0.000 claims abstract description 51
- 235000005289 dietary characteristics Nutrition 0.000 claims abstract description 40
- 238000013210 evaluation model Methods 0.000 claims abstract description 35
- 235000021004 dietary regimen Nutrition 0.000 claims abstract description 9
- 238000007781 pre-processing Methods 0.000 claims abstract description 8
- 238000003062 neural network model Methods 0.000 claims abstract description 7
- 230000006870 function Effects 0.000 claims description 48
- 239000011159 matrix material Substances 0.000 claims description 47
- 235000015097 nutrients Nutrition 0.000 claims description 39
- 239000004615 ingredient Substances 0.000 claims description 30
- 238000012549 training Methods 0.000 claims description 28
- 238000011156 evaluation Methods 0.000 claims description 26
- 235000012041 food component Nutrition 0.000 claims description 22
- 230000003044 adaptive effect Effects 0.000 claims description 19
- 230000000378 dietary effect Effects 0.000 claims description 13
- 230000004927 fusion Effects 0.000 claims description 11
- 235000006180 nutrition needs Nutrition 0.000 claims description 6
- 238000000354 decomposition reaction Methods 0.000 claims description 5
- 238000009395 breeding Methods 0.000 claims description 4
- 230000001488 breeding effect Effects 0.000 claims description 4
- 238000004590 computer program Methods 0.000 claims description 4
- 230000035772 mutation Effects 0.000 claims description 4
- 238000003860 storage Methods 0.000 claims description 3
- 239000013598 vector Substances 0.000 description 10
- 235000013305 food Nutrition 0.000 description 9
- 230000012010 growth Effects 0.000 description 8
- 108090000623 proteins and genes Proteins 0.000 description 7
- 239000000203 mixture Substances 0.000 description 6
- 238000011161 development Methods 0.000 description 5
- 230000018109 developmental process Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000001514 detection method Methods 0.000 description 4
- 238000013507 mapping Methods 0.000 description 4
- 235000012054 meals Nutrition 0.000 description 4
- 235000018343 nutrient deficiency Nutrition 0.000 description 4
- 238000005457 optimization Methods 0.000 description 4
- 238000009826 distribution Methods 0.000 description 3
- 230000037406 food intake Effects 0.000 description 3
- 230000036541 health Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 239000005417 food ingredient Substances 0.000 description 2
- 238000012163 sequencing technique Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 208000024891 symptom Diseases 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 206010010356 Congenital anomaly Diseases 0.000 description 1
- 102000008297 Nuclear Matrix-Associated Proteins Human genes 0.000 description 1
- 108010035916 Nuclear Matrix-Associated Proteins Proteins 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 235000013361 beverage Nutrition 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 230000004641 brain development Effects 0.000 description 1
- 235000014633 carbohydrates Nutrition 0.000 description 1
- 150000001720 carbohydrates Chemical class 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 230000003930 cognitive ability Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 235000020979 dietary recommendations Nutrition 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000003925 fat Substances 0.000 description 1
- 235000019197 fats Nutrition 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 229910052500 inorganic mineral Inorganic materials 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000004060 metabolic process Effects 0.000 description 1
- 239000011707 mineral Substances 0.000 description 1
- 235000010755 mineral Nutrition 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000000299 nuclear matrix Anatomy 0.000 description 1
- 230000037081 physical activity Effects 0.000 description 1
- 235000018102 proteins Nutrition 0.000 description 1
- 102000004169 proteins and genes Human genes 0.000 description 1
- 235000013343 vitamin Nutrition 0.000 description 1
- 229940088594 vitamin Drugs 0.000 description 1
- 229930003231 vitamin Natural products 0.000 description 1
- 239000011782 vitamin Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
- G16H20/60—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to nutrition control, e.g. diets
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/211—Selection of the most significant subset of features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
- G16H10/65—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records stored on portable record carriers, e.g. on smartcards, RFID tags or CD
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Epidemiology (AREA)
- Biomedical Technology (AREA)
- Primary Health Care (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Biophysics (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Nutrition Science (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention provides a nutrition balance assessment method and a system of a multi-label neural network for children, which relate to the field of nutrition and comprise the steps of recording diet information of a target object in a certain time period and acquiring motion information of the target object in the certain time period through wearable equipment, and extracting diet characteristics corresponding to the diet information and motion characteristics corresponding to the motion information through a feature selection algorithm after carrying out data preprocessing on the diet information and the motion information; according to the dietary characteristics, whether the dietary characteristics meet the nutritional requirements of the target object is evaluated through a nutritional evaluation model, wherein the nutritional evaluation model is constructed based on a multi-label neural network model and is used for evaluating whether various input information of an input model meets preset requirements; if the dietary characteristics do not meet the nutritional requirements of the target subject, providing a dietary regimen that meets the nutritional requirements of the target subject according to the nutritional requirement bias and intake balance of the target subject.
Description
Technical Field
The invention relates to nutrition technology, in particular to a nutrition balance assessment method and system for a multi-label neural network for children.
Background
In the growth process of children, the absorption of nutrients is influenced by congenital factors and acquired factors, and different people absorb different nutrients in the growth process, so that the situation that the children lack nutrients in the growth process is caused, but the fastest period of human physique and brain development is 0-6 years old, and the nutrition intake and growth and development state in the period affect the life of the children.
CN110033845a, a child nutrient deficiency risk rating system, comprising: the data acquisition module is used for acquiring gene detection data, child meal data and child body index data; the gene detection result evaluation module is used for evaluating the nutrient related genes according to the gene detection data; a meal intake nutrient level assessment module for assessing meal intake nutrient levels based on the child meal data; the nutrient deficiency symptom assessment module is used for assessing the nutrient deficiency symptoms of the children according to the physical index data of the children; and the comprehensive risk level evaluation module is used for comprehensively analyzing according to each evaluation result to determine the risk of nutrient deficiency.
CN111242427a discloses a method and a system for evaluating the relationship between nutrition and growth and development of children, wherein the method for evaluating the relationship between nutrition and growth and development of children comprises the following steps: personal information data of the name and the age of the child are collected, and height data, weight data and blood element index data of the child are collected; drawing a relationship curve image of nutrition and growth and development of the children according to the acquired data, and testing the psychology of the children; evaluating the development ability, the cognitive ability and the comprehensive ability of the children; and finally, storing and displaying the acquired data.
In the prior art, although the condition of the ingested nutrient can be determined by detection, an individual nutrient ingestion plan cannot be formulated for each child according to the physical condition and ingestion habit of each child.
Disclosure of Invention
The embodiment of the invention provides a nutrition balance assessment method and a nutrition balance assessment system for a multi-label neural network for children, which can assess whether the nutrition intake of the children is balanced and reasonable according to the information such as the actual intake nutrition position and the movement condition of the children.
In a first aspect of an embodiment of the present invention,
provided is a nutrition balance assessment method for a multi-label neural network for children, comprising:
recording diet information of a target object within a certain time period, acquiring motion information of the target object within the certain time period through wearable equipment, preprocessing the diet information and the motion information, and extracting diet characteristics corresponding to the diet information and motion characteristics corresponding to the motion information through a characteristic selection algorithm;
according to the dietary characteristics, combining the movement characteristics with the basic attribute characteristics of the target object, and evaluating whether the dietary characteristics meet the nutritional requirements of the target object through a pre-constructed nutritional evaluation model, wherein the nutritional evaluation model is constructed based on a multi-label neural network model and is used for evaluating whether various input information of an input model meets the preset requirements;
and if the dietary characteristic does not meet the nutritional requirement of the target object, providing a dietary scheme meeting the nutritional requirement of the target object according to the nutritional requirement deviation and the intake balance of the target object.
In an alternative embodiment of the present invention,
extracting the diet characteristics corresponding to the diet information and the exercise characteristics corresponding to the exercise information through a characteristic selection algorithm comprises the following steps:
normalizing the diet information and the exercise information, and determining a diet kernel matrix corresponding to the normalized diet information and a exercise kernel matrix corresponding to the normalized exercise information by using a self-adaptive kernel function, wherein the kernel matrix is used for indicating the inner product of the diet information and the exercise information in a high-dimensional space, and the self-adaptive kernel function;
performing decentralization treatment on the diet core matrix and the exercise core matrix respectively to obtain an adaptive diet matrix and an adaptive exercise matrix, performing eigenvalue decomposition on the adaptive diet matrix and the adaptive exercise matrix, and determining diet eigenvalue corresponding to the adaptive diet matrix and exercise eigenvalue corresponding to the adaptive exercise matrix;
and sequencing according to the sizes of the diet characteristic values and the exercise characteristic values, and taking the characteristics corresponding to the characteristic values, of which the space dimensions of the diet characteristic values and the exercise characteristic values meet the preset dimension conditions, as the diet characteristics corresponding to the diet information and the exercise characteristics corresponding to the exercise information.
In an alternative embodiment of the present invention,
according to the dietary characteristics, combining the exercise characteristics and the pre-acquired basic attribute characteristics of the target object, and evaluating whether the dietary characteristics meet the nutritional requirements of the target object through a pre-constructed nutritional evaluation model comprises:
determining a plurality of nutritional components corresponding to the dietary characteristics through the nutritional assessment model;
fusing the basic attribute characteristics and the motion characteristics of the target object to obtain attribute fusion characteristics, and determining energy demand information and nutrition demand information corresponding to the attribute fusion characteristics based on a preset nutrition attribute corresponding relation;
judging whether the multiple nutritional components are matched with the energy demand information and the nutrition demand information, if so, meeting the nutrition requirements of the target object, and if not, not meeting the nutrition requirements of the target object.
In an alternative embodiment of the present invention,
the method further comprises training a nutritional assessment model:
acquiring a nutrition evaluation training data set, wherein a sample of the nutrition evaluation training data set comprises a plurality of nutrition components, different nutrition components are regarded as different tasks, and a sample label corresponding to the sample of the nutrition evaluation training data set is determined;
initializing network weight and bias parameters of a nutrition evaluation model to be trained, inputting a nutrition evaluation data set to be trained into the nutrition evaluation model to be trained, determining a model output of the nutrition evaluation model and a prediction deviation of a sample label corresponding to a sample of the nutrition evaluation training data set;
based on the correlation between different nutritional ingredients, and assigning a corresponding nutritional weight for each nutritional ingredient, and assigning a corresponding correlation weight for the correlation between different nutritional ingredients, constructing a loss function in combination with the prediction bias;
according to the loss value of the loss function, determining the gradient of the network weight and the bias parameter of the nutrition evaluation model to be trained on the loss function through a back propagation algorithm, iteratively and automatically adjusting the learning rate and updating the network weight and the bias parameter of the nutrition evaluation model to be trained, so that the loss value of the loss function is minimized.
In an alternative embodiment of the present invention,
based on the correlation between the different nutritional ingredients, and assigning a corresponding nutritional weight to each nutritional ingredient, and assigning a corresponding correlation weight to the correlation between the different nutritional ingredients, in combination with the prediction bias, constructing a loss function comprises:
wherein,LOSSrepresenting the loss value corresponding to the loss function,i,jrespectively represent the firstiSeed nutrient composition and firstjThe seed is a nutrient component,Nthe amount of the nutrient components is indicated,w i represent the firstiThe nutrition weight corresponding to the nutrition ingredients of the seeds,p i 、y i respectively represent the firstiModel output and model number corresponding to seed nutrient compositioniSample tags corresponding to the seed nutrient components,r ij represent the firstiSeed nutrient composition and firstjThe correlation weight corresponding to the nutrient components,y j represent the firstjSample labels corresponding to nutritional ingredients.
In an alternative embodiment of the present invention,
providing a dietary regimen that meets the nutritional requirements of the target subject based on the nutritional requirement bias and intake balance of the target subject if the dietary characteristics do not meet the nutritional requirements of the target subject comprises:
if the dietary characteristics do not meet the nutritional requirements of the target object, determining an ideal intake of each nutritional component meeting the nutritional requirements of the target object according to the basic attribute characteristics of the target object and the nutritional requirements of the target object, determining an actual intake by combining the dietary information of the target object, and determining a nutritional demand deviation according to the ideal intake and the actual intake;
determining an average intake of the nutritional components of the target subject based on the dietary information, determining intake balance in combination with actual intake;
constructing an fitness function based on the nutrition demand deviation and the intake balance, sorting according to fitness values corresponding to the fitness function, and selecting an individual with the fitness value larger than a preset sorting threshold as a parent for breeding the next generation; performing cross operation on the selected parent individuals to generate offspring individuals, performing mutation operation on the crossed individuals, introducing random factors, and generating new offspring individuals;
determining a difference between the parent individual and the offspring individual,
if the difference value is larger than a preset reference threshold value, replacing the child individuals with the parent individuals;
and if the difference value is smaller than a preset reference threshold value, reserving the child individuals in combination with a preset acceptance threshold value, and taking the final reserved child individuals as a diet scheme meeting the nutritional requirements of the target object.
Second aspect of embodiments of the invention
There is provided a nutritional balance assessment system for a multi-labeled neural network for children, comprising:
the first unit is used for recording diet information of a target object in a certain time period and acquiring motion information of the target object in the certain time period through a wearable device, and extracting diet characteristics corresponding to the diet information and motion characteristics corresponding to the motion information through a feature selection algorithm after carrying out data preprocessing on the diet information and the motion information;
the second unit is used for evaluating whether the dietary characteristics meet the nutritional requirements of the target object through a pre-constructed nutritional evaluation model according to the dietary characteristics and combining the movement characteristics and the pre-acquired basic attribute characteristics of the target object, wherein the nutritional evaluation model is constructed based on a multi-label neural network model and is used for evaluating whether various input information of an input model meets the preset requirements;
and a third unit for providing a dietary regimen that meets the nutritional requirements of the target subject according to the nutritional requirement deviation and intake balance of the target subject if the dietary characteristics do not meet the nutritional requirements of the target subject.
In a third aspect of an embodiment of the present invention,
there is provided an electronic device including:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to invoke the instructions stored in the memory to perform the method described previously.
In a fourth aspect of an embodiment of the present invention,
there is provided a computer readable storage medium having stored thereon computer program instructions which, when executed by a processor, implement the method as described above.
The beneficial effects of the embodiments of the present invention may refer to the effects corresponding to technical features in the specific embodiments, and are not described herein.
Drawings
FIG. 1 is a flow chart of a method for nutritional balance assessment for a child's multi-labeled neural network according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a nutrition balance evaluation system for a multi-label neural network for children according to an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The technical scheme of the invention is described in detail below by specific examples. The following embodiments may be combined with each other, and some embodiments may not be repeated for the same or similar concepts or processes.
Fig. 1 is a flow chart of a nutrition balance evaluation method for a multi-label neural network for children according to an embodiment of the present invention, as shown in fig. 1, the method includes:
s1, recording diet information of a target object in a certain time period and acquiring motion information of the target object in the certain time period through a wearable device, preprocessing data of the diet information and the motion information, and extracting diet characteristics corresponding to the diet information and motion characteristics corresponding to the motion information through a characteristic selection algorithm;
the diet information refers to various relevant data and knowledge about foods and beverages, including information about ingredients, nutritional values and the like of the foods;
the exercise information refers to data and knowledge related to physical exercise, exercise and physical activity, and specifically comprises information such as exercise type, exercise intensity, exercise frequency and the like;
the dietary characteristics are representative characteristics or modes extracted after dietary information of a target object in a certain time period is processed and analyzed;
the motion characteristics are representative motion behaviors or modes extracted after motion information of a target object in a certain time period is processed and analyzed;
in an alternative embodiment of the present invention,
extracting the diet characteristics corresponding to the diet information and the exercise characteristics corresponding to the exercise information through a characteristic selection algorithm comprises the following steps:
normalizing the diet information and the exercise information, and determining a diet kernel matrix corresponding to the normalized diet information and a exercise kernel matrix corresponding to the normalized exercise information by using a self-adaptive kernel function, wherein the kernel matrix is used for indicating the inner product of the diet information and the exercise information in a high-dimensional space, and the self-adaptive kernel function;
the self-adaptive kernel function is a kernel function and is characterized in that parameters in the kernel function can be automatically adjusted to adapt to the characteristics of data, and the purpose is to improve the generalization capability of a model so that the model can be better adapted to the characteristics of different data sets.
Collecting and collating data of diet information and exercise information and ensuring consistency and availability of data formats;
for diet information and exercise information, converting text information into numerical characteristics or using the existing characteristic vector to represent, extracting characteristics and representing the characteristics as vector forms;
selecting a proper self-adaptive kernel function, and dynamically adjusting parameters of the kernel function according to the distribution of data;
calculating a kernel matrix of the diet information according to the selected self-adaptive kernel function;
normalizing the nuclear matrix of dietary information using means and standard deviations;
performing decentralization treatment on the diet core matrix and the exercise core matrix respectively to obtain an adaptive diet matrix and an adaptive exercise matrix, performing eigenvalue decomposition on the adaptive diet matrix and the adaptive exercise matrix, and determining diet eigenvalue corresponding to the adaptive diet matrix and exercise eigenvalue corresponding to the adaptive exercise matrix;
the diet eigenvalues, after matrix decentralization, are metrics that describe the importance of each dimension in the matrix of diet information. Each feature value corresponds to a feature vector, and the feature vector defines the direction of the diet information in the data space;
the motion feature value describes the importance of each dimension in the matrix of motion information. Similarly, each feature value corresponds to a feature vector, and the direction of the motion information in the data space is defined;
firstly, calculating the average value of each row of the diet core matrix, and subtracting the average value of the corresponding row from each element to change the matrix into a zero-average matrix to obtain a decentralized diet core matrix;
and calculating the average value of each row of the motion kernel matrix, and subtracting the average value of the corresponding row from each element to obtain the decentered motion kernel matrix.
Performing eigenvalue decomposition on the decentered diet kernel matrix and the decentered motion kernel matrix to obtain diet eigenvalue and corresponding eigenvector and motion eigenvalue and corresponding eigenvector;
and sequencing according to the sizes of the diet characteristic values and the exercise characteristic values, and taking the characteristics corresponding to the characteristic values, of which the space dimensions of the diet characteristic values and the exercise characteristic values meet the preset dimension conditions, as the diet characteristics corresponding to the diet information and the exercise characteristics corresponding to the exercise information.
Ordering the diet characteristic value and the exercise characteristic value according to the sequence from big to small;
determining the number of required dimensions, and selecting the number of the maximum eigenvalues and the corresponding eigenvectors to be reserved according to preset dimension conditions;
selecting feature vectors corresponding to the sequenced feature values, wherein the feature vectors are to be the features of diet information and motion information, and the number of the feature vectors is the number of the features meeting the preset dimension condition;
the selected feature vectors are regarded as features of diet information and exercise information, respectively.
In the step, the self-adaptive kernel function can adjust parameters of the kernel function according to data distribution, and the adaptability of the model is improved. This helps to more accurately capture the intrinsic structure of diet and exercise information;
the construction of the kernel matrix can capture the similarity and relevance between different information. By calculating the inner product of the kernel matrix, the similarity between diet information and exercise information can be measured in a high-dimensional space;
the data stability can be improved by decentralization, and the data cannot be influenced by bias in the processing process. The main characteristics of the data can be extracted by the decomposition of the characteristic values, so that the dimension of the data is reduced;
the most important feature vectors are selected according to the magnitude order of the feature values. This helps to reduce noise, preserve primary information, and improve accuracy of subsequent analysis;
in summary, this step helps to better understand the eating and exercise habits of the child, thereby more accurately assessing their nutritional balance. This facilitates the design of personalized, targeted interventions that promote the healthy growth of children.
S2, according to the dietary characteristics, combining the movement characteristics and the basic attribute characteristics of the target object, and evaluating whether the dietary characteristics meet the nutritional requirements of the target object through a pre-constructed nutritional evaluation model, wherein the nutritional evaluation model is constructed based on a multi-label neural network model and is used for evaluating whether various input information of an input model meets the preset requirements;
the nutrition evaluation model is a model for evaluating whether the dietary and exercise characteristics of the target object meet the nutrition requirements thereof;
the basic attribute features refer to basic information of the target object, including but not limited to age, sex, height, weight, etc. These features have important roles in assessing an individual's energy demand, nutritional metabolism, and overall health;
in an alternative embodiment of the present invention,
according to the dietary characteristics, combining the exercise characteristics and the pre-acquired basic attribute characteristics of the target object, and evaluating whether the dietary characteristics meet the nutritional requirements of the target object through a pre-constructed nutritional evaluation model comprises:
determining a plurality of nutritional components corresponding to the dietary characteristics through the nutritional assessment model;
preparing a database containing nutritional information of various foods, such as food ingredients database containing nutritional ingredients of various foods, such as proteins, carbohydrates, fats, vitamins, minerals, etc.;
matching the food in the diet feature with the food in the database by establishing a mapping relation, and mapping the diet feature to the corresponding food in the nutrition database;
calculating a plurality of nutritional ingredients corresponding to each dietary characteristic by using the nutritional ingredient information in the database;
fusing the basic attribute characteristics and the motion characteristics of the target object to obtain attribute fusion characteristics, and determining energy demand information and nutrition demand information corresponding to the attribute fusion characteristics based on a preset nutrition attribute corresponding relation;
collecting basic attribute characteristics of the target object, such as age, gender, height, weight and the like, and exercise characteristics, such as exercise type, intensity, frequency and the like;
the basic attribute features and the motion features of the target object are fused through simple splicing, weighted average or feature extraction methods, so that a comprehensive attribute fusion feature is obtained;
establishing a preset nutrition attribute corresponding relation, namely determining the relation between different attribute fusion characteristics, energy demand information and nutrition demand information;
mapping the attribute fusion characteristics to corresponding energy demand information and nutrition demand information according to a preset nutrition attribute corresponding relation;
and calculating the energy demand information and the nutrition demand information corresponding to the target object by using the relation obtained by mapping, and explaining the calculated energy demand information and nutrition demand information.
Judging whether the multiple nutritional components are matched with the energy demand information and the nutrition demand information, if so, meeting the nutrition requirements of the target object, and if not, not meeting the nutrition requirements of the target object.
Acquiring a plurality of nutritional components determined according to the nutritional assessment model;
matching the plurality of nutritional ingredient data with the energy demand information and the nutritional demand information;
setting a matching standard, namely defining matching conditions among various nutrient components, energy demand information and nutrition demand information;
judging whether the multiple nutritional ingredients are matched with the energy demand information and the nutritional demand information according to the set matching standard, if so, considering that the nutritional requirements are met, otherwise, not meeting the nutritional requirements;
and outputting a judging result to clearly indicate whether the target object meets the nutrition requirement and which aspects are possibly not matched.
In an alternative embodiment of the present invention,
the method further comprises training a nutritional assessment model:
acquiring a nutrition evaluation training data set, wherein a sample of the nutrition evaluation training data set comprises a plurality of nutrition components, different nutrition components are regarded as different tasks, and a sample label corresponding to the sample of the nutrition evaluation training data set is determined;
determining a data source from which the nutritional assessment training dataset was obtained;
collecting data related to nutritional assessment based on the determined data source;
cleaning and preprocessing the collected data, including missing value processing, abnormal value processing and data format unification;
performing feature engineering according to the nutrition evaluation target, including creating new features according to the features in the data to improve the performance of the model, and normalizing or normalizing the features to ensure the stability of model training;
initializing network weight and bias parameters of a nutrition evaluation model to be trained, inputting a nutrition evaluation data set to be trained into the nutrition evaluation model to be trained, determining a model output of the nutrition evaluation model and a prediction deviation of a sample label corresponding to a sample of the nutrition evaluation training data set;
initializing the weight and bias parameters of the neural network by using methods such as random initialization, loading of a pre-training model and the like according to the selected nutrition evaluation model;
selecting an appropriate loss function to measure the difference between the model output and the sample label;
setting learning rate and super parameters of other optimizers by selecting an optimization algorithm;
inputting the preprocessed training data set into the model and ensuring that the dimension of the input data is matched with the input layer of the model;
performing forward propagation operation, and calculating input data through each layer of the neural network to obtain the output of the model;
calculating a difference of the model output and the sample label using the selected loss function;
performing back propagation operation, transferring error gradient backwards through the network layer, and calculating gradient of loss to model parameters;
updating the weights and biases of the model according to the gradients using the selected optimizers;
repeatedly executing signing operation, and iterating the training model until training is completed;
inputting the test data set into a model to obtain a model output prediction deviation of a sample label corresponding to a sample of the nutrition evaluation training data set
Based on the correlation between different nutritional ingredients, and assigning a corresponding nutritional weight for each nutritional ingredient, and assigning a corresponding correlation weight for the correlation between different nutritional ingredients, constructing a loss function in combination with the prediction bias;
calculating the correlation between different nutritional ingredients using the correlation coefficients;
assigning weights to each nutritional component based on domain knowledge, experimental data, or other prior information;
weighting the correlations between different nutritional ingredients;
constructing a loss function based on the correlation and the correlation weight;
according to the loss value of the loss function, determining the gradient of the network weight and the bias parameter of the nutrition evaluation model to be trained on the loss function through a back propagation algorithm, iteratively and automatically adjusting the learning rate and updating the network weight and the bias parameter of the nutrition evaluation model to be trained, so that the loss value of the loss function is minimized.
Initializing weight and bias parameters in a model, and preparing a training data set comprising input data and corresponding target labels;
initializing a learning rate and iteration times, and performing iterative training;
inputting training data into a model, calculating the output of the model, calculating the difference between the model output and an actual sample label by using a defined loss function, calculating the gradient of the loss function to model parameters (weight and bias) by using a back propagation algorithm, dynamically adjusting the learning rate according to the requirement, and updating the parameters of the model according to the gradient and the learning rate by using a gradient descent or other optimization algorithms;
after each iteration, checking whether the loss value is small enough or tends to be stable, and if the loss value meets the condition of stopping training, stopping training;
in an alternative embodiment of the present invention,
based on the correlation between the different nutritional ingredients, and assigning a corresponding nutritional weight to each nutritional ingredient, and assigning a corresponding correlation weight to the correlation between the different nutritional ingredients, in combination with the prediction bias, constructing a loss function comprises:
wherein,LOSSrepresenting the loss value corresponding to the loss function,i,jrespectively represent the firstiSeed nutrient composition and firstjThe seed is a nutrient component,Nthe amount of the nutrient components is indicated,w i represent the firstiThe nutrition weight corresponding to the nutrition ingredients of the seeds,p i 、y i respectively represent the firstiModel output and model number corresponding to seed nutrient compositioniSample tags corresponding to the seed nutrient components,r ij represent the firstiSeed nutrient composition and firstjThe correlation weight corresponding to the nutrient components,y j represent the firstjCorresponding to the nutrient componentsSample tags.
In the function, the correlation and the weight among different nutritional components are considered, the model is more likely to generate a nutritional evaluation result which is more matched with the actual situation, personalized evaluation can be carried out according to the importance of each component by distributing the weight to the different nutritional components, so that the nutritional requirements of a target object are better met, and the correlation among different nutritional components can be better captured by introducing the correlation weight.
In this step, the nutrition evaluation model can accurately determine a plurality of nutritional components corresponding to the characteristics of the diet. This helps to fully understand the dietary status of the target subject;
basic attributes and motion characteristics of the target object are fused to obtain attribute fusion characteristics, and individual differences can be considered more comprehensively. The energy demand information and the nutrition demand information corresponding to the attribute fusion characteristics are determined through the preset nutrition attribute corresponding relation, so that the assessment is more personalized and accurate;
judging whether or not the plurality of nutrient components are matched with the energy demand information and the nutrition demand information can provide a judgment as to whether or not the target object satisfies the nutrition demand. This helps to adjust the diet and lifestyle in time to meet the health needs of the child;
acquiring a training data set containing multiple nutritional components, treating different nutritional components as different tasks, which helps the model to better learn and understand the relationship between the tasks;
constructing a loss function that includes a number of aspects of consideration, including weight distribution, correlation, and predictive bias, etc., can more fully scale the performance of the model. Optimizing a model through a back propagation algorithm, so that the comprehensive loss is minimized;
in summary, the step considers a plurality of factors, so that the assessment is more comprehensive and personalized, and the accuracy and the practicability of the nutrition assessment of the children are improved.
S3, if the dietary characteristics do not meet the nutritional requirements of the target object, providing a dietary scheme meeting the nutritional requirements of the target object according to the nutritional requirement deviation and intake balance of the target object.
In an alternative embodiment of the present invention,
providing a dietary regimen that meets the nutritional requirements of the target subject based on the nutritional requirement bias and intake balance of the target subject if the dietary characteristics do not meet the nutritional requirements of the target subject comprises:
if the dietary characteristics do not meet the nutritional requirements of the target object, determining an ideal intake of each nutritional component meeting the nutritional requirements of the target object according to the basic attribute characteristics of the target object and the nutritional requirements of the target object, determining an actual intake by combining the dietary information of the target object, and determining a nutritional demand deviation according to the ideal intake and the actual intake;
defining a nutrition standard according to medical and nutritional knowledge, setting a threshold value, and judging whether the dietary characteristics are in a reasonable range or not, thereby judging whether the dietary characteristics meet the nutrition requirements of a target object or not;
if not, determining an ideal intake of each nutritional ingredient according to specialized nutritional knowledge, medical standards, or recommended intake issued by a specialized institution;
determining actual diet data of the target subject, including intake of various foods, food ingredients, and the like;
calculating, for each nutrient component, a difference between the actual intake and the ideal intake to determine a nutrient demand deviation;
determining an average intake of the nutritional components of the target subject based on the dietary information, determining intake balance in combination with actual intake;
obtaining diet information of a target object, wherein the diet information comprises information such as the type of food to be ingested, the ingestion amount, the diet frequency and the like;
dividing the total amount of intake of the target subject in a certain time range by the time for each nutrient component to obtain an average intake;
for each nutrient component, calculating the difference between the actual intake and the average intake for assessing intake balance;
the calculated intake balance is interpreted to determine which nutrients are in excess or in deficiency.
Constructing an fitness function based on the nutrition demand deviation and the intake balance, sorting according to fitness values corresponding to the fitness function, and selecting an individual with the fitness value larger than a preset sorting threshold as a parent for breeding the next generation; performing cross operation on the selected parent individuals to generate offspring individuals, performing mutation operation on the crossed individuals, introducing random factors, and generating new offspring individuals;
the fitness function can comprehensively consider the nutrition demand deviation and intake balance to form an index for evaluating the health condition of the target object;
calculating the fitness function value of each individual, and sorting according to the fitness value from large to small;
selecting an individual with the fitness value larger than a preset ranking threshold value as a parent for breeding the next generation according to the ranking result;
exchanging the gene information of two parent individuals through cross operation to generate offspring individuals;
introducing random factors, and mutating the gene information of the individual to generate a new offspring individual;
combining parent individuals and generated offspring individuals to form a new generation population;
determining a difference between the parent individual and the offspring individual,
if the difference value is larger than a preset reference threshold value, replacing the child individuals with the parent individuals;
and if the difference value is smaller than a preset reference threshold value, reserving the child individuals in combination with a preset acceptance threshold value, and taking the final reserved child individuals as a diet scheme meeting the nutritional requirements of the target object.
For each individual, calculating the difference of the fitness values by adopting the absolute value difference of each fitness value or other measurement modes;
if the difference value is larger than a preset reference threshold value, selecting replacement, and replacing the parent individuals with the child individuals;
if the difference value is smaller than a preset reference threshold value, reserving offspring individuals by combining with a preset acceptance threshold value;
if a substitution is selected, the child individuals are substituted for the parent individuals. If a reservation is selected, reserving the offspring individuals;
the final diet regimen consisted of selected offspring individuals that met the nutritional requirements of the target subject.
In the step, by constructing the fitness function and comprehensively considering the nutrition demand deviation and the intake balance, the quality of each diet scheme can be more comprehensively evaluated. Such personalized assessment helps to formulate dietary advice more in line with actual needs based on the specific circumstances of the target subject;
the genetic algorithm is utilized for propagation and evolution, so that the diet scheme of each generation is optimized to a certain extent on the basis of the previous generation. By selecting individuals with higher fitness values as parents and randomness introduced by crossover and mutation operations, more excellent diet schemes can be quickly found;
by judging the difference between the parent and the offspring, a feedback-based adaptive optimization strategy is realized. For individuals with poor performance, a replacement strategy is adopted, and for individuals with good performance, the diversity of the search space is increased by keeping a certain proportion, so that the balance of global search and local optimization is facilitated;
through multi-generation evolution, the final offspring individual is a diet proposal which meets the nutrition requirement of the target object on the basis of considering a plurality of factors such as ideal intake, actual intake, fitness evaluation and the like. This approach helps provide a more comprehensive and scientific dietary recommendation to meet the nutritional needs of children.
In conclusion, the method is helpful for optimizing the search space and improving the individuation and the overall fitness of the diet scheme, and an effective method is provided for formulating the diet scheme which meets the actual demands of children.
Fig. 2 is a schematic structural diagram of a nutrition balance evaluation system for a multi-label neural network for children according to an embodiment of the present invention, as shown in fig. 2, the system includes:
the first unit is used for recording diet information of a target object in a certain time period and acquiring motion information of the target object in the certain time period through a wearable device, and extracting diet characteristics corresponding to the diet information and motion characteristics corresponding to the motion information through a feature selection algorithm after carrying out data preprocessing on the diet information and the motion information;
the second unit is used for evaluating whether the dietary characteristics meet the nutritional requirements of the target object through a pre-constructed nutritional evaluation model according to the dietary characteristics and combining the movement characteristics and the pre-acquired basic attribute characteristics of the target object, wherein the nutritional evaluation model is constructed based on a multi-label neural network model and is used for evaluating whether various input information of an input model meets the preset requirements;
and a third unit for providing a dietary regimen that meets the nutritional requirements of the target subject according to the nutritional requirement deviation and intake balance of the target subject if the dietary characteristics do not meet the nutritional requirements of the target subject.
The present invention may be a method, apparatus, system, and/or computer program product. The computer program product may include a computer readable storage medium having computer readable program instructions embodied thereon for performing various aspects of the present invention.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the invention.
Claims (6)
1. A method for nutritional balance assessment of a multi-labeled neural network for children, comprising:
recording diet information of a target object within a certain time period, acquiring motion information of the target object within the certain time period through wearable equipment, preprocessing the diet information and the motion information, and extracting diet characteristics corresponding to the diet information and motion characteristics corresponding to the motion information through a characteristic selection algorithm;
according to the dietary characteristics, combining the movement characteristics with the basic attribute characteristics of the target object, and evaluating whether the dietary characteristics meet the nutritional requirements of the target object through a pre-constructed nutritional evaluation model, wherein the nutritional evaluation model is constructed based on a multi-label neural network model and is used for evaluating whether various input information of an input model meets the preset requirements;
providing a dietary regimen that meets the nutritional requirements of the target subject according to the nutritional requirement deviation and intake balance of the target subject if the dietary characteristics do not meet the nutritional requirements of the target subject;
extracting the diet characteristics corresponding to the diet information and the exercise characteristics corresponding to the exercise information through a characteristic selection algorithm comprises the following steps:
normalizing the diet information and the exercise information, and determining a diet kernel matrix corresponding to the normalized diet information and a exercise kernel matrix corresponding to the normalized exercise information by using a self-adaptive kernel function, wherein the kernel matrix is used for indicating the inner product of the diet information and the exercise information in a high-dimensional space, and the self-adaptive kernel function;
performing decentralization treatment on the diet core matrix and the exercise core matrix respectively to obtain an adaptive diet matrix and an adaptive exercise matrix, performing eigenvalue decomposition on the adaptive diet matrix and the adaptive exercise matrix, and determining diet eigenvalue corresponding to the adaptive diet matrix and exercise eigenvalue corresponding to the adaptive exercise matrix;
sorting according to the sizes of the diet characteristic values and the exercise characteristic values, and taking the characteristics corresponding to the characteristic values, of which the space dimensions of the diet characteristic values and the exercise characteristic values meet preset dimension conditions, as diet characteristics corresponding to the diet information and exercise characteristics corresponding to the exercise information;
according to the dietary characteristics, combining the exercise characteristics and the pre-acquired basic attribute characteristics of the target object, and evaluating whether the dietary characteristics meet the nutritional requirements of the target object through a pre-constructed nutritional evaluation model comprises:
determining a plurality of nutritional components corresponding to the dietary characteristics through the nutritional assessment model;
fusing the basic attribute characteristics and the motion characteristics of the target object to obtain attribute fusion characteristics, and determining energy demand information and nutrition demand information corresponding to the attribute fusion characteristics based on a preset nutrition attribute corresponding relation;
judging whether the multiple nutritional components are matched with the energy demand information and the nutrition demand information, if so, meeting the nutrition requirements of the target object, and if not, not meeting the nutrition requirements of the target object;
the method further comprises training a nutritional assessment model:
acquiring a nutrition evaluation training data set, wherein a sample of the nutrition evaluation training data set comprises a plurality of nutrition components, different nutrition components are regarded as different tasks, and a sample label corresponding to the sample of the nutrition evaluation training data set is determined;
initializing network weight and bias parameters of a nutrition evaluation model to be trained, inputting a nutrition evaluation data set to be trained into the nutrition evaluation model to be trained, determining a model output of the nutrition evaluation model and a prediction deviation of a sample label corresponding to a sample of the nutrition evaluation training data set;
based on the correlation between different nutritional ingredients, and assigning a corresponding nutritional weight for each nutritional ingredient, and assigning a corresponding correlation weight for the correlation between different nutritional ingredients, constructing a loss function in combination with the prediction bias;
according to the loss value of the loss function, determining the gradient of the network weight and the bias parameter of the nutrition evaluation model to be trained on the loss function through a back propagation algorithm, iteratively and automatically adjusting the learning rate and updating the network weight and the bias parameter of the nutrition evaluation model to be trained, so that the loss value of the loss function is minimized.
2. The method of claim 1, wherein constructing a loss function based on the correlation between different nutritional ingredients and assigning a corresponding nutritional weight to each nutritional ingredient and a corresponding correlation weight to the correlation between different nutritional ingredients, in combination with the predictive bias, comprises:
；
wherein, LOSS represents the LOSS value corresponding to the LOSS function, i, j represents the ith nutrient component and the jth nutrient component respectively, N represents the number of nutrient components, wi represents the nutrition weight corresponding to the ith nutrient component, pi, yi represents the model output corresponding to the ith nutrient component and the sample label corresponding to the ith nutrient component respectively, rij represents the correlation weight corresponding to the ith nutrient component and the jth nutrient component, yj represents the sample label corresponding to the jth nutrient component.
3. The method of claim 1, wherein providing a dietary regimen that meets the nutritional requirements of the target subject based on the nutritional requirement bias and intake balance of the target subject if the dietary characteristics do not meet the nutritional requirements of the target subject comprises:
if the dietary characteristics do not meet the nutritional requirements of the target object, determining an ideal intake of each nutritional component meeting the nutritional requirements of the target object according to the basic attribute characteristics of the target object and the nutritional requirements of the target object, determining an actual intake by combining the dietary information of the target object, and determining a nutritional demand deviation according to the ideal intake and the actual intake;
determining an average intake of the nutritional components of the target subject based on the dietary information, determining intake balance in combination with actual intake;
constructing an fitness function based on the nutrition demand deviation and the intake balance, sorting according to fitness values corresponding to the fitness function, and selecting an individual with the fitness value larger than a preset sorting threshold as a parent for breeding the next generation; performing cross operation on the selected parent individuals to generate offspring individuals, performing mutation operation on the crossed individuals, introducing random factors, and generating new offspring individuals;
determining a difference between the parent individual and the offspring individual,
if the difference value is larger than a preset reference threshold value, replacing the child individuals with the parent individuals;
and if the difference value is smaller than a preset reference threshold value, reserving the child individuals in combination with a preset acceptance threshold value, and taking the final reserved child individuals as a diet scheme meeting the nutritional requirements of the target object.
4. A nutrition balance assessment system for a multi-labeled neural network for children, for implementing the nutrition balance assessment method for a multi-labeled neural network for children as claimed in any one of the preceding claims 1 to 3, characterized by comprising:
the first unit is used for recording diet information of a target object in a certain time period and acquiring motion information of the target object in the certain time period through a wearable device, and extracting diet characteristics corresponding to the diet information and motion characteristics corresponding to the motion information through a feature selection algorithm after carrying out data preprocessing on the diet information and the motion information;
the second unit is used for evaluating whether the dietary characteristics meet the nutritional requirements of the target object through a pre-constructed nutritional evaluation model according to the dietary characteristics and combining the movement characteristics and the pre-acquired basic attribute characteristics of the target object, wherein the nutritional evaluation model is constructed based on a multi-label neural network model and is used for evaluating whether various input information of an input model meets the preset requirements;
and a third unit for providing a dietary regimen that meets the nutritional requirements of the target subject according to the nutritional requirement deviation and intake balance of the target subject if the dietary characteristics do not meet the nutritional requirements of the target subject.
5. An electronic device, comprising:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to invoke the instructions stored in the memory to perform the method of any of claims 1 to 3.
6. A computer readable storage medium having stored thereon computer program instructions, which when executed by a processor, implement the method of any of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311396009.4A CN117133407B (en) | 2023-10-26 | 2023-10-26 | Nutritional balance assessment method and system for multi-label neural network for children |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311396009.4A CN117133407B (en) | 2023-10-26 | 2023-10-26 | Nutritional balance assessment method and system for multi-label neural network for children |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117133407A CN117133407A (en) | 2023-11-28 |
| CN117133407B true CN117133407B (en) | 2024-02-13 |
Family
ID=88854943
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311396009.4A Active CN117133407B (en) | 2023-10-26 | 2023-10-26 | Nutritional balance assessment method and system for multi-label neural network for children |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117133407B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117393109B (en) * | 2023-12-11 | 2024-03-22 | 亿慧云智能科技（深圳）股份有限公司 | Scene-adaptive diet monitoring method, device, equipment and storage medium |
| CN117594195B (en) * | 2024-01-18 | 2024-04-09 | 吉林大学 | Kidney disease patient diet analysis system and method based on big data |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111564200A (en) * | 2020-05-08 | 2020-08-21 | 深圳市万佳安人工智能数据技术有限公司 | Old people diet feature extraction device and method based on rapid random gradient descent |
| CN111916179A (en) * | 2019-05-08 | 2020-11-10 | 北京明熹一品电子商务有限公司 | Method for carrying out 'customized' diet nourishing model based on artificial intelligence self-adaption individual physical sign |
| CN115458123A (en) * | 2022-08-24 | 2022-12-09 | 南京励翱科技有限公司 | Personalized nutrition intake analysis method based on multi-source data analysis |
| CN116417115A (en) * | 2023-06-07 | 2023-07-11 | 北京四海汇智科技有限公司 | Personalized nutrition scheme recommendation method and system for gestational diabetes patients |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170249445A1 (en) * | 2014-09-12 | 2017-08-31 | Blacktree Fitness Technologies Inc. | Portable devices and methods for measuring nutritional intake |
-
2023
- 2023-10-26 CN CN202311396009.4A patent/CN117133407B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111916179A (en) * | 2019-05-08 | 2020-11-10 | 北京明熹一品电子商务有限公司 | Method for carrying out 'customized' diet nourishing model based on artificial intelligence self-adaption individual physical sign |
| CN111564200A (en) * | 2020-05-08 | 2020-08-21 | 深圳市万佳安人工智能数据技术有限公司 | Old people diet feature extraction device and method based on rapid random gradient descent |
| CN115458123A (en) * | 2022-08-24 | 2022-12-09 | 南京励翱科技有限公司 | Personalized nutrition intake analysis method based on multi-source data analysis |
| CN116417115A (en) * | 2023-06-07 | 2023-07-11 | 北京四海汇智科技有限公司 | Personalized nutrition scheme recommendation method and system for gestational diabetes patients |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117133407A (en) | 2023-11-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117133407B (en) | Nutritional balance assessment method and system for multi-label neural network for children | |
| Van Erp et al. | Shrinkage priors for Bayesian penalized regression | |
| Baldeon-Calisto et al. | AdaResU-Net: Multiobjective adaptive convolutional neural network for medical image segmentation | |
| CN116417115B (en) | Personalized nutrition scheme recommendation method and system for gestational diabetes patients | |
| CN117198466B (en) | Diet management method and system for kidney disease patients | |
| Towner et al. | Alternative statistical approaches to the use of data as evidence for hypotheses in human behavioral ecology | |
| Marbac et al. | Model-based clustering for conditionally correlated categorical data | |
| CN109583272B (en) | Footprint system capable of acquiring living state of human body | |
| Mishra et al. | Analysis of Indian Food Based on Machine learning Classification Models | |
| Bai et al. | Recommendation algorithm based on probabilistic matrix factorization with adaboost | |
| Ferguson et al. | Incorporating parameter estimability into model selection | |
| Kirchler et al. | Training normalizing flows from dependent data | |
| Jillani et al. | Efficient diagnosis of liver disease using deep learning technique | |
| Mbaluka et al. | Application of principal component analysis and hierarchical regression model on Kenya macroeconomic indicators | |
| Najaflou et al. | Recommended system for controlling malnutrition in Iranian children 6 to 12 years old using machine learning algorithms | |
| US20240087719A1 (en) | Methods and systems for calculating an edible score in a display interface | |
| CN116580817B (en) | Personalized nutrition guidance method and system based on machine learning | |
| Dindorf et al. | Enhancing biomechanical machine learning with limited data: generating realistic synthetic posture data using generative artificial intelligence | |
| Holloway et al. | Hierarchical analysis of production efficiency in a coastal trawl fishery | |
| US11862322B2 (en) | System and method for generating a dynamic weighted combination | |
| Kour et al. | An Advance Approach for Diabetes Detection by Implementing Machine Learning Algorithms | |
| US11688507B2 (en) | Systems and methods for generating a metabolic dysfunction nourishment program | |
| Bhushan et al. | Prediction Of Health Status Using Machine Learning Techniques | |
| Mbaluka et al. | Application of Principal Component Analysis and Hierarchical Regression Model on Kenya Macroeconomic Indicators | |
| Gür | Innovation in the dairy industry: forecasting cow cheese production with machine learning and deep learning models |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |