Children psychological development analysis system and method based on large language model

Abstract

The invention discloses a child psychological development analysis system and method based on a big language model, and the method comprises the steps: carrying out the semantic coding and time sequence feature aggregation of the language communication data and behavior habit data of a to-be-analyzed child object through employing a natural language understanding and processing technology based on the big language model and deep learning; and automatically evaluating whether the child object to be analyzed develops forward or not according to a multi-modal fusion representation feature obtained by remarkably fusing the time sequence semantic aggregation features of the language communication data and the behavioral habit data. Thus, by means of the large language model and the deep learning technology, the system can automatically process and analyze a large amount of data, and subjectivity and workload of manual interpretation are reduced. Meanwhile, important dynamic changes and trends of languages and behaviors in the psychological development process of the children can be captured, a more detailed basis is provided for psychological development analysis and psychological counseling of the children, and the intelligence degree of psychological counseling of the children is improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent analysis, and more specifically, to a system and method for analyzing children's psychological development based on a large language model. Background Art

[0002] In today's society, with the accelerated pace of life and increasing competitive pressure, the psychological challenges faced by children are also increasing. They not only have to deal with academic pressure, but also face social influences from peers, the media, and other aspects. These factors may have a significant impact on children's self-cognition, emotional management, and social skills. Therefore, understanding the laws of children's psychological development can help to discover and prevent problems in a timely manner, provide children with appropriate guidance and help, and promote their all-round and healthy development.

[0003] However, existing methods of assessing children's psychological development usually require professionals to make subjective judgments based on standardized tests or observations, which not only increases the subjectivity of the assessment, but also limits the universal applicability of the assessment results. In other words, due to the lack of a unified quantitative standard, there may be significant differences in interpretation between different assessors, resulting in reduced comparability of the assessment results. In addition, the limitations of manual interpretation in processing large-scale data make it impossible to efficiently identify and analyze subtle but critical complex relationships between data, resulting in some important psychological patterns and behavioral trends being overlooked, thereby limiting the breadth and depth of deep understanding of children's psychological development and affecting the timeliness and effectiveness of intervention measures.

[0004] Therefore, an optimized children's psychological development analysis system is desired. Summary of the invention

[0005] In order to solve the above-mentioned technical problems, the present application is proposed. The embodiment of the present application provides a child psychological development analysis system and method based on a large language model, which uses a natural language understanding and processing technology based on a large language model and deep learning to perform semantic coding and temporal feature aggregation on the language communication data and behavioral habit data of the child object to be analyzed, so as to automatically evaluate whether the child object to be analyzed is developing positively according to the multimodal fusion representation features obtained by significantly merging the temporal semantic aggregation features of the language communication data and the behavioral habit data. In this way, using a large language model and deep learning technology, the system can automatically process and analyze a large amount of data, reducing the subjectivity and workload of manual interpretation. At the same time, it can capture the important dynamic changes and trends of language and behavior in the process of children's psychological development, provide a more detailed basis for children's psychological development analysis and psychological counseling, and improve the intelligence level of children's psychological counseling.

[0006] According to one aspect of the present application, a children's psychological development analysis system based on a large language model is provided, which includes:

[0007] A language communication behavior habit data acquisition module is used to acquire the time queue of language communication data and the time queue of behavior habit data of the child object to be analyzed;

[0008] A language communication data semantic understanding module, used for using a language communication semantic encoder based on a large model to perform semantic understanding on each language communication data in the time queue of the language communication data to obtain a time queue of semantic encoding feature vectors of the language communication data;

[0009] A behavior habit data semantic understanding module, used for using a behavior habit semantic encoder based on a large model to perform semantic understanding on each behavior habit data in the time queue of the behavior habit data to obtain a time queue of semantic encoding feature vectors of the behavior habit data;

[0010] A language behavior pattern temporal semantic aggregation module, used for inputting the time queue of the semantic encoding feature vector of the language communication data and the time queue of the semantic encoding feature vector of the behavior habit data into a feature aggregation network guided by node feature significance decay to obtain a language pattern temporal semantic aggregation representation vector and a behavior pattern temporal semantic aggregation representation vector;

[0011] A language-behavior multimodal fusion module, used for inputting the language pattern temporal semantic aggregation representation vector and the behavior pattern temporal semantic aggregation representation vector into a significant fusion network based on feature principal component query matching to obtain a language-behavior multimodal significant fusion representation vector;

[0012] The psychological development analysis result generation module is used to obtain the psychological development analysis result based on the language-behavior multimodal significant fusion representation vector, and the psychological development analysis result is used to indicate whether it is developing positively.

[0013] According to another aspect of the present application, a method for analyzing children's psychological development based on a large language model is provided, which includes:

[0014] Obtaining a time queue of language communication data and a time queue of behavior habit data of the child subject to be analyzed;

[0015] Using a language communication semantic encoder based on a large model to perform semantic understanding on each language communication data in the time queue of the language communication data to obtain a time queue of semantic encoding feature vectors of the language communication data;

[0016] Using a behavior habit semantic encoder based on a large model to semantically understand each behavior habit data in the time queue of the behavior habit data to obtain a time queue of semantic encoding feature vectors of the behavior habit data;

[0017] Input the time queue of the semantic encoding feature vector of the language communication data and the time queue of the semantic encoding feature vector of the behavior habit data into a feature aggregation network guided by node feature significance decay to obtain a language pattern temporal semantic aggregation representation vector and a behavior pattern temporal semantic aggregation representation vector;

[0018] Inputting the language pattern temporal semantic aggregation representation vector and the behavior pattern temporal semantic aggregation representation vector into a significant fusion network based on feature principal component query matching to obtain a language-behavior multimodal significant fusion representation vector;

[0019] The language-behavior multimodal significant fusion representation vector is used to obtain a psychological development analysis result, and the psychological development analysis result is used to indicate whether it is developing positively.

[0020] Compared with the prior art, the present application provides a child psychological development analysis system and method based on a large language model, which uses natural language understanding and processing technology based on a large language model and deep learning to perform semantic encoding and temporal feature aggregation on the language communication data and behavioral habit data of the child to be analyzed, respectively, so as to automatically evaluate whether the child to be analyzed is developing positively based on the multimodal fusion representation features obtained by the significant fusion between the temporal semantic aggregation features of the language communication data and the behavioral habit data. In this way, using a large language model and deep learning technology, the system can automatically process and analyze a large amount of data, reducing the subjectivity and workload of manual interpretation. At the same time, it can capture the important dynamic changes and trends of language and behavior in the process of children's psychological development, provide a more detailed basis for children's psychological development analysis and psychological counseling, and improve the intelligence level of children's psychological counseling. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0022] Figure 1 A block diagram of a children's psychological development analysis system based on a large language model according to an embodiment of the present application;

[0023] Figure 2 A data flow diagram of a children's psychological development analysis system based on a large language model according to an embodiment of the present application;

[0024] Figure 3 It is a flowchart of a method for analyzing children's psychological development based on a large language model according to an embodiment of the present application. DETAILED DESCRIPTION

[0025] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0026] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0027] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0028] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. Meanwhile, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0029] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0030] Existing methods of assessing children's psychological development usually require professionals to make subjective judgments based on standardized tests or observations, which not only increases the subjectivity of the assessment, but also limits the universal applicability of the assessment results. In other words, due to the lack of a unified quantitative standard, there may be significant differences in interpretation between different assessors, resulting in reduced comparability of the assessment results. In addition, the limitations of manual interpretation in processing large-scale data make it impossible to efficiently identify and analyze subtle but critical complex relationships between data, resulting in some important psychological patterns and behavioral trends being ignored, thereby limiting the breadth and depth of a deep understanding of children's psychological development and affecting the timeliness and effectiveness of intervention measures. Therefore, an optimized children's psychological development analysis system is desired.

[0031] In the technical solution of the present application, a children's psychological development analysis system based on a large language model is proposed. Figure 1 It is a block diagram of a children's psychological development analysis system based on a large language model according to an embodiment of the present application. Figure 2 FIG. 1 is a data flow diagram of a children's psychological development analysis system based on a large language model according to an embodiment of the present application. Figure 1 and Figure 2 As shown, according to the embodiment of the present application, the child psychological development analysis system 300 based on the large language model includes: a language communication behavior habit data acquisition module 310, which is used to obtain the time queue of the language communication data and the time queue of the behavior habit data of the child object to be analyzed; a language communication data semantic understanding module 320, which is used to use the language communication semantic encoder based on the large model to perform semantic understanding on each language communication data in the time queue of the language communication data to obtain the time queue of the semantic coding feature vector of the language communication data; a behavior habit data semantic understanding module 330, which is used to use the behavior habit semantic encoder based on the large model to perform semantic understanding on each behavior habit data in the time queue of the behavior habit data to obtain the time queue of the semantic coding feature vector of the behavior habit data; language behavior pattern time queue The sequential semantic aggregation module 340 is used to input the time queue of the semantic encoding feature vector of the language communication data and the time queue of the semantic encoding feature vector of the behavior habit data into a feature aggregation network guided by node feature significance attenuation to obtain a language pattern temporal semantic aggregation representation vector and a behavior pattern temporal semantic aggregation representation vector; the language-behavior multimodal fusion module 350 is used to input the language pattern temporal semantic aggregation representation vector and the behavior pattern temporal semantic aggregation representation vector into a significant fusion network based on feature principal component query matching to obtain a language-behavior multimodal significant fusion representation vector; the psychological development analysis result generation module 360 ​​is used to obtain a psychological development analysis result based on the language-behavior multimodal significant fusion representation vector, and the psychological development analysis result is used to indicate whether the development is positive.

[0032] In particular, the language communication behavior habit data acquisition module 310 and the language communication data semantic understanding module 320 are used to obtain the time queue of language communication data and the time queue of behavior habit data of the child object to be analyzed; and use the language communication semantic encoder based on the large model to perform semantic understanding on each language communication data in the time queue of the language communication data to obtain the time queue of the semantic encoding feature vector of the language communication data. Considering that each language communication data in the time queue of the language communication data has a semantic association between the contexts, and the large language model has been pre-trained with a large amount of corpus, it has a very powerful natural language understanding ability and can accurately capture the semantic information in the language communication data. Extract semantically rich feature vectors. Based on this, in the technical solution of the present application, a language communication semantic encoder based on the large model is used to perform semantic understanding on each language communication data in the time queue of the language communication data to obtain the time queue of the semantic encoding feature vector of the language communication data.

[0033] In particular, the behavior habit data semantic understanding module 330 is used to use a large-model-based behavior habit semantic encoder to perform semantic understanding on each behavior habit data in the time queue of the behavior habit data to obtain a time queue of semantically encoded feature vectors of the behavior habit data. Similarly, considering that each behavior habit data in the time queue of the behavior habit data also has a semantic correlation relationship between the context of the behavior information, therefore, in the technical solution of the present application, a large-model-based behavior habit semantic encoder is used to perform semantic understanding on each behavior habit data in the time queue of the behavior habit data to obtain a time queue of semantically encoded feature vectors of the behavior habit data.

[0034] In particular, the language behavior pattern temporal semantic aggregation module 340 is used to input the time queue of the semantic encoding feature vector of the language communication data and the time queue of the semantic encoding feature vector of the behavior habit data into a feature aggregation network guided by the node feature significance decay to obtain a language pattern temporal semantic aggregation representation vector and a behavior pattern temporal semantic aggregation representation vector. Considering that the language communication and the behavior habit are dynamic processes that change over time, and the importance and significance of different time points are different, in order to effectively capture such temporal dynamic features and automatically identify the key time points in the language communication and behavior habits to enhance the semantic representation ability of the entire time dimension, in the technical solution of the present application, the time queue of the semantic encoding feature vector of the language communication data and the time queue of the semantic encoding feature vector of the behavior habit data are aggregated by significant temporal features by using a feature aggregation network guided by the node feature significance decay to obtain a language pattern temporal semantic aggregation representation vector and a behavior pattern temporal semantic aggregation representation vector. In particular, the feature aggregation network guided by the node feature significance decay is a feature extraction and aggregation model based on feature significance analysis. Here, the time queue of the semantic encoding feature vector of the language communication data is taken as an example for detailed description.

[0035] In an embodiment of the present application, the time queue of the semantically encoded feature vectors of the language communication data and the time queue of the semantically encoded feature vectors of the behavior habit data are input into a feature aggregation network guided by node feature significance attenuation to obtain a language pattern temporal semantic aggregation representation vector and a behavior pattern temporal semantic aggregation representation vector, including: first, calculating the feature significance description factor of each language communication data semantic encoding feature vector in the time queue of the language communication data semantic encoding feature vector, wherein the feature significance description factor is related to the mean and variance of each language communication data semantic encoding feature vector; that is, the mean and variance of each feature in the language communication data reflect the average importance and fluctuation degree of the feature in the entire time series, and are important statistics for describing feature significance, and by calculating the relative mean and variance of each feature, the importance of the feature relative to other features can be obtained, which is conducive to highlighting key features. Next, based on the distance span between each semantic encoding feature vector of the language communication data in the time queue of the semantic encoding feature vector of the language communication data and the current semantic encoding feature vector of the language communication data, the feature significance attenuation factor of each semantic encoding feature vector of the language communication data is constructed; here, by constructing the feature significance attenuation factor, it is indicated that the influence of the language mode far away from the current moment is small as time changes, so as to highlight the importance of recent features and give them a relatively high weight. Subsequently, based on the feature significance attenuation factor of each semantic encoding feature vector of the language communication data, the feature significance description factor of each semantic encoding feature vector of the language communication data is modulated to obtain a sequence of semantic feature significance attenuation description factors of the language communication data; further, the sequence of semantic feature significance attenuation description factors of the language communication data is input into the gated mask module to obtain a sequence of semantic feature significance attenuation weight factors of the language communication data; finally, the sequence of semantic feature significance attenuation weight factors of the language communication data is used as the sequence of weights, and the weighted sum of the time queue of the semantic encoding feature vector of the language communication data is calculated to obtain the temporal semantic aggregation representation vector of the language mode.

[0036] Among them, the process of calculating the feature significance description factor of each language communication data semantic encoding feature vector in the time queue of the language communication data semantic encoding feature vector includes: respectively calculating the mean and variance of the language communication data semantic encoding feature vector to obtain the language communication data semantic feature mean and the language communication data semantic feature variance; calculating the positional difference between the language communication data semantic encoding feature vector and the language communication data semantic feature mean to obtain the language communication data semantic differential vector; calculating the fourth power of each eigenvalue in the language communication data semantic differential vector to obtain the language communication data semantic modulation difference vector; calculating the expected value of the language communication data semantic modulation difference vector to obtain the language communication data expected value; dividing the language communication data expected value by the square of the language communication data semantic feature variance to obtain the feature significance description factor corresponding to the language communication data semantic encoding feature vector.

[0037] More specifically, based on the distance span between each semantic coding feature vector of the language communication data in the time queue of the semantic coding feature vector of the language communication data and the current semantic coding feature vector of the language communication data, the process of constructing the feature significance attenuation factor of each semantic coding feature vector of the language communication data includes: extracting the maximum value of each semantic coding feature vector of the language communication data to obtain a sequence of maximum values ​​of the semantic features of the language communication data; extracting the maximum value of the semantic coding feature vector of the current language communication data to obtain the maximum value of the semantic coding feature of the current language communication data; calculating the sequence of the maximum values ​​of the semantic features of the language communication data and the maximum value of the semantic coding feature vector of the current language communication data; The maximum values ​​of the coding features are subtracted by position to obtain a sequence of language communication data semantic feature offset values; the distance span values ​​between the semantic coding feature vectors of each language communication data and the current semantic coding feature vector of the language communication data are counted to obtain a sequence of language communication data semantic distance span values, and the squares of each value in the sequence of the semantic distance span values ​​of the language communication data are calculated to obtain a sequence of language communication data semantic distance span modulation values; the sequence of the language communication data semantic feature offset values ​​is divided by the sequence of the language communication data semantic distance span modulation values ​​to obtain a feature significance attenuation factor of each language communication data semantic coding feature vector.

[0038] More specifically, the process of inputting the sequence of semantic feature significance attenuation description factors of the language communication data into a gated mask module to obtain a sequence of semantic feature significance attenuation weight factors of the language communication data includes: performing normalization processing based on a Sigmoid function on each semantic feature significance attenuation description factor of the language communication data in the sequence of semantic feature significance attenuation description factors of the language communication data to obtain a set of normalized semantic feature significance attenuation description factors of the language communication data; and inputting each normalized semantic feature significance attenuation description factor of the language communication data in the set of normalized semantic feature significance attenuation description factors of the language communication data into a gated function for masking processing to obtain a sequence of semantic feature significance attenuation weight factors of the language communication data.

[0039] In summary, in the above embodiment, the time queue of the semantically encoded feature vector of the language communication data and the time queue of the semantically encoded feature vector of the behavior habit data are input into a feature aggregation network guided by node feature significance attenuation to obtain a language pattern temporal semantic aggregation representation vector and a behavior pattern temporal semantic aggregation representation vector, including: inputting the time queue of the semantically encoded feature vector of the language communication data into a feature aggregation network guided by node feature significance attenuation, and processing with the following feature aggregation formula to obtain the language pattern temporal semantic aggregation representation vector; wherein the feature aggregation formula is:

[0040] X={x1,x2,...,x k ,...,x n}

[0041]

[0042] w i =sigmoid(k i )

[0043] ws i =mask(w i )

[0044]

[0045] Where X represents the time sequence of the semantic encoding feature vector of the language communication data, x k 、x n and x i are the kth, nth and ith language communication data semantic coding feature vectors in the time queue of the language communication data semantic coding feature vectors, max(·) is the maximum value of the extracted vector, Count(x n →x i ) represents x n With x iThe distance span between i is the feature significance attenuation factor of the semantic encoding feature vector of the i-th language communication data, x i (j) is the eigenvalue of the (j)th position of the semantic encoding feature vector of the i-th language communication data, μ i and σ i 4 are the mean and square of the variance of the semantic encoding feature vector of the i-th language communication data, k i is the descriptive factor of the semantic feature attenuation of the language communication data corresponding to the semantic encoding feature vector of the i-th language communication data, sigmoid(·) is the sigmoid function, and w i is the ith normalized language communication data semantic feature significance attenuation description factor in the set of normalized language communication data semantic feature significance attenuation description factors, mask(·) is masking processing, ws i is the i-th language communication data semantic feature significance attenuation weight factor in the sequence of the language communication data semantic feature significance attenuation weight factors, θ is a predetermined threshold, and V1 is the language pattern temporal semantic aggregation representation vector. Similarly, the encoding method of the time queue of the behavior habit data semantic encoding feature vector is also shown in the above formula.

[0046] In particular, the language-behavior multimodal fusion module 350 is used to input the language pattern temporal semantic aggregation representation vector and the behavior pattern temporal semantic aggregation representation vector into a significant fusion network based on feature principal component query matching to obtain a language-behavior multimodal significant fusion representation vector. Considering that there is a semantic correlation between the language pattern temporal semantic aggregation representation vector and the behavior pattern temporal semantic aggregation representation vector, which contains the main component features between the language and behavior pattern language features, but also some unimportant feature information. Therefore, in order to enable the model to focus on those principal component features that are highly correlated between language and behavior, and highlight the role of these key features in the fusion process, in the technical solution of the present application, the language pattern temporal semantic aggregation representation vector and the behavior pattern temporal semantic aggregation representation vector are input into a significant fusion network based on feature principal component query matching to obtain a language-behavior multimodal significant fusion representation vector. It is worth mentioning that the significant fusion network based on feature principal component query matching is a feature extraction and fusion technology based on feature sparse processing, principal component analysis and significant fusion, which is used to generate sparse and significant feature association representations between feature vectors.

[0047] In the above embodiment, the language pattern temporal semantic aggregation representation vector and the behavior pattern temporal semantic aggregation representation vector are input into a significant fusion network based on feature principal component query matching to obtain a language-behavior multimodal significant fusion representation vector, including: first, the language pattern temporal semantic aggregation representation vector and the behavior pattern temporal semantic aggregation representation vector are standardized to obtain a standardized language pattern temporal semantic aggregation representation vector and a standardized behavior pattern temporal semantic aggregation representation vector; the difference influence of different dimensions and numerical ranges is eliminated by standardization, so that each feature dimension is on the same numerical scale. Then, the sample covariance matrix of the standardized language pattern temporal semantic aggregation representation vector and the standardized behavior pattern temporal semantic aggregation representation vector is calculated to obtain a language pattern sample covariance matrix and a behavior pattern sample covariance matrix; and the language pattern sample covariance matrix and the behavior pattern sample covariance matrix are subjected to feature vector extraction based on matrix decomposition to obtain a set of language pattern principal component feature vectors and a set of behavior pattern principal component feature vectors; wherein the principal component feature usually has a stronger expression ability and can better capture the potential key semantic information contained in the original feature. Subsequently, the set of the language pattern principal component feature vectors and the set of the behavior pattern principal component feature vectors are input into the maximum approximate query matching network to obtain the set of the best matching pairs of the language pattern principal component feature vectors and the behavior pattern principal component feature vectors; that is, the two sets of sets are subjected to the maximum approximate query matching based on the feature vectors, and the most relevant semantic query matching is performed by calculating the similarity between each feature, so as to find the best matching pair. Furthermore, the best matching pairs of each language pattern principal component feature vector and the behavior pattern principal component feature vector in the set of the best matching pairs of the language pattern principal component feature vector and the behavior pattern principal component feature vector are input into the semantic fine-grained gating joint module to obtain the set of language-behavior pattern principal component fusion feature vectors; here, the gating mechanism is used to perform the fine-grained semantic joint on the best matching pairs to enhance and strengthen the expression of the semantic relevance and importance in each set of matching pairs to obtain the set of language-behavior pattern principal component fusion feature vectors. Finally, the set of the language-behavior pattern principal component fusion feature vectors is cascaded to obtain the language-behavior multimodal significant fusion representation vector.

[0048] Among them, the process of standardizing the language pattern temporal semantic aggregation representation vector and the behavior pattern temporal semantic aggregation representation vector to obtain a standardized language pattern temporal semantic aggregation representation vector and a standardized behavior pattern temporal semantic aggregation representation vector includes: respectively calculating the mean and standard deviation of the language pattern temporal semantic aggregation representation vector to obtain the mean of the language pattern temporal semantic aggregation representation feature and the standard deviation of the language pattern temporal semantic aggregation representation feature; subtracting the language pattern temporal semantic aggregation representation vector from the mean of the language pattern temporal semantic aggregation representation feature by position, and then calculating the language pattern temporal semantic aggregation representation offset vector and the language pattern temporal semantic aggregation representation offset vector. The standard deviation of the pattern temporal semantic aggregation representation feature is divided by position to obtain the standardized language pattern temporal semantic aggregation representation vector; the mean and standard deviation of the behavior pattern temporal semantic aggregation representation vector are calculated respectively to obtain the mean of the behavior pattern temporal semantic aggregation representation feature and the standard deviation of the behavior pattern temporal semantic aggregation representation feature; after subtracting the behavior pattern temporal semantic aggregation representation vector from the mean of the behavior pattern temporal semantic aggregation representation feature by position, the calculated behavior pattern temporal semantic aggregation representation offset vector is divided by the standard deviation of the behavior pattern temporal semantic aggregation representation feature to obtain the standardized behavior pattern temporal semantic aggregation representation vector.

[0049] More specifically, the process of calculating the sample covariance matrix of the standardized language pattern temporal semantic aggregation representation vector and the standardized behavior pattern temporal semantic aggregation representation vector to obtain the language pattern sample covariance matrix and the behavior pattern sample covariance matrix includes: multiplying the transpose vector of the standardized language pattern temporal semantic aggregation representation vector with the standardized language pattern temporal semantic aggregation representation vector, and then dividing the obtained standardized language pattern temporal semantic aggregation association matrix by the value obtained by subtracting one from the length of the standardized language pattern temporal semantic aggregation representation vector by position to obtain the language pattern sample covariance matrix; multiplying the transpose vector of the standardized behavior pattern temporal semantic aggregation representation vector with the standardized behavior pattern temporal semantic aggregation representation vector, and then dividing the obtained standardized behavior pattern temporal semantic aggregation association matrix by the value obtained by subtracting one from the length of the standardized behavior pattern temporal semantic aggregation representation vector by position to obtain the behavior pattern sample covariance matrix.

[0050] More specifically, the process of inputting the set of language pattern principal component feature vectors and the set of behavior pattern principal component feature vectors into the maximum approximate query matching network to obtain a set of best matching pairs of language pattern principal component feature vectors and behavior pattern principal component feature vectors includes: extracting predetermined language pattern principal component feature vectors from the set of language pattern principal component feature vectors; calculating the cosine similarity between the predetermined language pattern principal component feature vector and each behavior pattern principal component feature vector in the set of behavior pattern principal component feature vectors to obtain a set of matching query similarities; and taking the behavior pattern principal component feature vector corresponding to the largest matching query similarity in the set of matching query similarities and the predetermined language pattern principal component feature vector as the best matching pair of the predetermined language pattern principal component feature vector and the behavior pattern principal component feature vector.

[0051] More specifically, the process of inputting the best matching pairs of each language pattern principal component feature vector and behavior pattern principal component feature vector in the set of the best matching pairs of the language pattern principal component feature vector and the behavior pattern principal component feature vector into the semantic fine-grained gated joint module to obtain a set of language-behavior pattern principal component fusion feature vectors includes: respectively calculating the positional difference, positional dot product and positional addition between the best matching pairs of the language pattern principal component feature vector and the behavior pattern principal component feature vector to obtain a language-behavior principal component pattern differential vector, a language-behavior principal component pattern dot product vector and a language-behavior principal component pattern sum vector; cascading the language-behavior principal component pattern differential vector, the language-behavior principal component pattern dot product vector and the language-behavior principal component pattern sum vector and performing one-dimensional convolution encoding to obtain a language-behavior principal component pattern multi-dimensional fusion vector; performing local window-based maximum pooling processing on the language-behavior principal component pattern multi-dimensional fusion vector to obtain the language-behavior pattern principal component fusion feature vector.

[0052] In summary, in the above embodiment, the language pattern temporal semantic aggregation representation vector and the behavior pattern temporal semantic aggregation representation vector are input into a significant fusion network based on feature principal component query matching to obtain a language-behavior multimodal significant fusion representation vector, including: inputting the language pattern temporal semantic aggregation representation vector and the behavior pattern temporal semantic aggregation representation vector into a significant fusion network based on feature principal component query matching, and processing them with the following significant fusion formula to obtain the language-behavior multimodal significant fusion representation vector; wherein the significant fusion formula is:

[0053]

[0054] C1=U1Λ1U1 T

[0055] C2=U2Λ2U2 T

[0056] U1=[v 11 ,v 12 ,…,v 1m ]

[0057] Λ1=diag(λ 11 ,λ 12 ,…,λ 1m )

[0058] U2=[v 21 ,v 22 ,…,v 2m ]

[0059] Λ2=diag(λ 21 ,λ 22 ,…,λ 2m )

[0060]

[0061]

[0062] v f =[v p1 ;v p2 ;...v pm ]

[0063] Among them, V1 and V2 represent the temporal semantic aggregation representation vector of the language pattern and the temporal semantic aggregation representation vector of the behavior pattern, μ1 and μ2 are the means of the temporal semantic aggregation representation vector of the language pattern and the temporal semantic aggregation representation vector of the behavior pattern, σ1 and σ2 are the standard deviations of the temporal semantic aggregation representation vector of the language pattern and the temporal semantic aggregation representation vector of the behavior pattern, respectively. 1norm and V 2norm are the temporal semantic aggregation representation vector of the standardized language pattern and the temporal semantic aggregation representation vector of the standardized behavior pattern, respectively, 1norm T and V 2norm T V 1norm and V 2norm The transposed vector of , n1 and n2 are the number of eigenvalues ​​in the standardized language pattern temporal semantic aggregation representation vector and the standardized behavior pattern temporal semantic aggregation representation vector, C1 and C2 are the language pattern sample covariance matrix and the behavior pattern sample covariance matrix, U1 and U2 are the language pattern principal component orthogonal matrix and the behavior pattern principal component orthogonal matrix, Λ1 and Λ2 are the language pattern diagonal matrix and the behavior pattern diagonal matrix, diag(λ11 ,λ 12 ,…,λ 1m ) is the matrix whose diagonal elements are λ 11 ,λ 12 ,…,λ 1m The language pattern diagonal matrix, λ 11 ,λ 12 ,…,λ 1m are the weight values ​​of the principal component feature vectors of each language mode, diag(λ 21 ,λ 22 ,…,λ 2m ) is the matrix whose diagonal elements are λ 21 ,λ 22 ,…,λ 2m The behavior pattern diagonal matrix, λ 21 ,λ 22 ,…,λ 2m are the weight values ​​of the principal component eigenvectors of each behavior mode, U1 T and U2 T are the transposed matrices of U1 and U2, respectively, 11 ,v 12 ,…,v 1m is the principal component feature vector of each language pattern in the set of the principal component feature vectors of the language pattern, v 21 ,v 22 ,…,v 2m is each behavior pattern principal component feature vector in the set of the behavior pattern principal component feature vectors, <v 1i ·v 2j >To calculate the v 1i and v 2j The inner product of the vectors between ||·|| is the norm of the calculated vector, To return the j value corresponding to the maximum value, k is the maximum approximate matching value. ⊙ and are position difference, position dot multiplication and position addition respectively, [·,·,·] is cascade processing, conv1D(·) is a one-dimensional convolutional encoding operation, MaxPool(·) is a maximum pooling operation, v pi is the i-th language-behavior pattern principal component fusion feature vector in the set of the language-behavior pattern principal component fusion feature vectors, m is the number of feature vectors in the set of the language-behavior pattern principal component fusion feature vectors, v f It is the language-behavior multimodal significant fusion representation vector.

[0064] In particular, the psychological development analysis result generation module 360 ​​is used to obtain the psychological development analysis result based on the language-behavior multimodal significant fusion representation vector, and the psychological development analysis result is used to indicate whether it is developing positively. In a specific example of the present application, the language-behavior multimodal significant fusion representation vector is input into the psychological development analysis module based on the classifier to obtain the psychological development analysis result, and the psychological development analysis result is used to indicate whether it is developing positively. That is, the language-behavior multimodal significant fusion representation vector obtained by significantly fusion of the language pattern temporal semantic aggregation representation vector and the behavior pattern temporal semantic aggregation representation vector is used for classification processing, so as to automatically evaluate whether the child object to be analyzed is developing positively. In this way, using a large language model and deep learning technology, the system can automatically process and analyze a large amount of data, reducing the subjectivity and workload of manual interpretation. At the same time, it can capture the important dynamic changes and trends in the process of children's psychological development, provide a more detailed basis for psychological counseling, and improve the intelligent degree of children's psychological counseling.

[0065] Preferably, considering that the language pattern temporal semantic aggregation representation vector and the behavior pattern temporal semantic aggregation representation vector respectively represent the global time-domain guided aggregation features of the language communication data and the behavior habit data based on the local time-domain semantic encoding feature temporal node significance attenuation, when performing significant fusion based on feature principal component query matching, due to the difference in feature principal component query matching significance caused by the difference in local-global time-domain aggregation distribution of the source semantics, it will lead to semantic aggregation fine-grained alignment conflicts, thereby causing the loss of aggregation key information, resulting in poor iterative consistency of the convergence mapping of the language-behavior multimodal significant fusion representation vector to the predetermined class probability category representation in the probability space, thereby reducing the accuracy of the classification results.

[0066] Based on this, in the preferred example, when the language-behavior multimodal significant fusion representation vector is input into the classifier-based psychological development analysis module to obtain the psychological development analysis result, the language-behavior multimodal significant fusion representation vector is optimized. Specifically, the optimization process includes the following steps: for the first eigenvalue v of the language-behavior multimodal significant fusion representation vector, i and the second eigenvalue v j, calculate the absolute value of the difference divided by the sum of the absolute values ​​and then correct the relative probability value, wherein the correction makes the relative probability value greater than zero and less than one; calculate the first product of the first eigenvalue multiplied by the relative probability value, the second product of the second eigenvalue multiplied by the absolute value of the difference between the relative probability value and one, the first quotient of the first eigenvalue divided by the absolute value of the difference between the relative probability value and one, and the second quotient of the second eigenvalue divided by the relative probability value; after adding the first product and the second product, further subtract the difference between the first quotient and the second quotient to obtain the first eigenvalue v i and the second eigenvalue v j The corresponding eigenvalue m of the corresponding optimization matrix i,j ; and, multiplying the optimization matrix with the language-behavior multimodal significant fusion representation vector to obtain an optimized language-behavior multimodal significant fusion representation vector.

[0067] Specifically, it is expressed as:

[0068]

[0069] m i,j ∈M,v i ∈V,v j(j≠i) ∈V

[0070]

[0071] Where V represents the language-behavior multimodal significant fusion representation vector, v i and v j They represent the first eigenvalue and the second eigenvalue of the language-behavior multimodal significant fusion representation vector, respectively, and p i,j Represents the relative probability value, m i,j represents the corresponding eigenvalue of the optimization matrix, M represents the optimization matrix, represents matrix multiplication, and V' represents the optimized language-behavior multimodal significant fusion representation vector.

[0072] Among them, the first eigenvalue v i and the second eigenvalue v j The corresponding relative probability value p i,j It is expressed as:

[0073]

[0074] Among them, v i and v j denote the first eigenvalue and the second eigenvalue of the language-behavior multimodal significant fusion representation vector, ε denotes a predetermined hyperparameter, and p i,jFinally, in the preferred example, the optimized language-behavior multimodal significant fusion representation vector is input into a classifier-based psychological development analysis module to obtain a psychological development analysis result.

[0075] Here, for the region-boundary integral relationship of the high-dimensional feature manifold of the language-behavior multimodal significant fusion representation vector, by defining the probability distribution of the feature set of the language-behavior multimodal significant fusion representation vector as a manifold constraint boundary, the high-dimensional manifold of the feature set of the language-behavior multimodal significant fusion representation vector, which is composed of eigenvalue pairs, is approached to avoid the mapping ambiguity of the diversified feature representation of the language-behavior multimodal significant fusion representation vector to the local manifold representation in the high-dimensional feature classification regression convergence space, and the execution iteration consistency of each local feature distribution in the mapping task is improved to improve the feature classification regression convergence effect, and improve the accuracy of the psychological development analysis results obtained by the language-behavior multimodal significant fusion feature vector input based on the psychological development analysis module of the classifier. In this way, using a large language model and deep learning technology, the system can automatically process and analyze a large amount of data, reducing the subjectivity and workload of manual interpretation. At the same time, it can capture the important dynamic changes and trends in the process of children's psychological development, provide a more detailed basis for psychological counseling, and thus improve the intelligent level of children's psychological counseling.

[0076] As described above, the child psychological development analysis system 300 based on the large language model according to the embodiment of the present application can be implemented in various wireless terminals, such as a server with a child psychological development analysis algorithm based on the large language model. In a possible implementation, the child psychological development analysis system 300 based on the large language model according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the child psychological development analysis system 300 based on the large language model can be a software module in the operating system of the wireless terminal, or can be an application developed for the wireless terminal; of course, the child psychological development analysis system 300 based on the large language model can also be one of the many hardware modules of the wireless terminal.

[0077] Alternatively, in another example, the child psychological development analysis system 300 based on a large language model and the wireless terminal may be separate devices, and the child psychological development analysis system 300 based on a large language model may be connected to the wireless terminal via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0078] Furthermore, a method for analyzing children's psychological development based on a large language model is also provided.

[0079] Figure 3 FIG. 1 is a flow chart of a method for analyzing children's psychological development based on a large language model according to an embodiment of the present application. Figure 3 As shown, according to the method for analyzing the psychological development of children based on a large language model according to an embodiment of the present application, the method includes the following steps: S1, obtaining a time queue of language communication data and a time queue of behavioral habit data of the child to be analyzed; S2, using a language communication semantic encoder based on a large model to perform semantic understanding on each language communication data in the time queue of the language communication data to obtain a time queue of semantically encoded feature vectors of the language communication data; S3, using a behavioral habit semantic encoder based on a large model to perform semantic understanding on each behavioral habit data in the time queue of the behavioral habit data to obtain a time queue of semantically encoded feature vectors of the behavioral habit data; S4, converting the language The time queue of the semantically encoded feature vectors of the communication data and the time queue of the semantically encoded feature vectors of the behavioral habit data are input into a feature aggregation network guided by node feature significance attenuation to obtain a language pattern temporal semantic aggregation representation vector and a behavior pattern temporal semantic aggregation representation vector; S5, the language pattern temporal semantic aggregation representation vector and the behavior pattern temporal semantic aggregation representation vector are input into a significant fusion network based on feature principal component query matching to obtain a language-behavior multimodal significant fusion representation vector; S6, the language-behavior multimodal significant fusion representation vector is used to obtain a psychological development analysis result, and the psychological development analysis result is used to indicate whether it is developing positively.

[0080] In summary, the method for analyzing the psychological development of children based on a large language model according to the embodiment of the present application is explained, which uses a natural language understanding and processing technology based on a large language model and deep learning to perform semantic coding and temporal feature aggregation on the language communication data and behavioral habit data of the child object to be analyzed, respectively, so as to automatically evaluate whether the child object to be analyzed is developing positively based on the multimodal fusion representation features obtained by significantly fusion between the temporal semantic aggregation features of the language communication data and the behavioral habit data. In this way, using a large language model and deep learning technology, the system can automatically process and analyze a large amount of data, reducing the subjectivity and workload of manual interpretation. At the same time, it can capture the important dynamic changes and trends of language and behavior in the process of children's psychological development, provide a more detailed basis for children's psychological development analysis and psychological counseling, and improve the intelligence level of children's psychological counseling.

[0081] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A children's psychological development analysis system based on a large language model, characterized in that: include: A language communication behavior habit data acquisition module is used to acquire the time queue of language communication data and the time queue of behavior habit data of the child object to be analyzed; A language communication data semantic understanding module, used for using a language communication semantic encoder based on a large model to perform semantic understanding on each language communication data in the time queue of the language communication data to obtain a time queue of semantic encoding feature vectors of the language communication data; A behavior habit data semantic understanding module, used for using a behavior habit semantic encoder based on a large model to perform semantic understanding on each behavior habit data in the time queue of the behavior habit data to obtain a time queue of semantic encoding feature vectors of the behavior habit data; A language behavior pattern temporal semantic aggregation module, used for inputting the time queue of the semantic encoding feature vector of the language communication data and the time queue of the semantic encoding feature vector of the behavior habit data into a feature aggregation network guided by node feature significance decay to obtain a language pattern temporal semantic aggregation representation vector and a behavior pattern temporal semantic aggregation representation vector; A language-behavior multimodal fusion module, used for inputting the language pattern temporal semantic aggregation representation vector and the behavior pattern temporal semantic aggregation representation vector into a significant fusion network based on feature principal component query matching to obtain a language-behavior multimodal significant fusion representation vector; The psychological development analysis result generation module is used to obtain the psychological development analysis result based on the language-behavior multimodal significant fusion representation vector, and the psychological development analysis result is used to indicate whether it is developing positively.

2. The children's psychological development analysis system based on a large language model according to claim 1 is characterized in that: The language behavior pattern temporal semantic aggregation module includes: A language communication data semantic feature significance description factor calculation unit, used to calculate the feature significance description factor of each language communication data semantic encoding feature vector in the time queue of the language communication data semantic encoding feature vector, wherein the feature significance description factor is related to the mean and variance of each language communication data semantic encoding feature vector; A language communication data semantic feature significance attenuation factor construction unit, used to construct a feature significance attenuation factor of each language communication data semantic encoding feature vector based on the distance span between each language communication data semantic encoding feature vector in the time queue of the language communication data semantic encoding feature vector and the current language communication data semantic encoding feature vector; A language communication data semantic feature significance attenuation factor modulation unit, used for modulating the feature significance description factor of each language communication data semantic encoding feature vector based on the feature significance attenuation factor of each language communication data semantic encoding feature vector to obtain a sequence of language communication data semantic feature significance attenuation description factors; A language communication data semantic feature significance attenuation factor masking unit, used for inputting the sequence of language communication data semantic feature significance attenuation description factors into a gated masking module to obtain a sequence of language communication data semantic feature significance attenuation weighting factors; The language pattern temporal semantic aggregation representation unit is used to calculate the weighted sum of the time queue of the semantic encoding feature vector of the language communication data using the sequence of the semantic feature significance attenuation weight factors of the language communication data as the sequence of weights to obtain the language pattern temporal semantic aggregation representation vector.

3. The children's psychological development analysis system based on a large language model according to claim 2 is characterized in that: The language communication data semantic feature significance description factor calculation unit is used to: Respectively calculating the mean and variance of the semantic encoding feature vector of the language communication data to obtain the mean of the semantic feature of the language communication data and the variance of the semantic feature of the language communication data; Calculating the positional difference between the semantic encoding feature vector of the language communication data and the mean value of the semantic feature of the language communication data to obtain a semantic differential vector of the language communication data; Calculating the fourth power of each eigenvalue in the language communication data semantic difference vector to obtain the language communication data semantic modulation difference vector; Calculating the expected value of the semantic modulation difference vector of the language communication data to obtain the expected value of the language communication data; The expected value of the language communication data is divided by the square of the variance of the semantic features of the language communication data to obtain a feature significance description factor corresponding to the semantic encoding feature vector of the language communication data.

4. The children's psychological development analysis system based on a large language model according to claim 3 is characterized in that: The language communication data semantic feature significance attenuation factor construction unit is used to: Extracting the maximum value among the semantic coding feature vectors of each of the language communication data to obtain a sequence of maximum values ​​of the semantic features of the language communication data; Extracting the maximum value of the semantic coding feature vector of the current language communication data to obtain the maximum value of the semantic coding feature of the current language communication data; Calculate the positional subtraction between the sequence of maximum values ​​of the semantic features of the language communication data and the maximum value of the semantic coding feature of the current language communication data to obtain a sequence of offset values ​​of the semantic features of the language communication data; Counting the distance span values ​​between the semantic encoding feature vectors of each language communication data and the current semantic encoding feature vector of the language communication data to obtain a sequence of semantic distance span values ​​of the language communication data, and calculating the squares of each value in the sequence of semantic distance span values ​​of the language communication data to obtain a sequence of semantic distance span modulation values ​​of the language communication data; The sequence of the language communication data semantic feature offset values ​​and the sequence of the language communication data semantic distance span modulation values ​​are divided by position to obtain feature significance attenuation factors of the semantic encoding feature vectors of each language communication data.

5. The children's psychological development analysis system based on a large language model according to claim 4 is characterized in that: The language communication data semantic feature significance attenuation factor mask unit is used to: Performing normalization processing based on a Sigmoid function on each of the language communication data semantic feature significance attenuation description factors in the sequence of the language communication data semantic feature significance attenuation description factors to obtain a set of normalized language communication data semantic feature significance attenuation description factors; Each normalized language communication data semantic feature significance attenuation description factor in the set of normalized language communication data semantic feature significance attenuation description factors is input into a gating function for masking processing to obtain a sequence of language communication data semantic feature significance attenuation weight factors.

6. The children's psychological development analysis system based on a large language model according to claim 5 is characterized in that: The language-behavior multimodal fusion module includes: A language behavior pattern temporal semantic feature standardization unit, used for standardizing the language pattern temporal semantic aggregation representation vector and the behavior pattern temporal semantic aggregation representation vector to obtain a standardized language pattern temporal semantic aggregation representation vector and a standardized behavior pattern temporal semantic aggregation representation vector; A language behavior pattern sample covariance matrix calculation unit, used to calculate the sample covariance matrix of the standardized language pattern temporal semantic aggregation representation vector and the standardized behavior pattern temporal semantic aggregation representation vector to obtain a language pattern sample covariance matrix and a behavior pattern sample covariance matrix; A language-behavior pattern principal component feature extraction unit, used for performing matrix decomposition-based feature vector extraction on the language pattern sample covariance matrix and the behavior pattern sample covariance matrix to obtain a set of language pattern principal component feature vectors and a set of behavior pattern principal component feature vectors; A language-behavior maximum approximation query unit, used for inputting the set of the language pattern principal component feature vectors and the set of the behavior pattern principal component feature vectors into a maximum approximation query matching network to obtain a set of best matching pairs of the language pattern principal component feature vectors and the behavior pattern principal component feature vectors; A language-behavior pattern principal component fusion unit, used for inputting each best matching pair of the language pattern principal component feature vector and the behavior pattern principal component feature vector in the set of best matching pairs of the language pattern principal component feature vector and the behavior pattern principal component feature vector into a semantic fine-grained gating joint module to obtain a set of language-behavior pattern principal component fusion feature vectors; A language-behavior multimodal significant fusion unit, used for cascading the set of the language-behavior pattern principal component fusion feature vectors to obtain the language-behavior multimodal significant fusion representation vector; Wherein, the language behavior pattern sample covariance matrix calculation unit is used to: After multiplying the transposed vector of the standardized language pattern temporal semantic aggregation representation vector by the standardized language pattern temporal semantic aggregation representation vector, the obtained standardized language pattern temporal semantic aggregation association matrix is ​​divided by the value obtained by subtracting one from the length of the standardized language pattern temporal semantic aggregation representation vector by position to obtain the language pattern sample covariance matrix; After multiplying the transposed vector of the standardized behavior pattern temporal semantic aggregation representation vector by the standardized behavior pattern temporal semantic aggregation representation vector, the obtained standardized behavior pattern temporal semantic aggregation association matrix is ​​divided by the value obtained by subtracting one from the length of the standardized behavior pattern temporal semantic aggregation representation vector to obtain the behavior pattern sample covariance matrix.

7. The children's psychological development analysis system based on a large language model according to claim 6 is characterized in that: The language behavior pattern temporal semantic feature standardization unit is used to: Calculating the mean and standard deviation of the language pattern temporal semantic aggregation representation vector respectively to obtain the language pattern temporal semantic aggregation representation feature mean and the language pattern temporal semantic aggregation representation feature standard deviation; After positionally subtracting the language pattern temporal semantic aggregation representation vector from the language pattern temporal semantic aggregation representation feature mean, positionally dividing the calculated language pattern temporal semantic aggregation representation offset vector and the language pattern temporal semantic aggregation representation feature standard deviation to obtain the standardized language pattern temporal semantic aggregation representation vector; Calculating the mean and standard deviation of the behavior pattern temporal semantic aggregation representation vector respectively to obtain the behavior pattern temporal semantic aggregation representation feature mean and the behavior pattern temporal semantic aggregation representation feature standard deviation; After positionally subtracting the behavior pattern temporal semantic aggregation representation vector from the mean of the behavior pattern temporal semantic aggregation representation features, the calculated behavior pattern temporal semantic aggregation representation offset vector is positionally divided by the standard deviation of the behavior pattern temporal semantic aggregation representation features to obtain the standardized behavior pattern temporal semantic aggregation representation vector.

8. The children's psychological development analysis system based on a large language model according to claim 7 is characterized in that: The language-behavior maximum approximation query unit is used to: Extracting a predetermined language pattern principal component feature vector from the set of language pattern principal component feature vectors; Calculating the cosine similarity between the predetermined language pattern principal component feature vector and each behavior pattern principal component feature vector in the set of behavior pattern principal component feature vectors to obtain a set of matching query similarities; The behavior pattern principal component feature vector corresponding to the largest matching query similarity in the set of matching query similarities and the predetermined language pattern principal component feature vector are used as the best matching pair of the predetermined language pattern principal component feature vector and the behavior pattern principal component feature vector.

9. The children's psychological development analysis system based on a large language model according to claim 8 is characterized in that: The language-behavior pattern principal component fusion unit is used to: Respectively calculating the positional difference, positional dot product and positional addition between the best matching pairs of the language pattern principal component feature vector and the behavior pattern principal component feature vector to obtain a language-behavior principal component pattern difference vector, a language-behavior principal component pattern dot product vector and a language-behavior principal component pattern sum vector; The language-behavior principal component pattern difference vector, the language-behavior principal component pattern dot product vector and the language-behavior principal component pattern sum vector are cascaded and then one-dimensional convolutional coding is performed to obtain a language-behavior principal component pattern multi-dimensional fusion vector; The multi-dimensional fusion vector of the language-behavior principal component pattern is subjected to a local window-based maximum pooling process to obtain the language-behavior pattern principal component fusion feature vector.

10. A method for analyzing children's psychological development based on a large language model, characterized in that: include: Obtaining a time queue of language communication data and a time queue of behavior habit data of the child subject to be analyzed; Using a language communication semantic encoder based on a large model to perform semantic understanding on each language communication data in the time queue of the language communication data to obtain a time queue of semantic encoding feature vectors of the language communication data; Using a behavior habit semantic encoder based on a large model to semantically understand each behavior habit data in the time queue of the behavior habit data to obtain a time queue of semantic encoding feature vectors of the behavior habit data; Input the time queue of the semantic encoding feature vector of the language communication data and the time queue of the semantic encoding feature vector of the behavior habit data into a feature aggregation network guided by node feature significance decay to obtain a language pattern temporal semantic aggregation representation vector and a behavior pattern temporal semantic aggregation representation vector; Inputting the language pattern temporal semantic aggregation representation vector and the behavior pattern temporal semantic aggregation representation vector into a significant fusion network based on feature principal component query matching to obtain a language-behavior multimodal significant fusion representation vector; The language-behavior multimodal significant fusion representation vector is used to obtain a psychological development analysis result, and the psychological development analysis result is used to indicate whether it is developing positively.

Images