CN117976148A - Machine learning-based prediction method and system for mental health problems of children - Google Patents
Machine learning-based prediction method and system for mental health problems of children Download PDFInfo
- Publication number
- CN117976148A CN117976148A CN202410385921.8A CN202410385921A CN117976148A CN 117976148 A CN117976148 A CN 117976148A CN 202410385921 A CN202410385921 A CN 202410385921A CN 117976148 A CN117976148 A CN 117976148A
- Authority
- CN
- China
- Prior art keywords
- data
- child
- psychological
- mental
- children
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 72
- 238000010801 machine learning Methods 0.000 title claims abstract description 21
- 230000004630 mental health Effects 0.000 title claims description 66
- 230000009323 psychological health Effects 0.000 claims abstract description 82
- 238000012706 support-vector machine Methods 0.000 claims abstract description 25
- 230000002159 abnormal effect Effects 0.000 claims abstract description 15
- 238000005070 sampling Methods 0.000 claims abstract description 10
- 230000003340 mental effect Effects 0.000 claims description 65
- 238000004140 cleaning Methods 0.000 claims description 48
- 238000004364 calculation method Methods 0.000 claims description 31
- 238000000605 extraction Methods 0.000 claims description 30
- 230000006870 function Effects 0.000 claims description 30
- 239000013598 vector Substances 0.000 claims description 27
- 238000012549 training Methods 0.000 claims description 24
- 230000008569 process Effects 0.000 claims description 22
- 238000007781 pre-processing Methods 0.000 claims description 18
- 238000010276 construction Methods 0.000 claims description 17
- 238000012417 linear regression Methods 0.000 claims description 9
- 238000005406 washing Methods 0.000 claims description 6
- 230000003044 adaptive effect Effects 0.000 claims description 3
- 238000004422 calculation algorithm Methods 0.000 claims description 3
- 238000012217 deletion Methods 0.000 claims description 3
- 230000037430 deletion Effects 0.000 claims description 3
- 230000006996 mental state Effects 0.000 claims description 3
- 230000015572 biosynthetic process Effects 0.000 claims 1
- 238000003786 synthesis reaction Methods 0.000 claims 1
- 230000006872 improvement Effects 0.000 description 5
- 230000002411 adverse Effects 0.000 description 3
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 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
- G16H20/70—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to mental therapies, e.g. psychological therapy or autogenous training
-
- 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
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- 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
- G16H50/50—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for simulation or modelling of medical disorders
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Epidemiology (AREA)
- Biomedical Technology (AREA)
- Primary Health Care (AREA)
- General Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Data Mining & Analysis (AREA)
- Child & Adolescent Psychology (AREA)
- Developmental Disabilities (AREA)
- Hospice & Palliative Care (AREA)
- Psychiatry (AREA)
- Psychology (AREA)
- Social Psychology (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses a machine learning-based method and system for predicting psychological health problems of children, and belongs to the technical field of psychological health prediction. According to the invention, a self-adaptive synthetic sampling method is adopted, and the data diversity is increased through a synthetic sample, so that a data set is balanced, and the prediction capability of a model on a psychological abnormal state is improved; the automatic encoder is adopted to extract the characteristics, so that the most representative characteristics can be automatically learned, potential characteristics can be found, and the model prediction performance is improved; and a psychological health prediction model based on a support vector machine is adopted to construct, and a decision boundary with strong adaptability is established by optimizing parameters, so that the accuracy of model prediction is improved.
Description
Technical Field
The invention belongs to the technical field of psychological health prediction, and particularly relates to a machine learning-based method and system for predicting psychological health problems of children.
Background
The prediction of the mental health problems of the children refers to the process of predicting the mental health problems possibly existing in the children by utilizing a statistical analysis and machine learning method, and aims to discover the mental health problems as soon as possible, thereby providing early intervention and support, helping to lighten the severity of the mental health problems, providing better treatment effect for the children and promoting the mental health development of the children.
However, in the existing prediction process of the psychological health problem of children, the technical problem that the normal psychological state occupies most of data, so that the data set is unbalanced and the model performance is adversely affected exists; the method has the technical problems that the method relates to various different types of data, the data has complexity and diversity, and a method for efficiently extracting the data features is lacked; the model prediction method has the technical problems that the model prediction accuracy is low due to the fact that the psychological health problem of children is complex and is influenced by various factors.
Disclosure of Invention
Aiming at the problems that the data set is unbalanced and the model performance is adversely affected due to the fact that the normal psychological state occupies most of data in the prediction process of the psychological health problem of the children, the self-adaptive synthetic sampling method is adopted, the data diversity is increased through a synthetic sample, the data set is balanced, and the prediction capability of the model on the psychological abnormal state is improved; aiming at the technical problems that in the prediction process of the psychological health problems of children, various different types of data are involved, the data have complexity and diversity, and a method for efficiently extracting data features is lacking, the scheme adopts an automatic encoder to extract features, so that the most representative features can be automatically learned, potential features can be found, and the prediction performance of a model is improved; aiming at the technical problem that the model prediction accuracy is low due to the fact that the psychological health problem of the children is complex and influenced by various factors in the psychological health problem prediction process of the children, the method is constructed by adopting the psychological health prediction model based on the support vector machine, and a decision boundary with strong adaptability is established by optimizing parameters, so that the model prediction accuracy is improved.
The technical scheme adopted by the invention is as follows: the invention provides a machine learning-based prediction method for psychological health problems of children, which comprises the following steps:
Step S1: data acquisition, namely acquiring psychological health Data 1 of the children;
Step S2: the Data preprocessing, specifically, data cleaning is performed to obtain child mental cleaning Data 2, missing value filling is performed on the child mental cleaning Data 2 to obtain child mental standard Data 3, and a self-adaptive synthetic sampling method is adopted to perform Data enhancement on the child mental standard Data 3 to obtain child balance mental Data 4;
Step S3: the feature extraction is specifically to optimize an automatic encoder parameter omega g and a decoder parameter omega h by minimizing a reconstruction error, calculate to obtain a trained encoder, and perform feature extraction on the child balance psychological Data 4 through the trained encoder, calculate to obtain the child psychological feature Fea;
step S4: the psychological health prediction Model is constructed, specifically, a support vector machine Model is initialized through a classification hyperplane Model to obtain an initial Model 0, the parameters of the support vector machine Model are optimized through a Lagrangian function, and the initial Model 0 is trained to obtain a psychological health prediction Model 1;
step S5: and predicting psychological health problems of children.
As a further improvement of the present solution, in step S1, the child mental health Data 1 includes child personal information, mental health status and mental condition labels, the child personal information is obtained from the medical system, the mental health status is obtained by filling in advantages and difficult questionnaires by parents, and the mental condition labels including a mental normal state and a mental abnormal state are obtained by statistically analyzing the mental health status.
As a further improvement of the present solution, in step S2, the data preprocessing includes the steps of:
Step S21: the Data cleaning method comprises the steps of carrying out Data cleaning on child mental health Data 1, wherein the Data cleaning comprises weight removal, Z score abnormal value deletion and standardization, and the child mental cleaning Data 2 is obtained through the Data cleaning;
step S22: filling of the missing value comprises the following steps:
step S221: the mental cleaning Data 2 of the child is obtained, and the calculation formula is as follows:
；
wherein, data 2 is child mental cleaning Data, da 1 is a1 st child mental cleaning Data sample, n is the number of child mental cleaning Data 2 samples, and the child mental cleaning Data samples comprise child information variables, mental state variables and mental label variables;
step S223: the missing value is initially filled, and particularly, the missing value of the child mental cleaning Data 2 is initially filled through mode;
step S224: adopting a cross section interpolation algorithm to fill in missing values through iteration, comprising the following steps:
step S2241: defining regression parameter vectors, wherein the calculation formula is as follows:
；
Where gi is a regression parameter vector, ai 1 is a1 st child psychology washing data sample variable, ai n is an nth child psychology washing data sample variable, and T is a transpose operation;
step S2242: and (3) estimating the missing value by establishing a linear regression model through regression parameter vectors, wherein the calculation formula is as follows:
；
Wherein y a is a missing value variable, h () is a linear regression function, y b is other variables than the missing value variable y a, a is a missing value variable index, and b is other variable indexes than the missing value variable y a;
Step S2243: filling the missing values of the child psychological cleaning Data 2 one by one through a linear regression model, and calculating to obtain child psychological standard Data 3;
step S23: the data enhancement is carried out by adopting an adaptive synthetic sampling method, which comprises the following steps:
step S231: the psychological standard Data 3 of the children is obtained, and the calculation formula is as follows:
；
wherein, data 3 is the psychological standard Data of children, at 1 is the psychological standard Data sample of the 1 st child, Is the number of the psychological standard data samples of the children;
Step S232: selecting a child psychological standard data sample in a psychological abnormal state, and obtaining k nearest neighbor samples by calculating Euclidean distances between the selected child psychological standard data sample and the child psychological standard data samples in other psychological abnormal states ；
Step S233: and calculating a synthetic sample, wherein the calculation formula is as follows:
；
Where ea u is a synthetic sample, at u is a current child mental standard data sample, Is a random nearest neighbor sample, u is a sample index of current psychological standard data of children, v is a sample index of the random nearest neighbor, and alpha is a random parameter with a value range of [0,1 ];
Step S234: and repeating the S232 and subsequent operations for X times to obtain X synthesized samples, and merging all the synthesized samples into the child mental standard Data 3 to obtain the child balance mental Data 4.
As a further improvement of the present solution, in step S3, the feature extraction, specifically, the feature extraction performed by using an automatic encoder, includes the following steps:
Step S31: the hidden layer coding is calculated through an automatic encoder function, and the calculation formula is as follows:
；
Where f is the hidden layer encoding, g () is the auto encoder function, z is the child balance psychology data sample, sigmoid () is the Sigmoid function, M g is the auto encoder weight, c g is the auto encoder bias vector, ω g is the auto encoder parameters including auto encoder weight M g and auto encoder bias vector c g;
step S32: reconstructing a hidden layer code by a decoder function, wherein the calculation formula is as follows:
；
in the method, in the process of the invention, Is a child psycho-reconstruction sample, h () is a decoder function, M h is a decoder weight, c h is a decoder bias vector, ω h is a decoder parameter, which includes a decoder weight M h and a decoder bias vector c h;
step S33: according to the psychological reconstruction sample of the child, calculating a reconstruction error, wherein the calculation formula is as follows:
；
in the method, in the process of the invention, Is the reconstruction error, m is the number of samples of child balance psychological data, j is the index of child balance psychological data samples and child psychological reconstruction samples, z j is the j-th child balance psychological data sample,/>Is the j-th child heart reconstruction sample;
Step S34: training an automatic encoder, in particular optimizing an automatic encoder parameter omega g and a decoder parameter omega h by minimizing a reconstruction error, and calculating to obtain a trained encoder;
Step S35: and carrying out feature extraction on the child balance psychological Data 4 through the trained encoder, and calculating to obtain the child psychological features Fea.
As a further improvement of the present solution, in step S4, the psychological health prediction model construction, specifically, the psychological health prediction model construction based on the support vector machine, includes the following steps:
step S41: initializing a support vector machine model, comprising the following steps:
Step S411: constructing a training set, and particularly dividing a training set Q from the psychological characteristics Fea of the children;
Step S412: establishing a classified hyperplane model, wherein the calculation formula is as follows:
；
Wherein, p is a classification hyperplane normal vector, q r is a training set sample, r is a training set sample index, and s is a classification hyperplane offset;
Step S413: initializing a support vector machine Model by classifying the hyperplane Model to obtain an initial Model 0;
Step S42: the Lagrangian function is constructed, and the calculation formula is as follows:
；
in the method, in the process of the invention, Is a Lagrangian function,/>Is the euclidean norm of the classification hyperplane normal vector, w is the training set sample number, mu r is the relaxation parameter, lab r is the training set sample label;
Step S43: and optimizing Model parameters of the support vector machine through a Lagrangian function, and training an initial Model 0 to obtain a mental health prediction Model 1.
As a further improvement of the present solution, in step S5, the prediction of the mental health problem of the child, specifically, the prediction is performed by using the mental health prediction Model 1, so as to classify the mental health of the child and generate a mental health report of the child.
The invention provides a machine learning-based children mental health problem prediction system, which comprises: the system comprises a data acquisition module, a data preprocessing module, a characteristic extraction module, a psychological health prediction model construction module and a psychological health problem prediction module for children;
The data acquisition module acquires the mental health data of the children and sends the mental health data of the children to the data preprocessing module;
the data preprocessing module receives the child mental health data from the data acquisition module, performs data cleaning, missing value filling and data enhancement operation on the child mental health data to obtain child balance mental data, and sends the child balance mental data to the feature extraction module;
The feature extraction module receives the child balance psychological data from the data preprocessing module, performs feature extraction on the child balance psychological data by adopting an automatic encoder to obtain child psychological features, and sends the child psychological features to the psychological health prediction model construction module;
The psychological health prediction model building module receives the psychological characteristics of the children from the characteristic extraction module, acquires data from the psychological characteristics of the children, trains through a support vector machine model to obtain a psychological health prediction model, and sends the psychological health prediction model to the psychological health problem prediction module of the children;
The psychological health problem prediction module receives the psychological health prediction model from the psychological health prediction model construction module, predicts the psychological health problem of the child through the psychological health prediction model, classifies the psychological health of the child and generates a psychological health report of the child.
By adopting the scheme, the beneficial effects obtained by the invention are as follows:
(1) Aiming at the technical problems that in the prediction process of the psychological health problem of children, the data set is unbalanced due to the fact that the normal psychological state occupies most of the data, and bad influence is generated on the model performance, the data diversity is increased through a synthetic sample by adopting the self-adaptive synthetic sampling method, so that the data set is balanced, and the prediction capability of the model on the psychological abnormal state is improved.
(2) Aiming at the technical problems that in the prediction process of the psychological health problems of children, various data of different types are involved, the data has complexity and diversity, and a method for efficiently extracting the data features is lacked, the scheme adopts an automatic encoder to extract the features, so that the most representative features can be automatically learned, potential features can be found, and the prediction performance of a model is improved.
(3) Aiming at the technical problem that the model prediction accuracy is low due to the fact that the psychological health problem of the children is complex and influenced by various factors in the psychological health problem prediction process of the children, the method is constructed by adopting the psychological health prediction model based on the support vector machine, and a decision boundary with strong adaptability is established by optimizing parameters, so that the model prediction accuracy is improved.
Drawings
FIG. 1 is a schematic flow chart of a machine learning-based method for predicting mental health problems of children;
FIG. 2 is a schematic diagram of a system for predicting mental health problems of children based on machine learning according to the present invention;
FIG. 3 is a flow chart of step S2;
FIG. 4 is a flow chart of step S3;
fig. 5 is a flow chart of step S4.
The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the invention; 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.
Referring to fig. 1, the method for predicting mental health problems of children based on machine learning provided by the invention comprises the following steps:
Step S1: data acquisition, namely acquiring psychological health Data 1 of the children;
Step S2: the Data preprocessing, specifically, data cleaning is performed to obtain child mental cleaning Data 2, missing value filling is performed on the child mental cleaning Data 2 to obtain child mental standard Data 3, and a self-adaptive synthetic sampling method is adopted to perform Data enhancement on the child mental standard Data 3 to obtain child balance mental Data 4;
Step S3: the feature extraction is specifically to optimize an automatic encoder parameter omega g and a decoder parameter omega h by minimizing a reconstruction error, calculate to obtain a trained encoder, and perform feature extraction on the child balance psychological Data 4 through the trained encoder, calculate to obtain the child psychological feature Fea;
step S4: the psychological health prediction Model is constructed, specifically, a support vector machine Model is initialized through a classification hyperplane Model to obtain an initial Model 0, the parameters of the support vector machine Model are optimized through a Lagrangian function, and the initial Model 0 is trained to obtain a psychological health prediction Model 1;
step S5: and predicting psychological health problems of children.
In a second embodiment, referring to fig. 1, in step S1, the mental health Data 1 includes personal information, mental health status and mental condition labels of the child, the personal information of the child is obtained from the medical system, the mental health status is obtained by filling in advantages and difficult questionnaires by parents, and the mental health status labels including a mental normal state and a mental abnormal state are obtained by statistically analyzing the mental health status.
Embodiment three, referring to fig. 1 and 3, the embodiment is based on the above embodiment, further, in step S2, the data preprocessing includes the following steps:
Step S21: the Data cleaning method comprises the steps of carrying out Data cleaning on child mental health Data 1, wherein the Data cleaning comprises weight removal, Z score abnormal value deletion and standardization, and the child mental cleaning Data 2 is obtained through the Data cleaning;
step S22: filling of the missing value comprises the following steps:
step S221: the mental cleaning Data 2 of the child is obtained, and the calculation formula is as follows:
；
wherein, data 2 is child mental cleaning Data, da 1 is a1 st child mental cleaning Data sample, n is the number of child mental cleaning Data 2 samples, and the child mental cleaning Data samples comprise child information variables, mental state variables and mental label variables;
step S223: the missing value is initially filled, and particularly, the missing value of the child mental cleaning Data 2 is initially filled through mode;
step S224: adopting a cross section interpolation algorithm to fill in missing values through iteration, comprising the following steps:
step S2241: defining regression parameter vectors, wherein the calculation formula is as follows:
；
Where gi is a regression parameter vector, ai 1 is a1 st child psychology washing data sample variable, ai n is an nth child psychology washing data sample variable, and T is a transpose operation;
step S2242: and (3) estimating the missing value by establishing a linear regression model through regression parameter vectors, wherein the calculation formula is as follows:
；
Wherein y a is a missing value variable, h () is a linear regression function, y b is other variables than the missing value variable y a, a is a missing value variable index, and b is other variable indexes than the missing value variable y a;
Step S2243: filling the missing values of the child psychological cleaning Data 2 one by one through a linear regression model, and calculating to obtain child psychological standard Data 3;
step S23: the data enhancement is carried out by adopting an adaptive synthetic sampling method, which comprises the following steps:
step S231: the psychological standard Data 3 of the children is obtained, and the calculation formula is as follows:
；
wherein, data 3 is the psychological standard Data of children, at 1 is the psychological standard Data sample of the 1 st child, Is the number of the psychological standard data samples of the children;
Step S232: selecting a child psychological standard data sample in a psychological abnormal state, and obtaining k nearest neighbor samples by calculating Euclidean distances between the selected child psychological standard data sample and the child psychological standard data samples in other psychological abnormal states ；
Step S233: according to the random nearest-neck sample, calculating a synthesized sample, wherein the calculation formula is as follows:
；
Where ea u is a synthetic sample, at u is a current child mental standard data sample, Is a random nearest neighbor sample, u is a sample index of current psychological standard data of children, v is a sample index of the random nearest neighbor, and alpha is a random parameter with a value range of [0,1 ];
Step S234: repeating S232 and the subsequent operation for X times to obtain X synthetic samples, and merging all the synthetic samples into the child mental standard Data 3 to obtain child balance mental Data 4;
By executing the above operation, the technical problems that the data set is unbalanced and the model performance is adversely affected due to the fact that the normal psychological states occupy most of the data in the prediction process of the psychological health problems of the children are solved.
In a fourth embodiment, referring to fig. 1 and fig. 4, the embodiment is further based on the above embodiment, and in step S3, the feature extraction, specifically, the feature extraction performed by using an automatic encoder, includes the following steps:
Step S31: the hidden layer coding is calculated through an automatic encoder function, and the calculation formula is as follows:
；
Where f is the hidden layer encoding, g () is the auto encoder function, z is the child balance psychology data sample, sigmoid () is the Sigmoid function, M g is the auto encoder weight, c g is the auto encoder bias vector, ω g is the auto encoder parameters including auto encoder weight M g and auto encoder bias vector c g;
step S32: reconstructing a hidden layer code by a decoder function, wherein the calculation formula is as follows:
；
in the method, in the process of the invention, Is a child psycho-reconstruction sample, h () is a decoder function, M h is a decoder weight, c h is a decoder bias vector, ω h is a decoder parameter, which includes a decoder weight M h and a decoder bias vector c h;
step S33: according to the psychological reconstruction sample of the child, calculating a reconstruction error, wherein the calculation formula is as follows:
；
in the method, in the process of the invention, Is the reconstruction error, m is the number of samples of child balance psychological data, j is the index of child balance psychological data samples and child psychological reconstruction samples, z j is the j-th child balance psychological data sample,/>Is the j-th child heart reconstruction sample;
Step S34: training an automatic encoder, in particular optimizing an automatic encoder parameter omega g and a decoder parameter omega h by minimizing a reconstruction error, and calculating to obtain a trained encoder;
step S35: feature extraction is carried out on the child balance psychological Data 4 through a trained encoder, and the child psychological feature Fea is obtained through calculation;
By executing the above operation, aiming at the technical problems that in the prediction process of the psychological health problems of children, various data of different types are involved, the data has complexity and diversity, and a method for efficiently extracting the data features is lacked, the scheme adopts an automatic encoder to extract the features, so that the most representative features can be automatically learned, the potential features can be found, and the prediction performance of the model is improved.
An embodiment five, referring to fig. 1 and 5, is based on the foregoing embodiment, further in step S4, the mental health prediction model construction, specifically, the mental health prediction model construction based on a support vector machine, includes the following steps:
step S41: initializing a support vector machine model, comprising the following steps:
Step S411: constructing a training set, and particularly dividing a training set Q from the psychological characteristics Fea of the children;
Step S412: establishing a classified hyperplane model, wherein the calculation formula is as follows:
；
Wherein, p is a classification hyperplane normal vector, q r is a training set sample, r is a training set sample index, and s is a classification hyperplane offset;
Step S413: initializing a support vector machine Model by classifying the hyperplane Model to obtain an initial Model 0;
step S42: the Lagrangian function is constructed and used for optimizing model parameters, and the calculation formula is as follows:
；
in the method, in the process of the invention, Is a Lagrangian function,/>Is the euclidean norm of the classification hyperplane normal vector, w is the training set sample number, mu r is the relaxation parameter, lab r is the training set sample label;
Step S43: optimizing Model parameters of a support vector machine through a Lagrangian function, and training an initial Model 0 to obtain a mental health prediction Model 1;
By executing the above operation, aiming at the technical problem that the model prediction accuracy is low because the psychological health problem of the children is complex and is influenced by various factors in the psychological health problem prediction process of the children, the method adopts the psychological health prediction model construction based on the support vector machine, and establishes a decision boundary with strong adaptability by optimizing parameters, so that the model prediction accuracy is improved.
In a sixth embodiment, referring to fig. 1, the embodiment is based on the foregoing embodiment, further, in step S5, the prediction of the mental health problem of the child, specifically, the prediction is performed by using a mental health prediction Model 1, so as to classify the mental health of the child and generate a mental health report of the child.
Embodiment seven, referring to fig. 2, based on the above embodiment, the system for predicting a mental health problem of a child based on machine learning provided by the present invention includes: the system comprises a data acquisition module, a data preprocessing module, a characteristic extraction module, a psychological health prediction model construction module and a psychological health problem prediction module for children;
The data acquisition module acquires the mental health data of the children and sends the mental health data of the children to the data preprocessing module;
the data preprocessing module receives the child mental health data from the data acquisition module, performs data cleaning, missing value filling and data enhancement operation on the child mental health data to obtain child balance mental data, and sends the child balance mental data to the feature extraction module;
The feature extraction module receives the child balance psychological data from the data preprocessing module, performs feature extraction on the child balance psychological data by adopting an automatic encoder to obtain child psychological features, and sends the child psychological features to the psychological health prediction model construction module;
The psychological health prediction model building module receives the psychological characteristics of the children from the characteristic extraction module, acquires data from the psychological characteristics of the children, trains through a support vector machine model to obtain a psychological health prediction model, and sends the psychological health prediction model to the psychological health problem prediction module of the children;
The psychological health problem prediction module receives the psychological health prediction model from the psychological health prediction model construction module, predicts the psychological health problem of the child through the psychological health prediction model, classifies the psychological health of the child and generates a psychological health report of the child.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
The invention and its embodiments have been described above with no limitation, and the actual construction is not limited to the embodiments of the invention as shown in the drawings. In summary, if one of ordinary skill in the art is informed by this disclosure, a structural manner and an embodiment similar to the technical solution should not be creatively devised without departing from the gist of the present invention.
Claims (9)
1. A machine learning-based prediction method for psychological health problems of children is characterized in that: the method comprises the following steps:
Step S1: data acquisition, namely acquiring psychological health Data 1 of the children;
Step S2: the Data preprocessing, specifically, data cleaning is performed to obtain child mental cleaning Data 2, missing value filling is performed on the child mental cleaning Data 2 to obtain child mental standard Data 3, and a self-adaptive synthetic sampling method is adopted to perform Data enhancement on the child mental standard Data 3 to obtain child balance mental Data 4;
Step S3: the feature extraction is specifically to optimize an automatic encoder parameter omega g and a decoder parameter omega h by minimizing a reconstruction error, calculate to obtain a trained encoder, and perform feature extraction on the child balance psychological Data 4 through the trained encoder, calculate to obtain the child psychological feature Fea;
step S4: the psychological health prediction Model is constructed, specifically, a support vector machine Model is initialized through a classification hyperplane Model to obtain an initial Model 0, the parameters of the support vector machine Model are optimized through a Lagrangian function, and the initial Model 0 is trained to obtain a psychological health prediction Model 1;
step S5: and predicting psychological health problems of children.
2. The machine learning based method for predicting mental health problems in children of claim 1, wherein: in step S2, the data preprocessing includes the following steps:
Step S21: the Data cleaning method comprises the steps of carrying out Data cleaning on child mental health Data 1, wherein the Data cleaning comprises weight removal, Z score abnormal value deletion and standardization, and the child mental cleaning Data 2 is obtained through the Data cleaning;
step S22: filling of the missing value comprises the following steps:
step S221: the mental cleaning Data 2 of the child is obtained, and the calculation formula is as follows:
；
wherein, data 2 is child mental cleaning Data, da 1 is a1 st child mental cleaning Data sample, n is the number of child mental cleaning Data 2 samples, and the child mental cleaning Data samples comprise child information variables, mental state variables and mental label variables;
step S223: the missing value is initially filled, and particularly, the missing value of the child mental cleaning Data 2 is initially filled through mode;
step S224: adopting a cross section interpolation algorithm to fill in missing values through iteration, comprising the following steps:
step S2241: defining regression parameter vectors, wherein the calculation formula is as follows:
；
Where gi is a regression parameter vector, ai 1 is a1 st child psychology washing data sample variable, ai n is an nth child psychology washing data sample variable, and T is a transpose operation;
step S2242: and (3) estimating the missing value by establishing a linear regression model through regression parameter vectors, wherein the calculation formula is as follows:
；
Wherein y a is a missing value variable, h () is a linear regression function, y b is other variables than the missing value variable y a, a is a missing value variable index, and b is other variable indexes than the missing value variable y a;
Step S2243: filling the missing values of the child psychological cleaning Data 2 one by one through a linear regression model, and calculating to obtain child psychological standard Data 3;
step S23: and adopting a self-adaptive synthetic sampling method to enhance the data.
3. A machine learning based method of predicting mental health problems in children as claimed in claim 2, wherein: in step S23, the data enhancement is performed by adopting an adaptive synthesis sampling method, which includes the following steps:
step S231: the psychological standard Data 3 of the children is obtained, and the calculation formula is as follows:
；
wherein, data 3 is the psychological standard Data of children, at 1 is the psychological standard Data sample of the 1 st child, Is the number of the psychological standard data samples of the children;
Step S232: selecting a child psychological standard data sample in a psychological abnormal state, and obtaining k nearest neighbor samples by calculating Euclidean distances between the selected child psychological standard data sample and the child psychological standard data samples in other psychological abnormal states ；
Step S233: and calculating a synthetic sample, wherein the calculation formula is as follows:
；
Where ea u is a synthetic sample, at u is a current child mental standard data sample, Is a random nearest neighbor sample, u is a sample index of current psychological standard data of children, v is a sample index of the random nearest neighbor, and alpha is a random parameter with a value range of [0,1 ];
Step S234: and repeating the S232 and subsequent operations for X times to obtain X synthesized samples, and merging all the synthesized samples into the child mental standard Data 3 to obtain the child balance mental Data 4.
4. A machine learning based method of predicting mental health problems in children as claimed in claim 3, wherein: in step S3, the feature extraction, specifically, feature extraction performed by using an automatic encoder, includes the following steps:
Step S31: the hidden layer coding is calculated through an automatic encoder function, and the calculation formula is as follows:
；
Where f is the hidden layer encoding, g () is the auto encoder function, z is the child balance psychology data sample, sigmoid () is the Sigmoid function, M g is the auto encoder weight, c g is the auto encoder bias vector, ω g is the auto encoder parameters including auto encoder weight M g and auto encoder bias vector c g;
step S32: reconstructing a hidden layer code by a decoder function, wherein the calculation formula is as follows:
；
in the method, in the process of the invention, Is a child psycho-reconstruction sample, h () is a decoder function, M h is a decoder weight, c h is a decoder bias vector, ω h is a decoder parameter, which includes a decoder weight M h and a decoder bias vector c h;
step S33: according to the psychological reconstruction sample of the child, calculating a reconstruction error, wherein the calculation formula is as follows:
；
in the method, in the process of the invention, Is the reconstruction error, m is the number of samples of child balance psychological data, j is the index of child balance psychological data samples and child psychological reconstruction samples, z j is the j-th child balance psychological data sample,/>Is the j-th child heart reconstruction sample;
Step S34: training an automatic encoder, in particular optimizing an automatic encoder parameter omega g and a decoder parameter omega h by minimizing a reconstruction error, and calculating to obtain a trained encoder;
Step S35: and carrying out feature extraction on the child balance psychological Data 4 through the trained encoder, and calculating to obtain the child psychological features Fea.
5. The machine learning based method for predicting mental health problems in children of claim 4, wherein: in step S4, the psychological health prediction model construction, specifically, the psychological health prediction model construction based on the support vector machine, includes the following steps:
step S41: initializing a support vector machine model, comprising the following steps:
Step S411: constructing a training set, and particularly dividing a training set Q from the psychological characteristics Fea of the children;
Step S412: establishing a classified hyperplane model, wherein the calculation formula is as follows:
；
Wherein, p is a classification hyperplane normal vector, q r is a training set sample, r is a training set sample index, and s is a classification hyperplane offset;
Step S413: initializing a support vector machine Model by classifying the hyperplane Model to obtain an initial Model 0;
Step S42: the Lagrangian function is constructed, and the calculation formula is as follows:
；
in the method, in the process of the invention, Is a Lagrangian function,/>Is the euclidean norm of the classification hyperplane normal vector, w is the training set sample number, mu r is the relaxation parameter, lab r is the training set sample label;
Step S43: and optimizing Model parameters of the support vector machine through a Lagrangian function, and training an initial Model 0 to obtain a mental health prediction Model 1.
6. The machine learning based method for predicting mental health problems in children of claim 5, wherein: in step S1, the child mental health Data 1 includes personal information, mental health status and mental status labels of the child, the personal information of the child is obtained from the medical system, the mental health status is obtained by filling in advantages and difficult questionnaires by parents, and the mental status labels including a mental normal state and a mental abnormal state are obtained by statistically analyzing the mental health status.
7. The machine learning based method for predicting mental health problems in children of claim 6, wherein: in step S5, the prediction of the mental health problem of the child is specifically performed by using a mental health prediction Model 1 to classify the mental health of the child and generate a mental health report of the child.
8. A machine learning-based mental health problem prediction system for implementing a machine learning-based mental health problem prediction method for children as set forth in any one of claims 1 to 7, characterized in that: the system comprises a data acquisition module, a data preprocessing module, a feature extraction module, a psychological health prediction model construction module and a child psychological health problem prediction module.
9. The machine learning based mental health problem prediction system for children of claim 8, wherein: the data acquisition module acquires the mental health data of the children and sends the mental health data of the children to the data preprocessing module;
the data preprocessing module receives the child mental health data from the data acquisition module, performs data cleaning, missing value filling and data enhancement operation on the child mental health data to obtain child balance mental data, and sends the child balance mental data to the feature extraction module;
The feature extraction module receives the child balance psychological data from the data preprocessing module, performs feature extraction on the child balance psychological data by adopting an automatic encoder to obtain child psychological features, and sends the child psychological features to the psychological health prediction model construction module;
The psychological health prediction model building module receives the psychological characteristics of the children from the characteristic extraction module, acquires data from the psychological characteristics of the children, trains through a support vector machine model to obtain a psychological health prediction model, and sends the psychological health prediction model to the psychological health problem prediction module of the children;
The psychological health problem prediction module receives the psychological health prediction model from the psychological health prediction model construction module, predicts the psychological health problem of the child through the psychological health prediction model, classifies the psychological health of the child and generates a psychological health report of the child.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410385921.8A CN117976148B (en) | 2024-04-01 | 2024-04-01 | Machine learning-based prediction method and system for mental health problems of children |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410385921.8A CN117976148B (en) | 2024-04-01 | 2024-04-01 | Machine learning-based prediction method and system for mental health problems of children |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117976148A true CN117976148A (en) | 2024-05-03 |
| CN117976148B CN117976148B (en) | 2024-06-21 |
Family
ID=90863615
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410385921.8A Active CN117976148B (en) | 2024-04-01 | 2024-04-01 | Machine learning-based prediction method and system for mental health problems of children |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117976148B (en) |
Citations (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109192310A (en) * | 2018-07-25 | 2019-01-11 | 同济大学 | A kind of undergraduate psychological behavior unusual fluctuation scheme Design method based on big data |
| CN112669905A (en) * | 2020-12-31 | 2021-04-16 | 中南民族大学 | RNA sequence coding potential prediction method and system based on data enhancement |
| CN113539492A (en) * | 2021-06-16 | 2021-10-22 | 甘肃省卫生健康统计信息中心(西北人口信息中心) | Urban health index prediction system, prediction analysis method and storage medium thereof |
| CN114171150A (en) * | 2021-12-10 | 2022-03-11 | 曲阜师范大学 | Health data missing value prediction method and device, computer equipment and storage medium |
| CN114724701A (en) * | 2022-03-11 | 2022-07-08 | 梁娜 | Noninvasive ventilation curative effect prediction system based on superposition integration algorithm and automatic encoder |
| CN114927230A (en) * | 2022-04-11 | 2022-08-19 | 四川大学华西医院 | Machine learning-based severe heart failure patient prognosis decision support system and method |
| US20220285028A1 (en) * | 2021-03-04 | 2022-09-08 | Abbott Laboratories | Technology to automatically identify the most relevant health failure risk factors |
| CN115036043A (en) * | 2022-05-24 | 2022-09-09 | 珠海清大声光电工程技术研发中心 | Block chain technology-based children mental health monitoring and management system and method thereof |
| CN116013501A (en) * | 2022-12-12 | 2023-04-25 | 烟台新旧动能转换研究院暨烟台科技成果转移转化示范基地 | Multi-mode arrhythmia classification auxiliary diagnosis system based on data expansion |
| KR20230090465A (en) * | 2021-12-15 | 2023-06-22 | 한양대학교 에리카산학협력단 | Apparatus and method for augmentating of data |
| CN116318834A (en) * | 2023-01-16 | 2023-06-23 | 南京信息工程大学 | Internet intrusion detection method and device based on optimized correlation vector machine |
| CN116680613A (en) * | 2023-06-09 | 2023-09-01 | 东北大学 | Human activity recognition comprehensive optimization method based on multi-scale metric learning |
| CN116825348A (en) * | 2023-06-14 | 2023-09-29 | 安徽大学 | University student mental health state assessment model and early warning method based on campus behavior data analysis |
| WO2023231995A1 (en) * | 2022-05-30 | 2023-12-07 | 浙大城市学院 | Transfer-learning-based life prediction and health assessment method for aero-engine |
| CN117421582A (en) * | 2023-11-08 | 2024-01-19 | 浙江正泰中自控制工程有限公司 | Equipment health analysis method based on multi-source data driving |
| CN117522222A (en) * | 2023-11-27 | 2024-02-06 | 深圳育成教育科技有限公司 | Education assessment software implementation method based on big data |
| KR20240022744A (en) * | 2022-08-12 | 2024-02-20 | 한국로봇융합연구원 | Child psychology analysis robot using pictures and method for Child psychology analysis |
| CN117649941A (en) * | 2024-01-30 | 2024-03-05 | 凝动万生医疗科技(武汉)有限公司 | Dynamic early warning system for hemorrhagic shock after orthopedics trauma |
| CN117711578A (en) * | 2024-02-06 | 2024-03-15 | 重庆医科大学绍兴柯桥医学检验技术研究中心 | Medical image data analysis management system |
| CN117747064A (en) * | 2023-12-06 | 2024-03-22 | 江苏富敏人工智能科技有限公司 | Depression clinical decision method and system based on AI |
| CN117766083A (en) * | 2023-11-30 | 2024-03-26 | 江苏省苏北人民医院 | Method for constructing warfarin anticoagulation quality and adverse reaction prediction model based on machine learning algorithm and application device thereof |
-
2024
- 2024-04-01 CN CN202410385921.8A patent/CN117976148B/en active Active
Patent Citations (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109192310A (en) * | 2018-07-25 | 2019-01-11 | 同济大学 | A kind of undergraduate psychological behavior unusual fluctuation scheme Design method based on big data |
| CN112669905A (en) * | 2020-12-31 | 2021-04-16 | 中南民族大学 | RNA sequence coding potential prediction method and system based on data enhancement |
| US20220285028A1 (en) * | 2021-03-04 | 2022-09-08 | Abbott Laboratories | Technology to automatically identify the most relevant health failure risk factors |
| CN113539492A (en) * | 2021-06-16 | 2021-10-22 | 甘肃省卫生健康统计信息中心(西北人口信息中心) | Urban health index prediction system, prediction analysis method and storage medium thereof |
| CN114171150A (en) * | 2021-12-10 | 2022-03-11 | 曲阜师范大学 | Health data missing value prediction method and device, computer equipment and storage medium |
| KR20230090465A (en) * | 2021-12-15 | 2023-06-22 | 한양대학교 에리카산학협력단 | Apparatus and method for augmentating of data |
| CN114724701A (en) * | 2022-03-11 | 2022-07-08 | 梁娜 | Noninvasive ventilation curative effect prediction system based on superposition integration algorithm and automatic encoder |
| CN114927230A (en) * | 2022-04-11 | 2022-08-19 | 四川大学华西医院 | Machine learning-based severe heart failure patient prognosis decision support system and method |
| CN115036043A (en) * | 2022-05-24 | 2022-09-09 | 珠海清大声光电工程技术研发中心 | Block chain technology-based children mental health monitoring and management system and method thereof |
| WO2023231995A1 (en) * | 2022-05-30 | 2023-12-07 | 浙大城市学院 | Transfer-learning-based life prediction and health assessment method for aero-engine |
| KR20240022744A (en) * | 2022-08-12 | 2024-02-20 | 한국로봇융합연구원 | Child psychology analysis robot using pictures and method for Child psychology analysis |
| CN116013501A (en) * | 2022-12-12 | 2023-04-25 | 烟台新旧动能转换研究院暨烟台科技成果转移转化示范基地 | Multi-mode arrhythmia classification auxiliary diagnosis system based on data expansion |
| CN116318834A (en) * | 2023-01-16 | 2023-06-23 | 南京信息工程大学 | Internet intrusion detection method and device based on optimized correlation vector machine |
| CN116680613A (en) * | 2023-06-09 | 2023-09-01 | 东北大学 | Human activity recognition comprehensive optimization method based on multi-scale metric learning |
| CN116825348A (en) * | 2023-06-14 | 2023-09-29 | 安徽大学 | University student mental health state assessment model and early warning method based on campus behavior data analysis |
| CN117421582A (en) * | 2023-11-08 | 2024-01-19 | 浙江正泰中自控制工程有限公司 | Equipment health analysis method based on multi-source data driving |
| CN117522222A (en) * | 2023-11-27 | 2024-02-06 | 深圳育成教育科技有限公司 | Education assessment software implementation method based on big data |
| CN117766083A (en) * | 2023-11-30 | 2024-03-26 | 江苏省苏北人民医院 | Method for constructing warfarin anticoagulation quality and adverse reaction prediction model based on machine learning algorithm and application device thereof |
| CN117747064A (en) * | 2023-12-06 | 2024-03-22 | 江苏富敏人工智能科技有限公司 | Depression clinical decision method and system based on AI |
| CN117649941A (en) * | 2024-01-30 | 2024-03-05 | 凝动万生医疗科技(武汉)有限公司 | Dynamic early warning system for hemorrhagic shock after orthopedics trauma |
| CN117711578A (en) * | 2024-02-06 | 2024-03-15 | 重庆医科大学绍兴柯桥医学检验技术研究中心 | Medical image data analysis management system |
Non-Patent Citations (7)
| Title |
|---|
| DOU BOZHENG，等: "Machine Learning Methods for Small Data Challenges in Molecular Science", CHEMICAL REVIEWS, vol. 123, no. 13, 9 July 2023 (2023-07-09), pages 8736 - 8780 * |
| 娄建楼;李燕;王琦;孙博;贾俊奇;: "基于改进堆栈降噪自编码器的锅炉设备在线监测数据清洗方法", 计算机集成制造系统, no. 12, 15 December 2019 (2019-12-15), pages 205 - 214 * |
| 方旭;薛景锋;宋昊;申雅峰;赵启迪;: "飞机结构健康监测系统数据处理及在线应用", 计测技术, no. 06, 28 December 2019 (2019-12-28), pages 4 - 10 * |
| 肖迪尹: "基于机器学习的中医脉诊辅助决策研究", 《中国中国优秀硕士学位论文全文数据库》, no. 1, 15 January 2022 (2022-01-15), pages 056 - 54 * |
| 赵一凡;卞良;张飞飞;: "基于布谷鸟算法优化支持向量机应用于胸痛三联征的分类诊断研究", 生物医学工程研究, no. 01, 25 March 2019 (2019-03-25), pages 59 - 63 * |
| 赵任乾: "《金融商业算法建模 基于Python和SAS》", 31 October 2021, 机械工业出版社, pages: 266 - 268 * |
| 闵金婵;: "基于IHS-FCM的儿童心理健康分析研究", 微型电脑应用, no. 08, 20 August 2020 (2020-08-20), pages 66 - 68 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117976148B (en) | 2024-06-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109102005B (en) | Small sample deep learning method based on shallow model knowledge migration | |
| CN107908688B (en) | A kind of data classification prediction technique and system based on improvement grey wolf optimization algorithm | |
| CN102314614B (en) | Image semantics classification method based on class-shared multiple kernel learning (MKL) | |
| CN112115967A (en) | Image increment learning method based on data protection | |
| CN118133689B (en) | Teaching scene-oriented simulation platform data processing method and device and electronic equipment | |
| CN112116957A (en) | Disease subtype prediction method, system, device and medium based on small sample | |
| CN114019370A (en) | Motor fault detection method based on gray level image and lightweight CNN-SVM model | |
| CN117974221B (en) | Electric vehicle charging station location selection method and system based on artificial intelligence | |
| CN116521863A (en) | Tag anti-noise text classification method based on semi-supervised learning | |
| CN114118146A (en) | Rolling bearing fault diagnosis method and system based on belief rule base | |
| CN113469013B (en) | Motor fault prediction method and system based on transfer learning and time sequence | |
| CN117976148B (en) | Machine learning-based prediction method and system for mental health problems of children | |
| CN108573512B (en) | Complex visual image reconstruction method based on depth coding and decoding dual model | |
| CN114495114B (en) | Text sequence recognition model calibration method based on CTC decoder | |
| CN116861250A (en) | Fault diagnosis model training method and device | |
| CN113297385B (en) | Multi-label text classification system and method based on improved GraphRNN | |
| CN115206539A (en) | Multi-label integrated classification method based on perioperative patient risk event data | |
| CN115062685A (en) | Fault diagnosis method and device, electronic equipment and storage medium | |
| CN110796195B (en) | Image classification method including online small sample excitation | |
| CN111967593A (en) | Method and system for processing abnormal data based on modeling | |
| CN116070120B (en) | Automatic identification method and system for multi-tag time sequence electrophysiological signals | |
| CN117648890B (en) | Semiconductor device modeling method and system based on artificial intelligence | |
| CN112836740B (en) | Markov-based open composite domain-based method for improving model domain adaptivity | |
| CN118378178B (en) | Transformer fault identification method and system based on residual map convolution neural network | |
| Lin et al. | Data Augmentation for Industrial Multivariate Time Series via a Spatial and Frequency Domain Knowledge GAN |
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 |