CN109920551A - Autism children social action performance characteristic analysis system based on machine learning - Google Patents
Autism children social action performance characteristic analysis system based on machine learning Download PDFInfo
- Publication number
- CN109920551A CN109920551A CN201910068956.8A CN201910068956A CN109920551A CN 109920551 A CN109920551 A CN 109920551A CN 201910068956 A CN201910068956 A CN 201910068956A CN 109920551 A CN109920551 A CN 109920551A
- Authority
- CN
- China
- Prior art keywords
- social
- analysis
- children
- learning
- machine learning
- 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.)
- Pending
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 75
- 206010003805 Autism Diseases 0.000 title claims abstract description 37
- 208000020706 Autistic disease Diseases 0.000 title claims abstract description 37
- 238000010801 machine learning Methods 0.000 title claims abstract description 31
- 230000011273 social behavior Effects 0.000 claims abstract description 38
- 238000000034 method Methods 0.000 claims abstract description 35
- 230000006870 function Effects 0.000 claims abstract description 14
- 238000012549 training Methods 0.000 claims description 23
- 238000003066 decision tree Methods 0.000 claims description 16
- 238000011156 evaluation Methods 0.000 claims description 15
- 230000006735 deficit Effects 0.000 claims description 12
- 238000012360 testing method Methods 0.000 claims description 8
- 238000003745 diagnosis Methods 0.000 claims description 7
- 238000013479 data entry Methods 0.000 claims description 6
- 238000011161 development Methods 0.000 claims description 5
- 230000010354 integration Effects 0.000 claims description 5
- 238000013473 artificial intelligence Methods 0.000 claims description 3
- 238000004891 communication Methods 0.000 claims description 3
- 230000008451 emotion Effects 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 230000001186 cumulative effect Effects 0.000 claims description 2
- 239000000284 extract Substances 0.000 claims 1
- 208000024891 symptom Diseases 0.000 abstract description 19
- 230000000694 effects Effects 0.000 abstract description 3
- 208000030251 communication disease Diseases 0.000 abstract 1
- 238000004364 calculation method Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 6
- 230000018109 developmental process Effects 0.000 description 4
- 238000011835 investigation Methods 0.000 description 4
- 238000012417 linear regression Methods 0.000 description 3
- 230000004044 response Effects 0.000 description 3
- 239000000654 additive Substances 0.000 description 2
- 230000000996 additive effect Effects 0.000 description 2
- 230000003542 behavioural effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000003997 social interaction Effects 0.000 description 2
- 238000013179 statistical model Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000003556 assay Methods 0.000 description 1
- 208000013404 behavioral symptom Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000008131 children development Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 230000000474 nursing effect Effects 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
Landscapes
- Electrically Operated Instructional Devices (AREA)
Abstract
The method for the autism children human communication disorders symptom analysis based on machine learning that the invention discloses a kind of comprising the steps of: obtain all types of autism children social behavior feature set；Obtain the analysis of cases report of previous autism children；System learns the information of typing；Update fitting function；New analysis of cases is carried out using updated fitting function.The autism children analysis of cases system based on machine learning that the invention also discloses a kind of.The present invention is based on the analysis methods of machine learning can be according to the analysis of cases information of existing children, and therefrom study is regular to relevant signature analysis automatically；And in later analytic activity, according to the characteristic parameter that newly inputs and study to analysis method be automatically that rehabilitation personnel recommends diagnostic result to be with a wide range of applications to greatly improve the efficiency and accuracy of rehabilitation.
Description
Technical Field
The invention belongs to the technical field of artificial intelligence, and particularly relates to a method and a system for analyzing social behavior characteristics of autistic children based on machine learning.
Background
Symptom analysis is an important ring in the rehabilitation activities of autistic children. At present, in the rehabilitation link of the autism children, a rehabilitation teacher generates new symptom analysis according to characteristic performance and on the basis of the previous symptom analysis with similar performance, and feeds the new symptom analysis back to the autism children.
The rehabilitation teacher generates a new symptom analysis to the autistic children based on the previous expression characteristics and the symptom analysis with similar expression in the past. This assay relies on previous analysis of symptoms with similar manifestations. If the type or the number of the symptoms of the current autistic children are changed, the corresponding symptom analysis is influenced. For the rehabilitation practitioner, if the combination of symptoms after change is different from any previous successful analysis case, it is difficult to accurately judge how the change in current performance affects the symptom analysis. Therefore, error analysis of symptoms is likely to occur, misdiagnosis is caused, and the autistic children cannot be trained correspondingly.
Disclosure of Invention
In order to overcome and solve the problems in the prior art, the invention provides a method and a system for intelligently analyzing the social behavior characteristics of autistic children based on machine learning.
The invention provides a method for analyzing social behavior expression characteristics of autistic children based on machine learning, which comprises the following steps of:
A. acquiring various autism child social behavior expression characteristic sets, and processing the characteristic sets through a social ability evaluation questionnaire to obtain a corrected autism child social behavior expression characteristic set;
B. acquiring existing case analysis report information of the social behavior expression characteristics of the autism children in the past;
C. inputting the set of the behavioral expression characteristics in the step A and the case analysis report information obtained in the step B into a machine learning training system based on an XGboost algorithm, wherein the training system learns the entered case analysis report information, and the learning refers to learning 47 items (namely, the sum of the number of evaluation items in fig. 4) of social basic skill training targets, 6 items of basic information of children with autism and a specific intervention scheme, extracting characteristic parameters from the social basic skill training targets, outputting splitting weights of all nodes of a decision tree, obtaining corresponding decision tree parameters through calculation, and finally obtaining a decision tree which meets expectations, namely a learning model;
D. performing case analysis on the social behavior expression characteristics of the autism children by using a learning model;
E. updating the learning model, and updating the social behavior expression characteristics of the autistic children in the step A and the case analysis report information in the step B;
F. and carrying out new case analysis by using the updated learning model, and carrying out case analysis on the social behavior expression characteristics of the next autism child.
In the step a, the different types of social performance symptoms include different types of performance characteristics. As shown in fig. 3, the social performance characteristics of autistic children include: social attention deficit, self-awareness deficit, social emotion deficit, and communication deficit.
In the step a, the set of social behavior expression characteristics of autistic children is shown in fig. 4: according to the invention, referring to an autism children development assessment table (social field) of middle-incomplete links, three blocks of basic ability, social skills and social etiquette before social contact in original social contact are split and recombined, and according to the social ability development characteristics of autism children, the social abilities are split and recombined into eight sub-blocks of social attention, self-awareness, non-spoken social skills, greeting, farewell, telephone etiquette and high-order etiquette, wherein each sub-block comprises a plurality of sub-items; the set of autistic child performance characteristics shown in fig. 4 is obtained, and the unique method of the present invention is applied in this "set" technical process, which is different from the existing method of obtaining a set by field evaluation:
social ability assessment questionnaire
The capability assessment questionnaire is a questionnaire for parents of autistic children, the questionnaire collects social capability information of autistic children by setting 47 related professional questions in the social capability field, the questions are wide in coverage, low in difficulty and easy to discriminate, and meanwhile, the capability assessment questionnaire is different from a non-programmed and systematic manual investigation method of a traditional rehabilitation teacher;
regarding the feature set collection method in the invention:
(1) description of the operation of the method: the capability assessment questionnaire is issued to parents of autistic children on line through APP and WeChat public numbers to record information, and can be completed without going to an off-line test base, so that the overall efficiency of diagnosis and rehabilitation is improved.
(2) Relationship of feature collection to the set shown in FIG. 4: the questionnaire design is completely corresponding to the 47 social ability indexes, and information in the corresponding ability index field is directly extracted for division and arrangement, so that a frame diagram shown in fig. 4 can be obtained;
(3) compared with the existing manual method, the method has the technical advantages that: the accuracy and the standardability are improved, the investigation mode of a rehabilitation teacher is relatively subjective and random, and the accurate and in-place investigation mode of each time cannot be guaranteed; the full coverage, and the template type questionnaire makes the investigation process always in the same standard, efficient and standard; the efficiency is improved, the tables and the diagrams automatically generated by the computer are convenient for mutual conversion and data processing, and the data integration time and the data integration process are shortened; the invention ensures the uniformity of survey modes and data arrangement, and the questionnaire evaluation and scene evaluation are established based on the same social ability evaluation system, so that the situation that observation variables cannot be connected front and back can not occur from the survey to the collected arrangement results;
in the step B, the case analysis report information refers to the autism child symptom diagnosis and the corresponding rehabilitation suggestion which are evaluated by a rehabilitation teacher; the case analysis report information of the autistic child social behavior expression characteristics, which is specified in the invention, comprises test results of a child social ability development positioning map, social skills, social etiquette and the like, corresponding test result analysis and rehabilitation suggestions and rehabilitation frequency recommendation;
wherein, the step C is specifically as follows:
taking data (the behavioral expression characteristic set in the step A and the case analysis report information in the step B) input by researchers as learning samples, extracting characteristic parameters from the sample data (47 social basic skill training targets, 6 autism children basic information and a specific intervention scheme) obtained in the step B, outputting splitting weights of all nodes of a decision tree, obtaining corresponding decision tree parameters through calculation, and finally obtaining a decision tree which is in line with expectation, namely a learning model.
The characteristic parameters refer to social attention deficit, self-consciousness deficit, social emotion deficit, communication deficit and 6-autism child basic information;
wherein, the basic information of the 6-year autism children refers to age, language ability, learning ability, sex, nursing condition and family background;
wherein, in obtaining the corresponding decision tree parameters by calculation, the calculation refers to that for a given data set with n samples and m features, the input set is:
D＝(xi，yi)(|D|＝n，xi∈Rm，yi∈R)；
and predicting the output by using K accumulated functions by using a tree ensemble model:
wherein, the learning result is judged by utilizing the common goodness-of-fit judgment index of R2, the learning result is used for measuring the proportion occupied by an independent variable explanation part in the variation of a dependent variable in statistics so as to judge the explanation force of a statistical model, and in the linear regression of the experiment, a decision coefficient is the square of a sample correlation coefficient; when other regression arguments are added, the decision coefficient becomes the square of the multiple correlation coefficient accordingly.
Yn, (e.g., autistic child 6 big base information as variable input) for a set of n-valued variables y1, y2... yn:
the overall square sum is then:
the regression sum of squares is:
the decision coefficient can be written as:
the closer the decision coefficient is to 1, the higher the fitting degree of the model is; as shown in fig. 8, substituting random data yields a goodness-of-fit coefficient of 0.7828, which will result in a higher model accuracy if the goodness-of-fit coefficient is closer to 1. The calculation method provided by the invention can greatly improve the rehabilitation efficiency and accuracy, and has wide application prospect.
Step D, case analysis is carried out on the social behavior expression characteristics of the autistic children, and social ability mastering rate judgment and social ability rehabilitation suggestions are provided in the analysis process and are unique to the method;
wherein said updating the learning model of step E specifically includes updating the new performance characteristic values and updating the entered relevant data of step A, B;
based on the method, the invention also provides an analysis system of the social behavior expression characteristics of the autistic children based on machine learning, and the system comprises the following components:
the data entry module comprises a front-end APP and a WeChat public number and is used for entering a set of autism child social behavior expression characteristics and existing case analysis report information of the autism child social behavior expression characteristics;
the server module is responsible for receiving the data in the data entry module and carrying out primary screening;
the learning training module is used for a machine learning training system based on the XGboost algorithm to learn the characteristic information and the diagnosis information of the autism children on the set and the case analysis report information to obtain a learning model; the machine learning training module comprises a PC end program and is used for data integration and model training;
the analysis module is used for analyzing the social behavior expression characteristics of the autism children through the learning model; the analysis module comprises a doc type text tool and is responsible for integrating and presenting final data and presenting the final data to a front-end APP and a WeChat public number.
Compared with the prior art, the invention has the beneficial effects that: according to the intelligent analysis method and system based on machine learning, the related characteristic analysis rules can be automatically learned from the existing case analysis information of the children with autism according to the existing case analysis information of the children with autism; in the later analysis process, a proper treatment scheme is recommended for a rehabilitation teacher according to the newly input characteristic expression (the characteristic expression of the autism child to be detected) and the analysis rule obtained through learning (the existing case analysis information of the autism child), and a diagnosis result is recommended for a rehabilitation worker, so that the accuracy and the efficiency of analyzing the symptoms of the autism child to be detected are greatly improved, and the efficiency and the accuracy of rehabilitation are also improved.
Drawings
FIG. 1 is a diagram of a machine learning model according to the present invention.
FIG. 2 is a general training process according to the present invention.
Fig. 3 shows the social behavior of autistic children in accordance with the present invention.
Fig. 4 is a set of social performance characteristics of autistic children in accordance with the present invention.
Fig. 5 is a schematic diagram of an analysis system for social performance characteristics of autistic children based on machine learning according to the present invention.
FIG. 6 is a schematic diagram of an embodiment of the present invention.
FIG. 7 is a diagram illustrating evaluation results according to an embodiment of the present invention.
FIG. 8 is an example of splitting weights of nodes of a decision tree in a learning model according to the present invention.
FIG. 9 is a schematic diagram of an analysis system according to the present invention.
Detailed Description
The present invention will be described in further detail with reference to the following specific examples and the accompanying drawings. The procedures, conditions, experimental methods and the like for carrying out the present invention are general knowledge and common general knowledge in the art except for the contents specifically mentioned below, and the present invention is not particularly limited.
Example 1
As shown in fig. 1, the method for analyzing social behavior characteristics of autistic children based on machine learning of the present invention includes the following steps (as shown in fig. 2):
A. obtaining a set of social behavioral symptoms of autistic children of a certain type of autistic children. Different types of social performance symptoms include different types of performance characteristics. For example, what needs to be obtained here is a full set of performance characteristics that are encompassed by a certain type of symptom;
B. the expression characteristics of the past symptoms are obtained. Acquiring the previous performance characteristics, taking the past characteristic information (including the test results of the children social ability development positioning map, the social skills, social etiquette and the like, the corresponding test result analysis and rehabilitation suggestion, and the rehabilitation frequency recommendation) as an input sample,and gives the corresponding rehabilitative teacher by the traditional methodCase analysis results are used as output samples and used for a training sample set in machine learning;
C. the entered information is learned. B, using the existing case analysis report information of the social behavior expression characteristics of the autistic children obtained in the step B as machine learning sample data, extracting characteristic parameters from the case analysis report information, outputting splitting weights of all nodes of a decision tree, obtaining corresponding decision tree parameters through calculation, and finally obtaining a decision tree which is a learning model and accords with expectation;
D. case analysis is carried out on the social behavior expression characteristics of the autistic children.
E. Updating the learning model, and updating the learning model of each type of product;
F. a new analysis is performed using the updated learning model. For new performance characteristic values, calculations can be performed using the learning model already obtained, so that an appropriate plan is recommended to the rehabilitee.
The learning model in the embodiment of the invention:
and generating an evaluation report according to the questionnaire data, carrying out artificial intelligence deduction, and forming an evaluation suggestion through mass data operation.
Wherein,
(1) combining 47 social interaction evaluation projects, introducing XGboost to carry out factor evaluation
The XGBoost algorithm is a branch of a full-name eXtreme Gradient Boosting algorithm and is an extensible machine learning system of the Gradient Boosting. The XGboost is an updated distributed gradient enhancement library and aims to realize high efficiency, flexibility and portability. The method realizes a machine learning algorithm under a Gradient Boosting framework. XGBoost provides parallel tree lifting (also known as GBDT, GBM) that can quickly and accurately solve many data science problems. The same code runs on the main distributed environment (Hadoop, SGE, MPI) and can solve the problem of over billions of samples.
1) Overview of the model
The core idea of the XGboost is the same as that of the GBDT, the XGboost is promoted according to the negative gradient direction of the loss function, the XGboost is a GBDT algorithm which is subjected to Taylor quadratic expansion, and some regular terms are added. For a given data set of n samples and m features
D＝(xi，yi)(|D|＝n，xi∈Rm，yi∈R)
A tree ensemble model predicts the output using K cumulative functions:
whereinIs the space of CART (regression tree). Where q represents the structure of each tree that can map each sample into a corresponding leaf node, and T is the number of leaf nodes in the tree. Each corresponding to a separate tree structure q and leaf weight w. Unlike decision trees, each regression tree contains a continuous score value at each leaf node, representing the score of the ith node. Is the score for sample x, i.e., the model prediction value. For each sample, it will be classified into leaf nodes using multiple decision rules in the tree, and the final prediction is obtained by accumulating the scores w in the corresponding leaves (the prediction result for each sample is the sum of the prediction scores for each tree). To learn the set of functions used in the model, the following regularization objectives are minimized:
where l is a microprotrusive loss function that measures the difference between the predicted value and the target value. The second penalty model complexity (sum of all regression trees). The term includes two parts, one is the total number of leaf nodes and one is the L2 regularization term derived from the leaf nodes. This additional regularization term can smooth the learning weights of each leaf node to avoid overfitting. Intuitively, the regularized objective will tend to choose a model that employs simple and predictive functions. When the regularization parameter is zero, the function becomes the conventional GDBT.
2)GDBT
The tree set model has a function as a parameter, so it cannot be updated directly using the conventional update method. Instead, an additive learning mode (additive training) training is adopted, a constant prediction is started, a new function is added every time to learn the current tree, and the current optimal tree model is found and added into the integral model:
therefore, the key is to learn the t-th tree and find the best ftIncrease ftAnd minimizing an objective function, whereinIs the predicted value of sample i at the t-th iteration:
after taking into account empirical errors, the objective function can be rewritten as:
after the empirical error second order taylor expansion:
so the final goal, after removing the constant term:
(2) judging the learning result by the R2 goodness-of-fit judgment index
The XGboost algorithm and 1000+ effective sample data are used for structure learning of evaluation completion, and a suitable intervention scheme is finally calculated by linear regression. The method is used for measuring the proportion of the independent variable interpretation part in the variation of the dependent variable in statistics, so as to judge the interpretation power of the statistical model. For simple linear regression, the decision coefficient is the square of the sample correlation coefficient. When other regression arguments are added, the decision coefficient becomes the square of the multiple correlation coefficient accordingly.
Yn, for a set of n-valued variables y1, y2... yn, there are:
the overall square sum is then:
the regression sum of squares is:
the decision coefficient can be written as:
the closer the decision coefficient is to 1, the higher the fitting degree of the model.
The step D is specifically as follows: as shown in fig. 6, an example is selected, key private information is hidden, data is entered, and an autistic child social interaction ability assessment report is generated:
first, evaluating the result
The evaluation results are shown in fig. 7.
Table 1: FIG. 7 concrete data logout
Note: the test results in the table are score/total score (accuracy)
The results analysis section is a detailed analysis of table 1.
Second, result analysis
The month age of the child is 99 months, the basic skill mastering rate before social contact is 57.14%, the social skill mastering rate is 50.00%, and the social etiquette mastering rate is 48.00%.
As can be seen from fig. 7, the child has zero items reaching the standard, and eight items of social attention, self-awareness, non-spoken social skills, greeting, farewell, phone etiquette, and high-level etiquette have not reached the standard. The children are generally socially basic skills (social attention and self-awareness), generally socially skills (non-spoken social skills and spoken social skills) and generally socially eties (greetings, billings, telephone etiquets and high-order etiquets) on the whole.
Third, rehabilitation suggestion
According to the age of the child and the social ability development sequence, the method suggests that a real scene is combined in daily life, and intervention is performed from the following five items of content through related children songs, stories and life activities.
Item 1, actively initiating caregiver attention
Intervention target:
actively induce carer attention
Intervention strategy:
1) actively induce carer attention
2) Actively induce carer attention
Item 2, smiling to respond to greetings of a stranger
Intervention target:
smiling response to greetings from strangers
Intervention strategy:
1) smiling response to greetings from strangers
2) Smiling response to greetings from strangers
Item 3, strangers approach the device and are in proper flash and hide
Intervention target:
when the stranger walks to the front of the stranger, the stranger looks at the stranger and starts to walk forwards by bypassing the stranger.
Intervention strategy:
1) the child is made to understand to be away from strangers by the related child songs and fairy tales.
2) When a person who is unknown to children and tries to contact limbs of the person who brings children out in daily life is close to the children, the children are prompted by language: do you know Ta? Can handle/let Ta? And the children are indicated to dodge by actions (for example, the back of the two hands is separated from strangers).
Item 4, meaning of the caregiver when the caregiver leaves
Intervention target:
when the carer leaves, the carer has the meaning of chasing the carer
Intervention strategy:
1) when the carer leaves, the carer has the meaning of chasing the carer
2) When the carer leaves, the carer has the meaning of chasing the carer
Item 5, know own article
Intervention target:
articles for realizing oneself
Intervention strategy:
1) articles for realizing oneself
2) Articles for realizing oneself
Fourth, intervention frequency
Not less than 3 times per week;
each time is not less than 30 minutes;
for example, as shown in fig. 8, the splitting weight of each node of the decision tree in the learning model is about 0.21 for age, about 0.19 for learning ability, about 0.19 and 0.04 for gender, about 0.09 and 0.03 for parental care and ancestral care, and about 0.07 and 0.02 for general family and high-known family; fig. 8 shows that, while giving data, the weighted values of the indicators are arranged in a descending order, and thus, in six items of basic information, the age has the largest influence factor on the learning model of the machine, and the language ability and the learning ability are the second order, and the mean square error, the average absolute error and the interpretable variance below the histogram are used for independently assisting in verifying the proportional magnitude relation between the observed total quantity and the observed individual quantity and the corresponding expected values, so that whether the model is updated or not is judged according to the magnitude of the values, and the difference between the model and the absolute ideal model (the regression value is 1) can be generally seen through the goodness of fit coefficient.
As shown in fig. 9, the system for analyzing social behavior characteristics of autistic children based on machine learning in the present embodiment includes:
the data entry module comprises a front-end APP and a WeChat public number and is used for entering a set of social behavior expression characteristics of autism children and existing case analysis report information of the social behavior expression characteristics of the autism children;
the server module is responsible for receiving the data in the data entry module and carrying out primary screening;
the learning training module is used for a machine learning training system based on the XGboost algorithm to learn the characteristic information and the diagnosis information of the autism children on the set and case analysis report information to obtain a learning model; the machine learning training module comprises a PC end program and is used for data integration and model training;
the analysis module is used for analyzing the social behavior expression characteristics of the autistic children through the learning model; the analysis module comprises a doc type text tool and is responsible for integrating and presenting final data and presenting the final data to a front-end APP and a WeChat public number.
The protection of the present invention is not limited to the above embodiments. Variations and advantages that may occur to those skilled in the art may be incorporated into the invention without departing from the spirit and scope of the inventive concept, which is set forth in the following claims.
Claims (10)
1. A method for analyzing social behavior performance characteristics of autistic children based on machine learning, which is characterized by comprising the following steps:
A. acquiring a set of social behavior expression characteristics of autistic children;
B. acquiring case analysis report information of the social behavior expression characteristics of autistic children;
C. the machine learning training system based on the XGboost algorithm is used for learning the feature information and the diagnosis information of the autism children on the set and case analysis report information, wherein the learning refers to learning 47 social basic skills and 6 basic information of the autism children to obtain a learning model; the learning model is a gradient lifting tree model;
D. and analyzing the social behavior expression characteristics of the autistic children through the learning model.
2. The method of analyzing social performance characteristics of autistic children based on machine learning of claim 1, further comprising updating the learning model, comprising:
E. updating the learning model, and updating the social behavior expression characteristics of the autistic children in the step A and the case analysis report information in the step B;
F. and analyzing social behavior expression characteristics of the next autistic child by using the updated learning model.
3. The method of machine learning-based analysis of social performance characteristics of autistic children as claimed in claim 1, wherein the autistic children social performance characteristics comprise: social attention deficit, self-awareness deficit, social emotion deficit, and communication deficit.
4. The method for analyzing social performance characteristics of autistic children according to claim 1, wherein in the step a, a social ability assessment questionnaire is adopted for gathering: the social ability assessment questionnaire is a questionnaire for parents of autistic children, and the questionnaire collects social ability information of autistic children by setting 47 social basic skills.
5. The method for analyzing the social behavior characteristics of autistic children according to claim 4, wherein the analysis of the social behavior characteristics of autistic children means that the learning model evaluates questionnaire data according to the social ability to generate an evaluation report, performs artificial intelligence deduction, and forms an evaluation suggestion through data operation.
6. The method for analyzing social performance characteristics of autistic children according to claim 1, wherein in the step B, the case analysis report information includes a children social ability development positioning map, social skills, social contact etiquette test results, corresponding test result analysis and rehabilitation suggestions and rehabilitation frequency recommendations.
7. The method for analyzing social behavior characteristics of autistic children according to claim 1, wherein in the step C, the machine learning training system extracts feature parameters from the sample data by using the set and the case analysis report information as the sample data, outputs splitting weights of nodes of the decision tree, and calculates parameters of the decision tree to obtain the learning model.
8. The method of analyzing social performance characteristics of autistic children according to claim 7, wherein the learning model determines the learning result by using a goodness-of-fit determination index of R ^ 2.
9. The method of claim 7, wherein the computing of the decision tree parameters is performed on a given data set of n samples and m features, and the input set is: d ═ xi，yi)(|D|＝n，xi∈Rm，yiE.r) and predicts the output using K cumulative functions with a tree ensemble model:
wherein the decision coefficient is the square of the sample correlation coefficient; when the regression argument is added, the decision coefficient is changed to the square of the multiple correlation coefficient accordingly; for a set of n-valued variables y1, y2... yn, the coefficients are determined as:
wherein,
10. an analysis system for social performance characteristics of autistic children based on machine learning, characterized in that the analysis method according to any of claims 1-9 is used, the system comprising the following:
the data entry module comprises a front-end APP and a WeChat public number and is used for entering a set of autism child social behavior expression characteristics and existing case analysis report information of the autism child social behavior expression characteristics;
the server module is responsible for receiving the data in the data entry module and carrying out primary screening;
the learning training module is used for a machine learning training system based on the XGboost algorithm to learn the characteristic information and the diagnosis information of the autism children on the set and the case analysis report information to obtain a learning model; the machine learning training module comprises a PC end program and is used for data integration and model training;
the analysis module is used for analyzing the social behavior expression characteristics of the autism children through the learning model; the analysis module comprises a doc type text tool and is responsible for integrating and presenting final data and presenting the final data to a front-end APP and a WeChat public number.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910068956.8A CN109920551A (en) | 2019-01-24 | 2019-01-24 | Autism children social action performance characteristic analysis system based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910068956.8A CN109920551A (en) | 2019-01-24 | 2019-01-24 | Autism children social action performance characteristic analysis system based on machine learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109920551A true CN109920551A (en) | 2019-06-21 |
Family
ID=66960661
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910068956.8A Pending CN109920551A (en) | 2019-01-24 | 2019-01-24 | Autism children social action performance characteristic analysis system based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109920551A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110237397A (en) * | 2019-07-11 | 2019-09-17 | 重庆市中医院 | A kind of brain cognitive training system |
| CN110674875A (en) * | 2019-09-25 | 2020-01-10 | 电子科技大学 | Pedestrian motion mode identification method based on deep hybrid model |
| CN111009321A (en) * | 2019-08-14 | 2020-04-14 | 电子科技大学 | Application method of machine learning classification model in juvenile autism auxiliary diagnosis |
| CN111461937A (en) * | 2020-03-31 | 2020-07-28 | 北京复米教育科技有限公司 | Automatic tracking method and system for autism intervention teaching course |
| CN111524577A (en) * | 2020-04-17 | 2020-08-11 | 广东医博荟健康管理有限公司 | Autism integral intervention treatment auxiliary system and method |
| CN112668809A (en) * | 2020-10-30 | 2021-04-16 | 华东师范大学 | Method for establishing autism child rehabilitation effect prediction model and method and system for predicting autism child rehabilitation effect |
| US20210174933A1 (en) * | 2019-12-09 | 2021-06-10 | Social Skills Training Pty Ltd | Social-Emotional Skills Improvement |
| US20240062897A1 (en) * | 2022-08-18 | 2024-02-22 | Montera d/b/a Forta | Artificial intelligence method for evaluation of medical conditions and severities |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2003075187A1 (en) * | 2002-03-02 | 2003-09-12 | Bl Systems Inc. | Apparatus and method for constructing data mining model using ensemble machines |
| US20040107413A1 (en) * | 2002-11-22 | 2004-06-03 | Bixler John D. | Autism treatment system and method |
| US20160180735A1 (en) * | 2014-12-19 | 2016-06-23 | International Business Machines Corporation | Coaching a participant in a conversation |
| US20160179786A1 (en) * | 2014-12-19 | 2016-06-23 | International Business Machines Corporation | Diagnosing autism spectrum disorder using natural language processing |
| US20170069216A1 (en) * | 2014-04-24 | 2017-03-09 | Cognoa, Inc. | Methods and apparatus to determine developmental progress with artificial intelligence and user input |
| CN107748736A (en) * | 2017-10-13 | 2018-03-02 | 河海大学 | A kind of multiple-factor Remote Sensing temperature space NO emissions reduction method based on random forest |
| CN107910070A (en) * | 2017-11-06 | 2018-04-13 | 广州优涵信息技术有限公司 | A kind of expert recommendation method based on expert's case library |
| CN108010568A (en) * | 2017-03-24 | 2018-05-08 | 华东师范大学 | A kind of rehabilitation training system based on more course plot modules |
| CN108346107A (en) * | 2017-12-28 | 2018-07-31 | 阿里巴巴集团控股有限公司 | A kind of social content Risk Identification Method, device and equipment |
| CN108538365A (en) * | 2017-03-24 | 2018-09-14 | 华东师范大学 | A kind of self-closing disease sociability assessment system based on data analysis technique |
| CN109256207A (en) * | 2018-08-29 | 2019-01-22 | 王雁 | A method of based on XGBoost+SVM hybrid machine Learner diagnosis keratoconus case |
-
2019
- 2019-01-24 CN CN201910068956.8A patent/CN109920551A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2003075187A1 (en) * | 2002-03-02 | 2003-09-12 | Bl Systems Inc. | Apparatus and method for constructing data mining model using ensemble machines |
| US20040107413A1 (en) * | 2002-11-22 | 2004-06-03 | Bixler John D. | Autism treatment system and method |
| US20170069216A1 (en) * | 2014-04-24 | 2017-03-09 | Cognoa, Inc. | Methods and apparatus to determine developmental progress with artificial intelligence and user input |
| US20160180735A1 (en) * | 2014-12-19 | 2016-06-23 | International Business Machines Corporation | Coaching a participant in a conversation |
| US20160179786A1 (en) * | 2014-12-19 | 2016-06-23 | International Business Machines Corporation | Diagnosing autism spectrum disorder using natural language processing |
| CN108010568A (en) * | 2017-03-24 | 2018-05-08 | 华东师范大学 | A kind of rehabilitation training system based on more course plot modules |
| CN108538365A (en) * | 2017-03-24 | 2018-09-14 | 华东师范大学 | A kind of self-closing disease sociability assessment system based on data analysis technique |
| CN107748736A (en) * | 2017-10-13 | 2018-03-02 | 河海大学 | A kind of multiple-factor Remote Sensing temperature space NO emissions reduction method based on random forest |
| CN107910070A (en) * | 2017-11-06 | 2018-04-13 | 广州优涵信息技术有限公司 | A kind of expert recommendation method based on expert's case library |
| CN108346107A (en) * | 2017-12-28 | 2018-07-31 | 阿里巴巴集团控股有限公司 | A kind of social content Risk Identification Method, device and equipment |
| CN109256207A (en) * | 2018-08-29 | 2019-01-22 | 王雁 | A method of based on XGBoost+SVM hybrid machine Learner diagnosis keratoconus case |
Non-Patent Citations (1)
| Title |
|---|
| 车少辉: "基于神经网络方法的杉木人工林林分生长模拟研究", 《中国博士学位论文全文数据库农业科技辑》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110237397A (en) * | 2019-07-11 | 2019-09-17 | 重庆市中医院 | A kind of brain cognitive training system |
| CN111009321A (en) * | 2019-08-14 | 2020-04-14 | 电子科技大学 | Application method of machine learning classification model in juvenile autism auxiliary diagnosis |
| CN110674875A (en) * | 2019-09-25 | 2020-01-10 | 电子科技大学 | Pedestrian motion mode identification method based on deep hybrid model |
| US20210174933A1 (en) * | 2019-12-09 | 2021-06-10 | Social Skills Training Pty Ltd | Social-Emotional Skills Improvement |
| CN111461937A (en) * | 2020-03-31 | 2020-07-28 | 北京复米教育科技有限公司 | Automatic tracking method and system for autism intervention teaching course |
| CN111524577A (en) * | 2020-04-17 | 2020-08-11 | 广东医博荟健康管理有限公司 | Autism integral intervention treatment auxiliary system and method |
| CN112668809A (en) * | 2020-10-30 | 2021-04-16 | 华东师范大学 | Method for establishing autism child rehabilitation effect prediction model and method and system for predicting autism child rehabilitation effect |
| US20240062897A1 (en) * | 2022-08-18 | 2024-02-22 | Montera d/b/a Forta | Artificial intelligence method for evaluation of medical conditions and severities |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109920551A (en) | Autism children social action performance characteristic analysis system based on machine learning | |
| CN110556129B (en) | Bimodal emotion recognition model training method and bimodal emotion recognition method | |
| Misiunas et al. | DEANN: A healthcare analytic methodology of data envelopment analysis and artificial neural networks for the prediction of organ recipient functional status | |
| CN106295186A (en) | A kind of method and system of aided disease diagnosis based on intelligent inference | |
| CN104573359B (en) | A kind of mass-rent labeled data integration method of task based access control difficulty and mark person's ability | |
| CN113284623B (en) | Personalized cognitive training task recommendation algorithm and system based on user ability | |
| Alonso et al. | Cooperation between expert knowledge and data mining discovered knowledge: Lessons learned | |
| WO2020224433A1 (en) | Target object attribute prediction method based on machine learning and related device | |
| CN107315775A (en) | A kind of index calculating platform and method | |
| CN106096286A (en) | Clinical path formulating method and device | |
| CN110046757B (en) | Outpatient clinic volume prediction system and prediction method based on LightGBM algorithm | |
| Di Garbo et al. | Concepts and methods for integrating language typology and sociolinguistics | |
| Elayidom et al. | A generalized data mining framework for placement chance prediction problems | |
| Viroonluecha et al. | Salary predictor system for thailand labour workforce using deep learning | |
| CN112836105A (en) | Large-scale student aerobic capacity clustering method based on movement physiological characterization fusion | |
| KR20110098286A (en) | Self health diagnosis system of oriental medicine using fuzzy inference method | |
| Stanković et al. | Artificial neural network model for prediction of students’ success in learning programming | |
| Li et al. | Wearable-based human activity recognition with spatio-temporal spiking neural networks | |
| Quach et al. | Evaluation of the efficiency of the optimization algorithms for transfer learning on the rice leaf disease dataset | |
| Jayaprakash et al. | Multinomial logistic regression method for early detection of autism spectrum disorders | |
| Pingle | Evaluation of mental stress using predictive analysis | |
| CN115147768B (en) | Fall risk assessment method and system | |
| CN116775860A (en) | Unsupervised opinion abstract generation method and system based on antagonism framework | |
| CN106021927A (en) | Dermatoglyph analysis and processing method based on big data | |
| Danousis et al. | A Machine-Learning-Based Motor and Cognitive Assessment Tool Using In-Game Data from the GAME2AWE Platform |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190621 |