Infant growth and development evaluation model system based on probability graph

Abstract

The invention discloses an infant growth and development evaluation model system based on a probability graph, and the system comprises a data collection module, a data preprocessing module, a probability graph model building module, a development prediction module and a factor matching module, and comprises the following steps: S1, collecting and obtaining the performance data of an infant in each development field through the data collection module, obtaining influence factor data; s2, preprocessing the collected data through a data preprocessing module to ensure the quality and availability of the data; s3, modeling the preprocessed data through a probability graph model building module, and building a probability graph model of child growth and development; s4, predicting future development by adopting a probabilistic reasoning algorithm based on the probabilistic graph model through a development prediction module, and generating a development prediction report; and S5, analyzing the probability graph model through a factor matching module, and matching a personalized growth and development scheme for each child.

Description

Technical Field

[0001] The present invention relates to the field of evaluation technologies, and particularly to an evaluation model system for the growth and development of young children based on a probabilistic graph. Background Art

[0002] The young child evaluation technology aims to quantify the development levels of young children in dimensions such as cognition, language, emotion, social interaction, and movement through scientific methods, providing a basis for intervention for educators and parents.

[0003] Existing evaluation methods for the growth and development of young children often rely on subjective observations and simple statistical indicators, lacking a comprehensive consideration of the multi-dimensional, dynamic, and complex nature of young children's development, and the evaluation dimensions are relatively single.

[0004] At the same time, manually sorting data takes a long time, making it difficult to meet the needs of real-time evaluation. The evaluation process is cumbersome and inefficient, and it is difficult to accurately capture the key factors affecting the growth of young children and their interaction relationships.

[0005] In addition, through manual scoring, it is easily affected by subjective judgments and also lacks data-based prediction capabilities and dynamics.

[0006] Therefore, there is an urgent need for a more scientific and intelligent evaluation method to improve the accuracy and efficiency of the evaluation of the growth and development of young children. Summary of the Invention

[0007] The purpose of the present invention is to provide an evaluation model system for the growth and development of young children based on a probabilistic graph to overcome the deficiencies in the prior art.

[0008] To achieve the above purpose, the present invention provides the following technical solutions: The present application discloses an evaluation model system for the growth and development of young children based on a probabilistic graph, including a data collection module, a data preprocessing module, a probabilistic graph model construction module, a development prediction module, and a factor matching module, and includes the following steps: S1: Collect and obtain the performance data of young children in various development fields and the influencing factor data through the data collection module; S2: Preprocess the collected data through the data preprocessing module to ensure the quality and usability of the data; S3: Model the preprocessed data through the probabilistic graph model construction module to construct a probabilistic graph model for the growth and development of young children; S4: Through the development prediction module, based on the probabilistic graph model, use a probabilistic inference algorithm to predict future development and generate a development prediction report; S5: Through the factor matching module, analyze the probabilistic graph model and match a personalized growth and development plan for each young child.

[0009] Preferably, the S1 includes the following: data preparation and data collection; wherein, the data preparation includes several evaluation dimensions and evaluation indicators; and data of several children are collected according to the several evaluation dimensions and evaluation indicators to generate several children data results.

[0010] Preferably, the preprocessing in the S2 includes cleaning, normalization and missing value processing, and interference items existing in the data are screened out to ensure the effectiveness of the data.

[0011] Preferably, the S3 includes the following sub-steps: S31: Screen out data individuals with significance; S32: Calculate the influence weight of the data screened out in S31; S33: Establish a probabilistic graphical model based on the influence weight of the data.

[0012] The influence weight can be calculated by a non-linear function based on the significance probability value and the adjustment parameter.

[0013] Preferably, the S31 includes the following sub-steps: S311: Conduct statistical analysis on the preprocessed data and calculate the probability distribution of each dimension therein; S312: Calculate the significance threshold of the current data probability through the significance threshold formula; S313: When the probability of the data individual exceeds the significance threshold, it is determined that the data individual has significance.

[0014] Preferably, the S3 also includes the following: Every other certain time, automatic traversal is performed, and re-modeling is carried out through the probabilistic graphical model construction module to update the probability distribution and the influence weight; and the newly generated probability value is compared with the previous probability value, and calibration is carried out according to the fluctuation range of the new and old probability values; when the model is automatically updated periodically, the historical data retention period can be set according to the data, and then the probability fluctuation range is calibrated based on the data scale change rate.

[0015] Preferably, the S4 includes the following: According to the data on one of the dimensions in the probabilistic graphical model, comprehensive evaluation is carried out through the several performances of the children on this dimension; when predicting several dimensions, according to the data on several dimensions in the probabilistic graphical model, the performances of the children under several dimensions based on different dimensions are respectively averaged, and then comprehensive evaluation is carried out; when making multi-dimensional predictions, the conditional probability distribution and the average value weighting method can be adopted to improve the credibility of the prediction results.

[0016] Preferably, the S5 includes the following sub-steps: S51: Construct a directed graph based on the significant influence relationships existing among all factors in the probabilistic graph model; S52: Select one of the factors in the directed graph and extract the associated relevant nodes; S53: Evaluate according to the data information among the associated nodes; S54: Formulate a targeted strategy based on the data information and the evaluation results.

[0017] Meanwhile, after the directed graph is constructed, an intervention strategy can be generated based on the node influence weights, and a positive incentive and negative factor avoidance scheme can be adopted.

[0018] This application also discloses an evaluation device for children's growth and development based on a probabilistic graph, including a memory and one or more processors. Executable code is stored in the memory. When the one or more processors execute the executable code, it is used to implement the above-mentioned evaluation model system for children's growth and development based on a probabilistic graph.

[0019] This application also discloses a computer-readable storage medium, on which a program is stored. When the program is executed by a processor, it implements the above-mentioned evaluation model system for children's growth and development based on a probabilistic graph.

[0020] Advantages of the present invention: (1), The evaluation model system for children's growth and development based on a probabilistic graph of the present invention can comprehensively and dynamically evaluate the growth and development status of children, improving the accuracy and scientific nature of the evaluation; (2), Through the probabilistic graph model and the development prediction algorithm, the system of the present invention can accurately predict the future development trend of children, providing forward-looking guidance for early childhood education; (3), Through the factor matching module, the present invention adopts automatic matching of significant influencing factors, helping educators and parents better understand the development needs of children and formulating personalized education plans; (4), Through the systematic evaluation process and intelligent analysis function, the present invention greatly improves the efficiency and convenience of the evaluation of children's growth and development; (5), This system can be widely applied to fields such as kindergartens, early education institutions, and family education, and has important social value and market prospects.

[0021] The features and advantages of the present invention will be described in detail through embodiments in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 is a flowchart of the steps of the evaluation model system for children's growth and development based on a probabilistic graph of the present invention; Figure 2 is a schematic structural diagram of the present invention; Figure 3 is a schematic diagram of the directed acyclic graph structure of the present invention; Figure 4 is a schematic diagram of the child nodes and influence of the adaptability of the present invention to the new environment; Figure 5 is a schematic diagram of the device of the present invention; Figure 6 is the result of traversing all possible values of the evaluation dimension of the present invention. Detailed implementation manners

[0023] To make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. However, it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the scope of the present invention. In addition, in the following description, the descriptions of well-known structures and technologies are omitted to avoid unnecessarily confusing the concepts of the present invention.

[0024] Refer to Figures 1 to 4 , an infant growth and development evaluation model system based on a probabilistic graph is provided in an embodiment of the present invention, including a data acquisition module, a data preprocessing module, a probabilistic graph model construction module, a development prediction module, and a factor matching module.

[0025] Data acquisition module: responsible for collecting the performance data of infants in various development fields (such as cognition, language, social interaction, emotion, movement, etc.) and the relevant influencing factor data.

[0026] Data preprocessing module: responsible for cleaning, normalizing, and handling missing values of the collected data to ensure the quality and usability of the data.

[0027] Probabilistic graph model construction module: responsible for modeling the preprocessed data to construct a probabilistic graph model for infant growth and development. The nodes in the probabilistic graph represent different development indicators and influencing factors, and the edges represent the dependence relationship and influence weight between them.

[0028] Development prediction module: responsible for predicting the future development of infants based on the constructed probabilistic graph model by using a probabilistic inference algorithm and generating a development prediction report.

[0029] Factor matching module: responsible for automatically analyzing the probabilistic graph model, identifying the key factors that are significantly significant for infant development, and matching personalized growth and development plans for each infant.

[0030] The present invention uses a directed acyclic graph structure to model the complex relationships in the growth and development of young children. In a directed acyclic graph, each node represents an influencing factor and specific evaluation indicators for that factor. If there is a significant dependency between different node evaluations, it can be marked with a directed connection line. The starting point of the connection line represents the "cause", the ending point represents the "effect", and the weight value on the connection line quantifies the influence between the "cause" and the "effect".

[0031] Example 1 Data preparation: To make the entire modeling process concise and clear, 5 evaluation dimensions and related evaluation indicators are designed (refer to Table 1), and the data results of 50 young children are prepared (refer to Table 2).

[0032] Data processing: Based on Table 2, simple statistical analysis is carried out. When traversing each evaluation dimension to determine a certain value, the probability distribution of different values of other evaluation dimensions is obtained (refer to Table 3, showing the probabilities of different values of other dimensions when the value of "interest in new things" is 1).

[0033] Refer to Figure 6 , by traversing all possible value possibilities of all evaluation dimensions, the following results can be obtained.

[0034] When the probability exceeds the significance threshold, it can be considered significant ( Figure 6 all the yellow-filled values in). Significance threshold formula: Where N is the number of possible values of the evaluation dimension (all are 3 in the above table, and the value range of N must be greater than or equal to 2), C is a tuning parameter (default: 10%), so D is equal to 60% in this example.

[0035] Calculate the influence weight y: For all significant values, apply the following formula: Where C1 and C2 are tuning parameters (default: C1 = 0.46, C2 = 0.85), and f(x) is the function: Where D is the significance threshold.

[0036] Figure 6 For all the yellow-labeled content in Drawing: Based on the calculation results in Table 4, if there are line segments with influence and the influence is greater than 0, a directed connection line can be drawn (refer to Figure 3 ) Self-optimization and improvement technology of probabilistic graphical models The influence calculation of probabilistic graphical models is based on statistics, so it depends on the size of the data volume and the accuracy of data collection. When the number of children with data exceeds 2000 (N>2000, N = 50 in the above example), both the probability and influence values will tend to be stable.

[0037] For evaluation dimensions, there are concepts such as objective / semi-objective and semi-subjective / subjective. For example, for the evaluation of "frequency of active questions", since the specific scores have clear and observable criteria, it belongs to objective evaluation. For "exploration behavior", since the specific score evaluation tends to the subjective feeling of the evaluator, it belongs to semi-objective and semi-subjective evaluation.

[0038] Daily data records are obtained through three channels: Automatic collection by intelligent devices: For example, through wearable devices, visual recognition devices, etc., automatically collect objective data and complete statistics and records.

[0039] Teachers' manual data recording: Teachers use various convenient input methods such as checklists and observation records, and with the assistance of artificial intelligence, complete the preprocessing and classification of data (such as automatically checking after content analysis according to the input text information, etc.).

[0040] Parents of children record data: Teachers of children supplement data on their own children through daily observation records, checklist records, etc.

[0041] Every other period (such as one week / one month / one quarter), the system will automatically traverse and update the probability distribution and influence of each evaluation dimension (repeating the modeling process). During this process, historical data will be retained to a certain extent according to the objective / semi-objective and semi-subjective / subjective attributes of the evaluation dimension. For example: Retain data for the past 6 months for objective data; only retain data for the past 3 months for semi-objective and semi-subjective data; only retain data for the past 1 month for subjective data.

[0042] If the newly generated probability value fluctuates by more than 10% compared to the previous probability value, the following calibration operations will be performed: 1. Compare the change in the scale of statistical data (change in N). If the positive change in N (increase) exceeds 30%, no calibration will be performed.

[0043] 2. If the positive change in N is in the range of 10%-30%, reduce the fluctuation of the probability value by 20% for calibration; 3. If the positive change of N is within 10%, the fluctuation of the probability value is reduced by 50% for calibration; 4. If the negative change (decrease) of N exceeds 30%, the fluctuation of the probability value is reduced by 80% for calibration; 5. If the negative change of N is in the range of 10% - 30%, the fluctuation of the probability value is reduced by 50% for calibration; 6. If the negative change of N is within 10%, the fluctuation of the probability value is reduced by 20% for calibration.

[0044] Recalculate the connections and influence values between each node according to the new probability distribution.

[0045] Infant Development Prediction Algorithm Through Figure 6 the probability distribution, it is possible to judge the possibility of other unknown attributes under any known conditions.

[0046] Taking Figure 6 data as an example: When it is known that a certain infant "is very persistent when encountering challenges (=3)", it is possible to judge the possibility of each evaluation index in dimensions such as "interest in new things", "frequency of asking questions actively", "adaptation to new environment", and "exploration behavior" of this infant (see Table 5). If there is no evaluation data on the "adaptation to new environment" of this infant, but through the above table, it can be judged that the possibility of this child "not being adapted (will be more resistant and show fear emotions)" is 0%; the possibility of "being able to adapt, but may show restraint" is 36.4%; the possibility of "being very adapted and able to integrate into the environment quickly" is 63.6%.

[0047] Based on this data, teachers can focus more on those children who are not adapted in the corresponding teaching arrangement activities.

[0048] For the case where multiple conditions are known, such as knowing two data "being very persistent when encountering challenges (=3)" and "actively exploring (=3)", the following results can be obtained after querying the probability distribution table (Table 6). Based on this result, the average value of the possibilities under each dimension is obtained, as shown in Table 7: According to this table, the possibility that this child "is not adaptable (will be more resistant and show fear)" is 0%; the possibility that "can adapt, but may show restraint" is 40.95%; the possibility that "is very adaptable and can quickly integrate into the environment" is 59.05%. Moreover, since this data is calculated from two data sources, the credibility will be higher than that of data from a single data source.

[0049] Personalized matching algorithm In the previous text Figure 3 , the significant influence relationships between all evaluation factors are presented in the form of a directed graph, which can be used to find all significant related factors that affect a certain factor. This is described through the following examples: 1. The teacher hopes to improve a certain child's adaptability to the new environment; 2. Search for Figure 3 the nodes related to "adaptability to the new environment" in the directed graph, and extract all associated first-degree nodes, that is, the sub-nodes at the next level; As Figure 4 shown: 1. Can adapt to the new environment: Often ask questions actively: This factor has a strong positive influence on "can adapt to the new environment" (influence level is 1.41). This means that children who often ask questions actively are more likely to adapt to the new environment.

[0050] Tend to give up easily when facing challenges: This factor has a small negative influence on "can adapt to the new environment" (influence level is 0.38). This means that children who tend to give up easily when facing challenges have weaker abilities to adapt to the new environment.

[0051] 2. Very adaptable to the new environment: Often ask questions actively: This factor also has a strong positive influence on "very adaptable to the new environment" (influence level is 1.41), the same as "can adapt to the new environment".

[0052] Very persistent when facing challenges: This factor has a strong positive influence on "very adaptable to the new environment" (influence level is 1.28). This means that children who are very persistent when facing challenges are more likely to adapt to the new environment.

[0053] Sometimes explore: This factor has a small positive influence on "very adaptable to the new environment" (influence level is 0.42).

[0054] 3. Not adaptable to the new environment: Occasionally ask questions actively: This factor has a small negative influence on "not adaptable to the new environment" (influence level is 0.38). This means that children who occasionally ask questions actively have weaker abilities to adapt to the new environment.

[0055] Tend to give up easily when faced with challenges: This factor has a relatively large positive impact on "not adapting to the new environment" (the influence is 1.28). This means that children who tend to give up easily when faced with challenges are more likely to show discomfort in the new environment.

[0056] Reject exploration: This factor has a relatively small positive impact on "not adapting to the new environment" (the influence is 0.53). This means that children who reject exploration are more likely to show discomfort in the new environment.

[0057] If teachers hope to improve children's adaptability to the new environment, based on the content listed in this figure, a series of targeted strategies can be adopted. First of all, encouraging children to ask questions actively frequently is a key measure. This can not only enhance their curiosity and thirst for knowledge, but also help them feel more confident and secure in the new environment, thus effectively improving their adaptability. Secondly, cultivating children's perseverance when faced with challenges is equally important. By setting tasks or games with appropriate difficulties and letting children experience the process from difficulties to problem-solving, their perseverance and problem-solving abilities can be exercised, thereby promoting their good adaptation to the new environment. In addition, for those children who tend to give up or reject exploration easily when faced with challenges, teachers should give more attention and support, provide positive feedback and encouragement, reduce their sense of frustration, and stimulate their desire to explore. Through these methods, children can be helped to establish a positive attitude towards the new environment and ultimately achieve better adaptation effects.

[0058] An embodiment of the evaluation device for children's growth and development based on a probabilistic graph of the present invention can be applied to any device with data processing capabilities. This device with data processing capabilities can be a device or apparatus such as a computer. The device embodiment can be implemented by software, or by hardware or a combination of software and hardware. Taking software implementation as an example, as a logically meaningful device, it is formed by the processor of any device with data processing capabilities reading the corresponding computer program instructions in the non-volatile memory into the memory for operation. From the hardware level, as Figure 5 shown, it is a hardware structure diagram of any device with data processing capabilities where the evaluation device for children's growth and development based on a probabilistic graph of the present invention is located. In addition to Figure 5 the shown processor, memory, network interface, and non-volatile memory, the device with data processing capabilities where the embodiment of the device is located usually also includes other hardware according to the actual functions of the device with data processing capabilities, which will not be elaborated here. The implementation processes of the functions and roles of each unit in the above device are specifically described in the implementation processes of the corresponding steps in the above method and will not be elaborated here.

[0059] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to the descriptions of the method embodiments. The device embodiments described above are only illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of the present invention. A person of ordinary skill in the art can understand and implement it without creative work.

[0060] An embodiment of the present invention further provides a computer-readable storage medium, on which a program is stored. When the program is executed by a processor, it implements a device for evaluating the growth and development of infants and young children based on a probabilistic graph in the above embodiments.

[0061] The computer-readable storage medium may be an internal storage unit of any device with data processing capabilities described in any of the foregoing embodiments, such as a hard disk or memory. The computer-readable storage medium may also be an external storage device of any device with data processing capabilities, such as a plug-in hard disk, a Smart Media Card (SMC), an SD card, a Flash Card, etc. equipped on the device. Further, the computer-readable storage medium may also include both an internal storage unit and an external storage device of any device with data processing capabilities. The computer-readable storage medium is used to store the computer program and other programs and data required by any device with data processing capabilities, and can also be used to temporarily store the data that has been output or will be output.

[0062] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, or improvements made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A probability graph-based evaluation model system for children's growth and development, characterized in that: It includes a data collection module, a data preprocessing module, a probabilistic graphical model construction module, a development prediction module, and a factor matching module, and includes the following steps: S1: Collect the performance data of children in various development fields and influencing factor data through the data collection module; S2: Preprocess the collected data through the data preprocessing module to ensure the quality and usability of the data; S3: Model the preprocessed data through the probabilistic graphical model construction module to construct a probabilistic graphical model for children's growth and development; S4: Through the development prediction module, based on the probabilistic graphical model, use a probabilistic inference algorithm to predict future development and generate a development prediction report; S5: Through the factor matching module, analyze the probabilistic graphical model to match a personalized growth and development plan for each child.

2. The system of an infant growth and development evaluation model based on a probabilistic graph according to claim 1, wherein: The S1 includes the following content: including data preparation and data collection; among them, data preparation includes several evaluation dimensions and evaluation indicators; and data collection is carried out on several children according to the several evaluation dimensions and evaluation indicators to generate several children's data results.

3. The system of an infant growth and development evaluation model based on a probabilistic graph as described in claim 1, wherein: The preprocessing in S2 includes cleaning, normalization, and missing value processing, and screening out the interference items existing in the data to ensure the effectiveness of the data.

4. The evaluation model system for the growth and development of young children based on a probabilistic graph according to claim 1, wherein: The S3 includes the following sub-steps: S31: Screen out significant data individuals; S32: Calculate the influence weights of the data screened out in S31; S33: Based on the influence weights of the data, establish a probabilistic graphical model.

5. The system of an infant growth and development assessment model based on a probabilistic graph according to claim 4, characterized in that: The S31 includes the following sub-steps: S311: Conduct statistical analysis on the preprocessed data and calculate the probability distribution of each dimension therein; S312: Calculate the significance threshold of the current data probability through the significance threshold formula; S313: When the probability of a data individual exceeds the significance threshold, it is determined that the data individual is significant.

6. The system of an infant growth and development assessment model based on a probabilistic graph according to claim 4, characterized in that: The S3 also includes the following content: Every other certain time, perform an automatic traversal, re-model through the probabilistic graphical model construction module, update the probability distribution and influence weights; and compare the newly generated probability value with the previous probability value, and calibrate according to the fluctuation range of the new and old probability values.

7. The system of an infant growth and development assessment model based on a probabilistic graph according to claim 1, characterized in that: The S4 includes the following content: According to the data on one of the dimensions in the probabilistic graphical model, conduct a comprehensive evaluation through the several performances of children on this dimension; when predicting for several dimensions, according to the data on several dimensions in the probabilistic graphical model, take the average value of the performances made by children under several dimensions based on different dimensions respectively, and then conduct a comprehensive evaluation.

8. The system of an infant growth and development evaluation model based on a probabilistic graph according to claim 1, wherein: The S5 includes the following sub-steps: S51: Construct a directed graph according to the significant influence relationships existing among all factors in the probabilistic graphical model; S52: Select one of the factors in the directed graph and extract the associated relevant nodes; S53: Evaluate according to the data information between the associated nodes; S54: Develop targeted strategies according to the data information and evaluation results.

9. An evaluation device for the growth and development of young children based on a probabilistic graph, characterized in that: It includes a memory and one or more processors. Executable code is stored in the memory. When the one or more processors execute the executable code, it is used to implement a probabilistic graph-based infant growth and development assessment model system according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: A program is stored thereon. When the program is executed by a processor, it implements a probabilistic graph-based infant growth and development assessment model system according to any one of claims 1 to 8.

Images