Encrypted communication system for non-suicide self-injury behavior intervention of depressive teenagers

Abstract

The invention relates to the technical field of depression teenager behavior intervention, and discloses an encryption communication system for depression teenager non-suicide self-injury behavior intervention, and the system comprises a data collection module which is used for obtaining communication behavior data and emotion state parameters, and setting an encryption communication interval; the behavior analysis module can divide behavior nodes and generate interaction feature vectors; the dynamic evaluation module can extract high-risk indexes and establish encryption intervention rules; the channel optimization module is responsible for identifying the security level and calculating the data fluctuation density; the key management module can derive an optimal key threshold value and generate a key deviation sequence; and the strategy execution module is used for integrating and forming an intervention scheme. According to the system, real-time monitoring and accurate intervention of self-injury behaviors of the depressed teenagers are realized through multi-dimensional data acquisition, dynamic behavior analysis and intelligent encryption intervention, the intervention timeliness, the individuation level and the data security are improved, and an efficient and safe technical scheme is provided for mental health intervention of the teenagers.

Description

Technical Field

[0001] The present invention relates to the technical field of behavioral intervention for depressed adolescents, and in particular to an encrypted communication system for intervention in non-suicidal self-harming behaviors of depressed adolescents. Background Art

[0002] With increasing attention being paid to adolescent mental health issues, non-suicidal self-injury (NSSI) among depressed adolescents has become a significant public health issue worldwide. Such behaviors not only directly harm adolescents' physical health but can also exacerbate psychological problems, increase suicide risk, and place a heavy burden on families and society. Currently, interventions for NSSI in depressed adolescents primarily rely on traditional methods such as psychological counseling and medication. However, these approaches are often plagued by issues such as delayed intervention, insufficient personalization, and poor data security.

[0003] In terms of intervention timeliness, traditional intervention models rely on periodic consultations or passive responses after crises occur, making it difficult to capture real-time signals of high-risk behaviors exposed by adolescents in daily communication scenarios. For example, data such as the content and frequency of adolescents' text messages on social media and instant messaging often reveal early signs of mood swings and self-harm tendencies, but existing technologies lack systems capable of collecting, analyzing, and implementing interventions based on this dynamic data in real time.

[0004] When it comes to personalized intervention, the behavioral patterns and psychological states of different depressed adolescents vary significantly, making it difficult for traditional "one-size-fits-all" intervention strategies to accurately match individual needs. For example, some adolescents may exhibit increased mood swings and risk of self-harm during specific time periods (such as late at night), but existing systems are unable to dynamically adjust intervention plans based on individual behavioral characteristics.

[0005] Data security and privacy protection are another key challenge. The communication data and emotional state parameters of depressed adolescents are highly sensitive information. Traditional communication systems are prone to data leakage during transmission and storage, leading adolescents to resist participating in interventions due to privacy concerns. Furthermore, existing encryption technologies are mostly general-purpose solutions, lacking dynamic encryption mechanisms that are deeply integrated with behavioral intervention scenarios, making it difficult to flexibly adjust intervention strategies while ensuring data security.

[0006] In existing technologies, although some studies have attempted to identify self-harm risks through data analysis, the following defects are common: First, the data collection dimension is single, focusing only on one aspect of emotional state or communication behavior, and lacking the fusion analysis of multi-dimensional data; second, encryption technology and intervention strategy are independent of each other, and encryption strategy cannot be dynamically optimized according to real-time behavior analysis results, resulting in the inability to form a synergistic effect between intervention measures and data security assurance; third, there is a lack of a closed-loop intervention execution mechanism, and it is impossible to automatically generate and execute personalized intervention plans based on real-time data fluctuations, resulting in insufficient intervention efficiency and accuracy.

[0007] Therefore, there is an urgent need for an integrated system that can realize real-time multi-dimensional data collection, dynamic behavior analysis, and intelligent encryption intervention to address the problems of intervention lag, insufficient personalization, and data security risks in existing technologies, and provide more effective technical support for the precise intervention of NSSI in depressed adolescents. Summary of the Invention

[0008] The purpose of the present invention is to provide an encrypted communication system for intervening in non-suicidal self-harm behaviors of depressed adolescents, so as to solve the problems raised in the above background technology.

[0009] To achieve the above objectives, the present invention provides the following technical solution: an encrypted communication system for intervening in non-suicidal self-injurious behaviors in depressed adolescents, the system comprising:

[0010] A data acquisition module is used to obtain communication behavior data and emotional state parameters of a target user authorized by the target user, and set an encrypted communication interval that matches the user's behavior characteristics, wherein the encrypted communication interval is the obtained data to be encrypted;

[0011] The behavior analysis module is used to divide the encrypted communication interval into multiple behavior nodes, perform pattern recognition on the communication data of each behavior node, and generate the interaction feature vector corresponding to the behavior node;

[0012] A dynamic assessment module is used to extract high-risk behavior indicators from the interaction feature vector, establish encryption intervention rules associated with the behavior nodes, and obtain key control parameters corresponding to the encryption intervention rules;

[0013] The channel optimization module is used to identify the communication security level in the key control parameters, dynamically encrypt high-risk behavior indicators based on the communication security level, and calculate the data fluctuation density of each behavior node under different encryption strategies;

[0014] The key management module is used to derive the optimal key threshold according to the data fluctuation density and generate a key deviation sequence by comparing the current communication data density with the optimal key threshold;

[0015] The policy execution module is used to parse the key deviation sequence and integrate the key deviation sequence into a communication intervention execution plan based on the data fluctuation characteristics of the behavior node.

[0016] Preferably, the implementation of the behavior analysis module includes: constructing a user feature library corresponding to the behavior node, the user feature library including communication behavior data and behavior parameter vectors mapped to emotional state parameters;

[0017] Perform similar behavior matching on the behavior parameter vectors, and divide the behavior parameter vectors into behavior cluster groups according to the matching results; extract the distribution center points of the communication data from the behavior cluster groups, and set the distribution center points as behavior nodes.

[0018] Preferably, the behavior clustering groups for dividing the behavior parameter vectors further include:

[0019] According to the communication frequency and emotion fluctuation index in the behavior parameter vector, the communication duration, interaction interval and emotion intensity parameters are extracted, and the behavior feature label is generated based on the communication duration, interaction interval and emotion intensity parameters;

[0020] The behavioral feature labels are associated with the behavioral parameter vectors. By calculating the behavioral similarity between the feature labels, the parameter vectors with similarity higher than the preset behavioral threshold are screened to form a behavioral cluster group.

[0021] Preferably, the implementation method of generating the interaction feature vector corresponding to the behavior node includes:

[0022] For each behavior node, according to the time series distribution of the behavior node in the encrypted communication interval, the data fluctuation parameters of the behavior node within the preset period are obtained, and the communication fluctuation coefficient of the node is calculated;

[0023] When the communication fluctuation coefficient exceeds the first risk threshold, the node is marked as a high-risk node, and its communication data is extracted to form an interactive feature vector; when the communication fluctuation coefficient is lower than the first risk threshold, the node is marked as a safe node, and the communication data of the node's adjacent nodes are density-superimposed, and the superimposed data is reconstructed into an interactive feature vector.

[0024] Preferably, the implementation of the dynamic evaluation module includes:

[0025] Separate the communication data ratio, abnormal interaction ratio, and emotion fluctuation parameters from the interaction feature vector, and generate encrypted intervention rules for the behavior nodes based on the communication data ratio, abnormal interaction ratio, and emotion fluctuation parameters;

[0026] If the number of behavior nodes covered by the current encryption intervention rule is less than the preset behavior threshold, the interaction feature vectors of adjacent behavior nodes are traversed, and the communication indicators not included in the encryption intervention rules of the adjacent nodes are added to the current rule.

[0027] Preferably, the channel optimization module is implemented by: obtaining a timing parameter of a key validity period and a fluctuation parameter of data encryption strength in a communication security level;

[0028] An encryption state network associated with timing parameters and fluctuation parameters is constructed. According to the switching probability of various paths in the encryption state network, the data fluctuation density under different encryption strategies is determined.

[0029] Preferably, building an encrypted state network further includes:

[0030] Identify the periodic pattern of the timing parameters, and if the current periodic pattern completely matches the preset communication period, set the timing parameters as the starting point of the encrypted state network;

[0031] Calculate the switching correlation between the timing parameters and the fluctuation parameters, and generate the intermediate points and end points of the encrypted state network in descending order of switching correlation;

[0032] The state of the termination point is reversely verified. When the switching correlation of the termination point is lower than the preset switching threshold, it is output as the final link of the encrypted state network.

[0033] Preferably, the implementation of calculating the data fluctuation density includes:

[0034] Statistically calculate the mean value and range of the timing parameters of each terminal point in the encrypted state network, and calculate the global covariance of all node parameters;

[0035] The mean value of the timing parameters of a single termination point is subtracted from the mean value of the timing parameters of the adjacent nodes, and the difference is divided by the global covariance to obtain the timing density coefficient; at the same time, the ratio of the fluctuation parameter range to the global covariance is calculated, and the weighted sum of the ratio and the timing density coefficient is taken as the data fluctuation density of the node.

[0036] Preferably, the implementation of deriving the optimal key threshold includes:

[0037] Extracting the encryption mode in the historical data that is closest to the current data fluctuation density, and calculating the Euclidean distance between the data fluctuation density in the closest encryption mode and the current data fluctuation density in the time series distribution as the first key reference value;

[0038] Counting the difference in the number of peak points between the data fluctuation density in the closest encryption mode and the current data fluctuation density, and using the difference as a reference value for the second key;

[0039] Based on the linear combination of the first key reference value and the second key reference value, an optimal key threshold in the preset key threshold table is matched.

[0040] Preferably, the implementation of the policy execution module includes: dividing the data fluctuation domain and the reverse fluctuation domain according to the data fluctuation trend of each node in the key deviation sequence;

[0041] The convergence frequency of key deviation in the forward fluctuation domain and the diffusion frequency of key deviation in the reverse fluctuation domain are extracted, and the two are dynamically reconciled according to the communication weight of the behavior node to generate the configuration parameters of the communication intervention execution plan.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] The data collection module captures target users' communication behavior and emotional state parameters, and sets encrypted communication intervals that match their behavioral characteristics, enabling real-time capture of multi-dimensional dynamic data on adolescents. This design transcends the limitations of traditional intervention methods, which rely solely on single-source data collection. It can more comprehensively reflect users' true state and provide a rich and accurate data foundation for subsequent behavioral analysis. For example, by integrating parameters such as communication frequency, interaction intervals, and emotional intensity, the system can accurately identify behavioral differences between different users, laying the foundation for personalized intervention.

[0044] The behavior analysis module achieves in-depth structured analysis of communication data by building a user feature library, dividing behavior clusters, and setting behavior nodes. This module not only categorizes data with similar characteristics by matching similar behaviors, but also extracts distribution centers as behavior nodes, transforming complex communication data into identifiable and analyzable structured units. This hierarchical analysis mechanism helps the system quickly locate key behavior nodes. For example, behavioral feature labels generated by parameters such as communication duration and mood swing index can accurately identify high-risk behavior patterns that may pose a risk of self-harm, providing clear targets for subsequent dynamic assessments.

[0045] The dynamic assessment module extracts high-risk behavior indicators from interaction feature vectors and establishes encrypted intervention rules. It also supplements communication indicators by traversing adjacent nodes, ensuring the comprehensiveness and dynamic adaptability of intervention rules. This module dynamically adjusts intervention strategies based on real-time data. For example, when the number of behavior nodes covered by the current rules is insufficient, it automatically integrates information from adjacent nodes to avoid intervention loopholes caused by incomplete data. This mechanism enables the system to respond promptly to behavioral changes, improving the timeliness and targeted nature of interventions. For example, identifying an abnormal communication fluctuation coefficient can quickly trigger the marking of high-risk nodes and the generation of intervention rules.

[0046] The channel optimization module achieves a deep integration of encryption strategies and behavioral characteristics by constructing an encryption state network and calculating data fluctuation density. This module not only considers the timing parameters of key validity periods and the fluctuation parameters of data encryption strength, but also dynamically optimizes encryption strategies through switching probability and correlation analysis. For example, when an increase in data fluctuation density is detected, the system automatically increases encryption strength and adjusts key validity periods. This ensures that intervention information is accurately and securely transmitted to the intended user while maintaining data security, avoiding security vulnerabilities or intervention delays caused by fixed encryption strategies.

[0047] The key management module dynamically adapts key parameters to data fluctuations by deriving optimal key thresholds and generating key deviation sequences. Based on comparative analysis of historical and real-time data, this module accurately calculates the optimal key thresholds that adapt to current behavioral characteristics. For example, it matches a pre-set threshold table using a linear combination of Euclidean distance and peak point difference, enabling key parameters to adjust in real time to data fluctuations. This dynamic key management mechanism not only improves the security of data encryption but also provides precise control parameters for the policy execution module, ensuring the effectiveness of intervention plans.

[0048] The policy execution module generates personalized communication intervention execution plans by dividing fluctuation domains and reconciling convergence and diffusion frequencies. This module dynamically adjusts the configuration parameters of the intervention strategy based on the data fluctuation trends of each behavioral node, for example, adopting different intervention levels and methods for positive and negative fluctuation domains. This differentiated intervention mechanism enables the system to provide precise intervention measures tailored to the behavioral characteristics of different users, such as strengthening real-time monitoring and psychological counseling for nodes with severe emotional fluctuations, thereby improving the accuracy and effectiveness of interventions. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 This is a working principle diagram of the encrypted communication system for intervention of non-suicidal self-injurious behavior of depressed adolescents according to the present invention;

[0050] Figure 2 Flowchart of the method for dividing behavior nodes;

[0051] Figure 3 Design diagram for feature label association and cluster group screening;

[0052] Figure 4 Flowchart for high-risk node identification and feature vector reconstruction. DETAILED DESCRIPTION

[0053] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0054] See also Figure 1-Figure 4 The present invention relates to an encrypted communication system for intervention in non-suicidal self-harm behaviors in depressed adolescents. The system uses multiple modules working together to monitor the target user's communication behavior, conduct risk assessments, and implement encrypted interventions. Specifically, the system includes the following steps:

[0055] Data collection module: obtains the target user's communication behavior data (such as communication frequency, duration, interaction objects, etc.) and emotional state parameters (such as heart rate variability, voice intonation fluctuations, etc. collected through wearable devices or APPs) authorized by the target user, and sets a matching encrypted communication interval based on user behavior characteristics (such as historical communication patterns, emotional fluctuation cycles). This interval is the time range for subsequent data processing. The target user's communication behavior data and emotional state parameters authorized by the target user refer to the communication behavior data and emotional state parameters involved in this application that have been authorized by the target user and do not involve any violation of personal data privacy protection requirements.

[0056] Behavior Analysis Module: This module divides the encrypted communication interval into multiple behavior nodes, each corresponding to communication data from a specific time period. Pattern recognition is performed on the communication data for each behavior node, generating an interaction feature vector corresponding to that node. This vector contains multidimensional features reflecting the user's communication patterns and emotional state.

[0057] Dynamic assessment module: extracts high-risk behavior indicators (such as abnormal communication frequency, drastic emotional fluctuations, etc.) from the interaction feature vector, establishes encryption intervention rules associated with the behavior nodes, and obtains the key control parameters corresponding to the rules (such as encryption strength, key validity period, etc.).

[0058] Channel optimization module: Identifies the communication security level (such as high, medium, and low risk levels) in the key control parameters, dynamically encrypts high-risk behavior indicators based on the security level, and calculates the data fluctuation density of each behavior node under different encryption strategies. This density reflects the severity of data changes.

[0059] Key management module: derives the optimal key threshold based on the data fluctuation density, and generates a key deviation sequence by comparing the current communication data density with the optimal key threshold. This sequence records the difference between the actual data and the ideal state.

[0060] Policy execution module: Analyzes the key deviation sequence and, based on the data fluctuation characteristics of the behavior nodes, integrates the deviation sequence into a specific communication intervention execution plan, such as adjusting the communication content filtering rules and triggering the early warning mechanism.

[0061] The technical solution of the present invention is further described in detail below with reference to specific embodiments.

[0062] Example 1:

[0063] This embodiment relates to the specific implementation of the behavior analysis module. The behavior analysis module first constructs a user feature library corresponding to the behavior node, which contains communication behavior data and behavior parameter vectors mapped to emotional state parameters. Among them, the communication behavior data covers various types of interaction information of the target user within the encrypted communication interval, such as the time of communication initiation, duration of communication, frequency of communication with different interaction objects, number and type of messages sent and received (text, voice, picture, etc.); emotional state parameters are collected through terminals such as wearable devices and mobile phone APPs, including but not limited to heart rate variability data, fluctuation amplitude of voice intonation, emotional intensity index of facial expression recognition, etc. After normalization, these data are mapped into multi-dimensional behavior parameter vectors. Each vector corresponds to the user behavior and emotional state within a specific time window. For example, taking 15 minutes as a time window, a vector consisting of dimensions such as communication frequency, average communication duration, and heart rate variability coefficient within the window is generated.

[0064] When matching similar behaviors for behavioral parameter vectors, classic similarity calculation methods in the field of data mining are used, such as the cosine similarity algorithm or the Euclidean distance measurement method. By calculating the similarity values ​​between different behavioral parameter vectors, vectors with similarities higher than the initial set threshold are divided into the same behavioral cluster group. In the process of dividing behavioral cluster groups, it is necessary to further extract more targeted feature parameters based on the communication frequency and emotional fluctuation index in the behavioral parameter vector, including: extracting the number of communications per unit time (i.e., communication frequency) from the communication frequency dimension, extracting the standard deviation of the emotional parameter within the set time period (i.e., emotional fluctuation intensity parameter) from the emotional fluctuation index dimension, and calculating the time interval between two adjacent communications (i.e., interaction interval parameter) in combination with the communication initiation and end time in the communication behavior data. Based on the communication duration, interaction interval, and emotion intensity parameters extracted above, behavioral feature labels are generated that can intuitively reflect user behavior patterns and emotional characteristics. For example, when the communication frequency corresponding to a certain behavior parameter vector is high (e.g., more than 5 times per hour), the interaction interval is short (e.g., the average interval is less than 10 minutes), and the emotion intensity parameter is large (e.g., the standard deviation exceeds the preset threshold), a "high-frequency, short-interval, anxious interaction" label is generated; if the communication frequency is low (e.g., less than 1 time per hour), the interaction interval is long (e.g., the average interval is more than 1 hour), and the emotion intensity parameter is small, a "low-frequency, long-interval, calm interaction" label is generated.

[0065] After associating the behavioral feature labels with the corresponding behavioral parameter vectors, the behavioral similarity between the feature labels is calculated to further screen the cluster group members. The behavioral similarity calculation here is based on the semantic association of the feature parameters contained in the labels. For example, the labels "high-frequency short-interval anxious interaction" and "high-frequency long-interval anxious interaction" both contain the semantics of "high frequency" and "anxious interaction", and their similarity is higher than that of the label "low-frequency long-interval calm interaction". In specific implementation, word vector models in natural language processing (such as Word2Vec) can be used to convert labels into vector form. The similarity value is determined by calculating the cosine distance between vectors, and parameter vectors with similarity above the preset behavioral threshold are screened out. Finally, several behavioral cluster groups are formed, and the behavioral parameter vectors within each cluster group have similar communication patterns and emotional characteristics.

[0066] After the behavioral clusters are divided, the distribution centroid of the communication data needs to be extracted from each cluster. This centroid serves as the representative feature of the cluster and is used to set the behavior node. The distribution centroid is calculated by taking the arithmetic mean of the dimensions of all behavioral parameter vectors within the cluster to generate a new vector, which serves as the distribution centroid for that cluster. For example, if a cluster contains 100 behavioral parameter vectors, each of which contains data on the three dimensions of communication frequency, interaction interval, and emotional intensity, the communication frequency dimension of the distribution centroid is the average of the communication frequencies of all 100 vectors. The same applies to the interaction interval and emotional intensity dimensions. After the distribution centroid is set as a behavior node, each behavior node corresponds to a time point or time period with typical behavioral characteristics, which is used for subsequent dynamic analysis and risk assessment of the target user's communication behavior. In this way, the behavior analysis module can transform complex user communication data and emotional states into structured behavior nodes and corresponding interaction feature vectors, providing basic data support for the subsequent dynamic assessment module to identify high-risk behavior indicators and establish encrypted intervention rules. The entire process uses a data-driven approach to achieve automatic classification of behavioral patterns and node setting, avoiding the subjectivity and randomness of human intervention and ensuring the objectivity and accuracy of behavioral analysis results.

[0067] Example 2:

[0068] This embodiment further refines the process of generating interaction feature vectors corresponding to behavior nodes. For each behavior node, its temporal distribution within the encrypted communication interval must first be determined, namely its temporal position (e.g., at the beginning, middle, or end of the entire interval) and the time interval between adjacent nodes. Based on this temporal distribution, the node's data fluctuation parameters within a preset period are obtained. The preset period can be flexibly set based on the actual application scenario, for example, 1 hour for high-frequency communication users and 1 day for low-frequency communication users. Data fluctuation parameters include, but are not limited to, the rate of change of communication frequency within the preset period (i.e., the ratio of the difference between the current period's communication frequency and the previous period's communication frequency to the previous period's frequency), the standard deviation of the emotion parameter (reflecting the stability of the emotional state), and the coefficient of variation of the interaction interval (measuring the degree of fluctuation in the interaction interval). Using these parameters, the node's communication fluctuation coefficient is calculated. This coefficient is obtained by comprehensively weighting each data fluctuation parameter. The weighting can be pre-set based on the importance of each parameter to the risk assessment, for example, the communication frequency change rate is weighted 40%, the emotion parameter standard deviation is weighted 30%, and the interaction interval coefficient of variation is weighted 30%.

[0069] When the communication fluctuation coefficient exceeds the first risk threshold, the system determines that the node is a high-risk node. At this time, the communication data of the node is directly extracted to form an interaction feature vector. The scope of communication data extraction includes but is not limited to: content keywords of text communication, voiceprint features of voice communication (such as intonation and speech speed changes), identification information of the interaction object (such as phone number, social account ID), specific timestamps of communication initiation and termination, etc. After normalization, these data are combined into a multi-dimensional interaction feature vector in a preset dimensional order. For example, a vector is composed of dimensions such as [keyword frequency, intonation fluctuation, interaction object unfamiliarity, timestamp offset], which is used for subsequent dynamic assessment module analysis of high-risk behavior indicators.

[0070] When the communication fluctuation coefficient falls below the first risk threshold, the node is marked as a safe node. Because the data fluctuations of a safe node are relatively small, analyzing its communication data alone may not effectively reflect user behavior. Therefore, it is necessary to perform density overlay processing on the communication data of its neighboring nodes. The range of neighboring nodes can be set to N nodes before and after the current safe node. The value of N is determined by the length of the encrypted communication interval and the node partitioning density. For example, if the interval length is 24 hours and the node interval is 1 hour, N can be set to 1 or 2.

[0071] The specific method of density superposition is to splice the communication data of the safety node and its adjacent nodes in chronological order, and perform deduplication or weighted averaging on duplicate or conflicting data. For example, if the previous node of safety node t is t-1 and the next node is t+1, and the communication frequencies of the three nodes are f(t-1), f(t), and f(t+1), respectively, then the communication frequencies after superposition are:

[0072]

[0073] For the keyword frequency of the text content, the keyword lists of the three nodes can be merged, and the number of occurrences of each keyword can be counted and normalized.

[0074] After the data superposition is completed, the superimposed data needs to be reconstructed into an interaction feature vector. The reconstruction process must follow the dimensional definition consistent with the interaction feature vector of the high-risk node to ensure the consistency of subsequent analysis. For example, if the interaction feature vector of the high-risk node contains four dimensions, the data of the safe node after density superposition must also be mapped to these four dimensions. In specific implementation, for data that cannot be directly mapped to dimensions, feature engineering methods can be used for conversion, such as converting timestamp information into a relative offset from a preset period, and converting the depth of the social relationship chain of the interacting object into a strangeness index. In addition, in order to avoid the explosion of vector dimensions due to the excessive amount of superimposed data, dimensionality reduction techniques such as principal component analysis (PCA) can be used to compress the data and retain the principal components that best reflect the behavioral characteristics.

[0075] When processing secure nodes, edge cases must also be considered. For example, when a secure node is at the beginning of an encrypted communication interval, its forward neighboring nodes do not exist. In this case, only the data of its backward neighboring nodes is superimposed. If the secure node is at the end of the interval, only the data of its forward neighboring nodes is superimposed. For cases with only one neighboring node, an asymmetric weighting method can be used for superposition. For example, in the superimposed data of the starting node, the current node's data weighting accounts for 70%, and the backward node's data weighting accounts for 30%, to highlight the fundamental role of the current node.

[0076] The interaction feature vectors generated in this way, whether for high-risk or safe nodes, effectively capture the target user's communication patterns and emotional state at different behavior nodes. The vectors for high-risk nodes directly reflect potential abnormal behavior data, providing a basis for timely triggering encryption intervention rules. The vectors for safe nodes, by overlaying and reconstructing adjacent data, avoid missing potential behavioral trends due to bland data from a single node. For example, the overlaid data of multiple consecutive safe nodes may reveal slowly accumulating emotional fluctuations or shifts in communication patterns, providing more comprehensive data support for the system's long-term risk assessment. The entire process, through dynamic threshold judgment and data overlay strategies, achieves differentiated treatment of behavior nodes of different risk levels. This ensures a timely response to high-risk behavior while enhancing trend analysis of safe behavior, enabling the interaction feature vectors to more accurately reflect the user's true behavior within the encrypted communication range.

[0077] Example 3:

[0078] This embodiment describes the specific implementation of the dynamic evaluation module. The core function of the dynamic evaluation module is to extract key parameters from the interaction feature vector, establish encrypted intervention rules associated with the behavior node, and dynamically optimize the rule content according to the rule coverage. First, the module separates three types of key parameters from the interaction feature vector: communication data ratio, abnormal interaction ratio, and emotional fluctuation parameters. The communication data ratio reflects the proportion of different types of communication content in the overall data, such as the proportion of text messages to the total number of communications, the proportion of voice call duration to the total communication duration, etc.; the abnormal interaction ratio is used to measure the frequency of interaction with unconventional objects (such as unfamiliar numbers, unlabeled contacts) or abnormal communication patterns (such as high-frequency communication late at night); the emotional fluctuation parameter is calculated based on physiological indicators (such as heart rate, skin conductance) collected by wearable devices or APPs and the results of emotional analysis of communication content (such as the emotional polarity score of text messages), such as the ratio of low-frequency components (LF) to high-frequency components (HF) of heart rate variability (HRV), the fluctuation amplitude of Mel-frequency cepstral coefficients (MFCC) of speech signals, etc.

[0079] Based on the three types of parameters mentioned above, the module generates encryption intervention rules for behavior nodes. The rule generation logic is based on preset risk assessment conditions. For example, when the abnormal interaction ratio of a behavior node exceeds 20% and the HRVLF / HF ratio in the emotion fluctuation parameter is greater than 3, a rule is generated to "deeply encrypt the communication content of this node." If the voice call duration in the communication data ratio exceeds 60% and the emotional polarity score in the emotion fluctuation parameter is less than -0.5 (a negative value indicates negative emotion), a rule is generated to "trigger real-time communication content monitoring and log recording." These rules transform abstract interaction features into executable encryption intervention actions through a combination of parameter thresholds. Each rule corresponds to a unique behavior node and specific key control parameters (such as encryption algorithm type and key update frequency).

[0080] When the number of behavior nodes covered by the current encryption intervention rule is less than the preset behavior threshold (for example, the preset threshold is 10 nodes, and the current rule only covers 6 nodes), the module starts the rule expansion mechanism. The specific process is: traverse all adjacent nodes of the current behavior node (the range of adjacent nodes can be set to M nodes before and after, M is determined according to the node density of the encrypted communication interval, such as one node per hour when M = 2), and extract the communication indicators in the interaction feature vector of the adjacent nodes that are not included in the current rules. For example, if the current rule mainly focuses on the proportion of abnormal interactions and emotional fluctuation parameters, and there are indicators such as sudden changes in communication frequency (such as a 50% increase compared to the previous node) or abnormal shortening of the interaction interval (such as shortening from an average of 30 minutes to 5 minutes) in the interaction feature vector of the adjacent nodes that are not included in the rules, then these indicators will be added as new conditions to the current rules.

[0081] When expanding rules, the validity of newly added communication indicators must be verified. This verification method involves analyzing historical data to determine the correlation between the indicator and high-risk behaviors. For example, the correlation coefficient between communication frequency mutation indicators and records of non-suicidal self-harm behaviors can be calculated. If the absolute value of the coefficient is greater than 0.4 (a preset validity threshold), the indicator is considered to have risk warning value and can be included in the rule. If the coefficient is below the threshold, the indicator is ignored to avoid rule redundancy. This data-driven verification mechanism ensures the scientific validity and effectiveness of newly added rule conditions.

[0082] After rule expansion is complete, the newly generated rules must be checked for logical consistency to avoid conditional conflicts or overlapping coverage. For example, if the original rule stipulates "encryption when the proportion of abnormal interactions is greater than 20%," and the new rule stipulates "encryption when the proportion of abnormal interactions is greater than 15% and the communication frequency suddenly changes," it is necessary to ensure that the execution priority or complementary relationship between the two is clear. This can be achieved by setting rule levels (such as basic rules and extended rules) or merging conditions (such as "abnormal interaction proportion > 15% and (communication frequency suddenly changes or emotional fluctuation parameters > threshold)").

[0083] Another key link in the dynamic assessment module is the acquisition of key control parameters. Each encryption intervention rule corresponds to a specific set of key control parameters, which are dynamically adjusted according to the risk level and intervention intensity of the rule. For example, for the "deep content encryption" rule, the key control parameters may include: using the AES-256 encryption algorithm (a high-security algorithm), shortening the key validity period to 30 minutes (50% shorter than the regular validity period), and setting the encrypted data shard size to 1KB (smaller shards improve transmission security); and for the "real-time communication monitoring" rule, the parameters may be: using the AES-128 algorithm, keeping the key validity period at the default of 1 hour, and the data shard size to 4KB. The setting of parameters is based on cryptographic principles and communication security policies to ensure that behavioral nodes with different risk levels receive matching encryption protection.

[0084] During rule execution, the module continuously monitors changes in the parameters of behavioral nodes. When a node's parameters exceed the threshold set by the rule, the corresponding encryption intervention action is immediately triggered and the rule execution log is recorded. The log content includes: rule trigger time, behavioral node identifier, trigger parameter value, executed encryption policy, and key control parameter details. These logs provide data support for subsequent system optimization and risk tracing.

[0085] Through the above process, the dynamic assessment module achieves automated mapping from interaction features to intervention rules, as well as dynamic expansion and optimization of rule coverage. This module not only implements precise encryption intervention on currently identified high-risk behavior nodes, but also proactively identifies potential risk trends through adjacent node feature analysis, expanding the foresight and comprehensiveness of the rules. The entire process is based on a data-driven rule generation and verification mechanism, avoiding the limitations of manual rule-making. This ensures that encryption intervention strategies accurately match target users' communication behaviors and emotional state changes in real time, providing scientific and dynamic technical support for early intervention in non-suicidal self-harm behaviors in depressed adolescents.

[0086] Example 4:

[0087] This embodiment involves a detailed process of the channel optimization module. The channel optimization module first obtains two types of core parameters in the communication security level: the timing parameters of the key validity period and the fluctuation parameters of the data encryption strength. The timing parameters are used to describe the distribution characteristics of the key validity period on the time axis, such as the time point of key replacement, the length of the cycle (such as changing the key once an hour), the matching degree between the start time of the validity period and the peak period of user communication, etc.; the fluctuation parameters reflect the changes in data encryption strength over time, including the complexity of the encryption algorithm (such as switching from AES-128 to AES-256), the adjustment of the key length (such as expanding from 128 bits to 256 bits), the change in the size of the data fragment (such as reducing from 4KB to 1KB), etc.

[0088] When building an encrypted state network, the first step is to identify the periodic pattern of the timing parameters. Periodic pattern analysis is achieved by extracting the temporal characteristics of the timing parameters. For example, using Fourier transform to detect the main frequency component of the key validity period replacement cycle, or using a sliding window to count the number of key replacements within a fixed time period to determine the cycle length. If the current periodic pattern completely matches the preset communication cycle (e.g., the high-frequency communication period from 9:00 AM to 10:00 PM daily, and the rest of the period is low-frequency, based on the target user's historical communication data statistics) (i.e., the period length is consistent and the start time coincides), the starting point of the current period of the timing parameters is set as the starting point of the encrypted state network, serving as the time base for the entire network.

[0089] Calculate the switching correlation between the timing parameters and the volatility parameters. The correlation calculation is based on the co-occurrence frequency of the two types of parameters in historical data. For example, in the past 100 key changes, 80 of them were accompanied by an increase in encryption strength. The correlation between the timing parameters (key changes) and the volatility parameters (encryption strength increase) is 80%. By traversing all possible parameter combinations, a correlation matrix is ​​generated and sorted from high to low by correlation, and the midpoint and end point of the encryption state network are determined in turn. The midpoint represents the intermediate state of different parameter combinations, such as "the key validity period is shortened to 30 minutes and the encryption algorithm is changed to AES-256"; the end point represents the final parameter stable state, such as "the key validity period is 30 minutes, AES-256 encryption, and the shard size is 1KB."

[0090] When performing reverse state verification on a termination point, start from the termination point and trace back along the parameter switching path to the starting point, checking whether each switching step conforms to the parameter association logic in the historical data. For example, if the parameter combination corresponding to the termination point is "key validity period of 15 minutes, ChaCha20 encryption", and the backtracking path is "ChaCha20 encryption → key validity period shortened to 15 minutes", it is necessary to verify whether there is a case in the historical data where the encryption algorithm is first adjusted and then the key validity period is shortened. If so, and the correlation is above a preset switching threshold (such as 50%), the termination point is confirmed to be valid. If not, or the correlation is below the threshold, the termination point is removed and the path is regenerated until the final link that meets the conditions is found.

[0091] To calculate data fluctuation density, we first perform parameter statistics for each endpoint in the encrypted state network. The mean of the timing parameter is the average duration of the key validity period corresponding to that endpoint (e.g., the average of multiple samplings), and the range of the fluctuation parameter is the difference between the maximum and minimum values ​​of the data encryption strength parameter (e.g., the range of the encryption algorithm complexity score). The global covariance is calculated by calculating the covariance matrix of the timing parameters and fluctuation parameters for all nodes, reflecting the overall correlation between the two parameters in the entire network.

[0092] For a single endpoint, the time series density coefficient is calculated by subtracting the mean of the timing parameter at that endpoint from the mean of the timing parameter at its neighboring nodes, taking the absolute value, and dividing it by the corresponding dimension of the global covariance. For example, if the mean of the timing parameter at endpoint A is 30 minutes, the mean of the timing parameter at neighboring node B is 45 minutes, and the variance of the timing parameter in the global covariance is 225, then the time series density coefficient is |30-45| / √225=15 / 15=1. The volatility parameter ratio is the range of the volatility parameter at that endpoint divided by the square root of the variance of the volatility parameter in the global covariance. For example, if the range of the volatility parameter is 10 and the variance of the volatility parameter in the global covariance is 25, then the ratio is 10 / 5=2.

[0093] The data fluctuation density is calculated by weighting the time series density coefficient and the fluctuation parameter ratio according to preset weights. Weighting must be based on the communication security policy. For example, in high-risk scenarios, the time series density coefficient weight is set to 0.7, and the fluctuation parameter ratio weight is set to 0.3 to highlight the impact of key validity period changes on data stability. In low-risk scenarios, the weights can be set to 0.5:0.5. The resulting value reflects the severity of data fluctuations under the encryption policy corresponding to the endpoint. A larger value indicates more unstable data, requiring adjustment to the encryption policy to improve security.

[0094] In practice, the channel optimization module dynamically adjusts the encryption strategy for each behavioral node by updating the encryption state network and data fluctuation density in real time. For example, when a behavioral node is marked as a high-risk node, the module determines the communication security level as "high" based on its corresponding key control parameters. This triggers a high-security encryption state network path and selects a termination point parameter combination with low data fluctuation density (such as a combination with a short key validity period, high encryption strength, and low parameter fluctuation). Dynamically encrypt the high-risk behavior indicators to ensure the security of sensitive data during transmission.

[0095] The entire channel optimization process achieves a dynamic match between encryption strategies and user behavior risks through parameter correlation analysis, network modeling, and density calculation. The construction of the encryption state network is based on historical data and communication cycle characteristics to ensure the logic and effectiveness of parameter switching. The calculation of data fluctuation density quantifies the stability of the encryption strategy from the dual dimensions of timing and intensity, providing a key basis for the key management module to derive the optimal key threshold. This module not only improves the security of the communication channel but also reduces communication delays or resource consumption caused by excessive encryption through a dynamic adjustment mechanism, achieving a balance between security and communication efficiency. It is suitable for the encryption protection needs of real-time communication data in intervention scenarios for non-suicidal self-harm behaviors in depressed adolescents.

[0096] Example 5:

[0097] This embodiment describes the implementation details of the key management module and the policy execution module. When the key management module derives the optimal key threshold, it first filters the encryption mode that is closest to the current data fluctuation density from the historical data. The historical data is stored in the system database, and each encryption mode contains the data fluctuation density value, the corresponding key parameters and the time series distribution characteristics. The screening process calculates the absolute difference between the current data fluctuation density and the historical records, and selects the top K modes with the smallest difference (K is a preset value, such as 5) as the candidate set. For each candidate mode, calculate its Euclidean distance with the current data in the time series distribution, that is, compare the similarity of the key validity period change curve or encryption strength fluctuation curve on the time axis. For example, if the encryption strength of the current data fluctuates greatly in the 14:00-16:00 period, and the fluctuation trend of a historical mode in the same time period is similar, then the Euclidean distance between the two is small, and this value is used as the first key reference value.

[0098] Count the difference in the number of peak points between the current data fluctuation density and the historical encryption pattern. A peak point is defined as a point where the data fluctuation density exceeds a local threshold (such as a point where it exceeds the density value of two adjacent nodes). The number of differences is the absolute value of the difference between the number of peak points of the current data and the historical pattern. For example, if the current data has 4 peak points and a historical pattern has 6, the number of differences is 2, which is used as the second key reference value. The first reference value reflects the similarity of the time series distribution, and the second reference value reflects the difference in the severity of the fluctuation. Together, they represent the overall matching degree between the current data and the historical pattern.

[0099] Based on a linear combination of the first and second key reference values ​​(e.g., their sum or weighted sum, where the weight can be set to 1:1), a preset key threshold table is queried. This table is a two-dimensional mapping table, with the first reference value interval on the horizontal axis and the second reference value interval on the vertical axis. Each cell stores the corresponding optimal key threshold (e.g., key validity period, encryption strength coefficient, etc.). For example, when the first reference value is 0.8 and the second reference value is 1, the optimal key threshold is determined by table lookup to be "key validity period 45 minutes, encryption strength level 3." The threshold table is generated through historical data training, using cluster analysis or regression models to fit the correlation between parameters to ensure the rationality of the threshold.

[0100] When the policy execution module parses the key deviation sequence, it first divides the sequence into positive and negative fluctuation domains based on the data fluctuation trend (increasing, decreasing, or stable) of each node. The positive fluctuation domain refers to the interval where the data fluctuation density of multiple consecutive nodes shows an upward trend, while the negative fluctuation domain refers to the interval where the data fluctuation density shows a downward trend. For example, if the density values ​​of nodes i, i+1, and i+2 in the sequence are 0.6, 0.7, and 0.8, respectively, then nodes i through i+2 belong to the positive fluctuation domain. If the density values ​​of nodes j, j+1, and j+2 are 0.9, 0.7, and 0.5, then they belong to the negative fluctuation domain.

[0101] In the positive fluctuation domain, the convergence frequency of the key deviation is extracted, that is, the ratio of the number of nodes with gradually decreasing deviation values ​​(the difference between the current data density and the optimal key threshold) to the total number of nodes in the domain. For example, if there are 5 nodes in the positive fluctuation domain, and the deviation values ​​of 3 nodes decrease from 0.3 to 0.1, 0.2 to 0.1, and 0.4 to 0.2, then the convergence frequency is 3 / 5 = 60%. In the negative fluctuation domain, the diffusion frequency of the key deviation is extracted, that is, the ratio of the number of nodes with gradually increasing deviation values. For example, if there are 4 nodes in the negative fluctuation domain, and the deviation values ​​of 2 nodes increase from 0.1 to 0.3 and 0.2 to 0.5, then the diffusion frequency is 2 / 4 = 50%.

[0102] The convergence frequency and diffusion frequency are dynamically reconciled based on the behavior node's communication weight. The communication weight reflects the node's importance and can be set based on factors such as the closeness of the communication partner (e.g., family and friends have higher weights) and the sensitivity of the communication content (e.g., topics involving self-harm have higher weights). It ranges from 0 to 1. The reconciliation is performed by multiplying the convergence frequency in the positive fluctuation domain by the average weight of the nodes in that domain, and the diffusion frequency in the negative fluctuation domain by the average weight of the nodes in that domain. The two are summed to form the final reconciled value. For example, if the average weight in the positive fluctuation domain is 0.8 and the convergence frequency is 60%, the contribution value is 0.8 × 0.6 = 0.48; if the average weight in the negative fluctuation domain is 0.7 and the diffusion frequency is 50%, the contribution value is 0.7 × 0.5 = 0.35, resulting in a reconciliation value of 0.48 + 0.35 = 0.83.

[0103] The configuration parameters for the communication intervention execution plan are generated based on the reconciliation value. These parameters include the encryption strength adjustment range (e.g., when the reconciliation value exceeds 0.8, the encryption strength is increased by one level), the key update frequency (e.g., when the reconciliation value falls below 0.5, the key validity period is extended to 1.5 times the default value), and the warning trigger conditions (e.g., when the reconciliation value is between 0.5 and 0.8, a medium-level warning is triggered and a log is recorded). These parameters are associated with the reconciliation value through preset mapping rules, such as piecewise functions or fuzzy logic rules, to ensure the flexibility and targeted nature of the intervention strategy.

[0104] When executing an intervention plan, the system adjusts the encryption strategy for each behavioral node based on the configured parameters. For example, for nodes with higher weights in the positive fluctuation domain, if the harmonized value indicates that the deviation has converged but still presents a high risk, high encryption strength can be maintained and the key validity period can be shortened. For nodes with lower weights in the negative fluctuation domain, if the deviation diffusion trend is flat, encryption strength can be appropriately reduced to reduce resource consumption. This entire process achieves dynamic optimization of the intervention strategy through data fluctuation trend analysis and weight reconciliation, ensuring strict protection for high-risk nodes while avoiding excessive intervention in low-risk nodes, thereby improving the overall operational efficiency of the system.

[0105] The key management module and policy enforcement module form a closed-loop encryption policy adjustment mechanism through data-driven threshold derivation and deviation analysis. The former ensures the scientific accuracy of key thresholds based on historical pattern matching and parameter difference calculation; the latter achieves precise intervention plans by balancing fluctuation trends and weights. Working together, the system dynamically adjusts encryption policies based on the target user's real-time communication data, ensuring communication security while providing flexible and efficient technical support for interventions against non-suicidal self-injurious behaviors in depressed adolescents.

[0106] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0107] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. An encrypted communication system for intervention of non-suicidal self-harm behaviors in depressed adolescents, characterized by: include: A data acquisition module is used to obtain communication behavior data and emotional state parameters of a target user authorized by the target user, and set an encrypted communication interval that matches the user's behavior characteristics, wherein the encrypted communication interval is the obtained data to be encrypted; The behavior analysis module is used to divide the encrypted communication interval into multiple behavior nodes, perform pattern recognition on the communication data of each behavior node, and generate the interaction feature vector corresponding to the behavior node; A dynamic assessment module is used to extract high-risk behavior indicators from the interaction feature vector, establish encryption intervention rules associated with the behavior nodes, and obtain key control parameters corresponding to the encryption intervention rules; The channel optimization module is used to identify the communication security level in the key control parameters, dynamically encrypt high-risk behavior indicators based on the communication security level, and calculate the data fluctuation density of each behavior node under different encryption strategies; The key management module is used to derive the optimal key threshold according to the data fluctuation density and generate a key deviation sequence by comparing the current communication data density with the optimal key threshold; The policy execution module is used to parse the key deviation sequence and integrate the key deviation sequence into a communication intervention execution plan based on the data fluctuation characteristics of the behavior node.

2. The encrypted communication system for intervention of non-suicidal self-injurious behavior in depressed adolescents according to claim 1, characterized in that: The implementation of the behavior analysis module includes: building a user feature library corresponding to the behavior node, the user feature library containing communication behavior data and behavior parameter vectors mapped to emotional state parameters; Perform similar behavior matching on the behavior parameter vectors, and divide the behavior parameter vectors into behavior cluster groups according to the matching results; extract the distribution center points of the communication data from the behavior cluster groups, and set the distribution center points as behavior nodes.

3. The encrypted communication system for intervention of non-suicidal self-injurious behavior in depressed adolescents according to claim 2, characterized in that: The behavior clustering groups that divide the behavior parameter vectors also include: According to the communication frequency and emotion fluctuation index in the behavior parameter vector, the communication duration, interaction interval and emotion intensity parameters are extracted, and the behavior feature label is generated based on the communication duration, interaction interval and emotion intensity parameters; The behavioral feature labels are associated with the behavioral parameter vectors. By calculating the behavioral similarity between the feature labels, the parameter vectors with similarity higher than the preset behavioral threshold are screened to form a behavioral cluster group.

4. The encrypted communication system for intervention of non-suicidal self-injurious behavior in depressed adolescents according to claim 1, characterized in that: The implementation methods for generating the interaction feature vector corresponding to the behavior node include: For each behavior node, according to the time series distribution of the behavior node in the encrypted communication interval, the data fluctuation parameters of the behavior node within the preset period are obtained, and the communication fluctuation coefficient of the node is calculated; When the communication fluctuation coefficient exceeds the first risk threshold, the node is marked as a high-risk node, and its communication data is extracted to form an interactive feature vector; when the communication fluctuation coefficient is lower than the first risk threshold, the node is marked as a safe node, and the communication data of the node's adjacent nodes are density-superimposed, and the superimposed data is reconstructed into an interactive feature vector.

5. The encrypted communication system for intervention of non-suicidal self-injurious behavior in depressed adolescents according to claim 1, characterized in that: The implementation of the dynamic evaluation module includes: Separate the communication data ratio, abnormal interaction ratio, and emotion fluctuation parameters from the interaction feature vector, and generate encrypted intervention rules for the behavior node based on these parameters; If the number of behavior nodes covered by the current encryption intervention rule is less than the preset behavior threshold, the interaction feature vectors of adjacent behavior nodes are traversed, and the communication indicators not included in the encryption intervention rules of the adjacent nodes are added to the current rule.

6. The encrypted communication system for intervention of non-suicidal self-harm behaviors in depressed adolescents according to claim 1, characterized in that: The channel optimization module is implemented by: obtaining the timing parameters of the key validity period and the fluctuation parameters of the data encryption strength in the communication security level; An encryption state network associated with timing parameters and fluctuation parameters is constructed. According to the switching probability of various paths in the encryption state network, the data fluctuation density under different encryption strategies is determined.

7. The encrypted communication system for intervention of non-suicidal self-harm behavior in depressed adolescents according to claim 6, characterized in that: Building a crypto-state network also includes: Identify the periodic pattern of the timing parameters, and if the current periodic pattern completely matches the preset communication period, set the timing parameters as the starting point of the encrypted state network; Calculate the switching correlation between the timing parameters and the fluctuation parameters, and generate the intermediate points and end points of the encrypted state network in descending order of switching correlation; The state of the termination point is reversely verified. When the switching correlation of the termination point is lower than the preset switching threshold, it is output as the final link of the encrypted state network.

8. The encrypted communication system for intervention of non-suicidal self-injurious behavior in depressed adolescents according to claim 7, characterized in that: The implementation methods for calculating data fluctuation density include: Statistically calculate the mean value and range of the timing parameters of each terminal point in the encrypted state network, and calculate the global covariance of all node parameters; The mean value of the timing parameters of a single termination point is subtracted from the mean value of the timing parameters of the adjacent nodes, and the difference is divided by the global covariance to obtain the timing density coefficient; at the same time, the ratio of the fluctuation parameter range to the global covariance is calculated, and the weighted sum of the ratio and the timing density coefficient is taken as the data fluctuation density of the node.

9. The encrypted communication system for intervention of non-suicidal self-injurious behavior in depressed adolescents according to claim 1, characterized in that: The implementation methods for deriving the optimal key threshold include: Extracting the encryption mode in the historical data that is closest to the current data fluctuation density, and calculating the Euclidean distance between the data fluctuation density in the closest encryption mode and the current data fluctuation density in the time series distribution as the first key reference value; Counting the difference in the number of peak points between the data fluctuation density in the closest encryption mode and the current data fluctuation density, and using the difference as a reference value for the second key; Based on the linear combination of the first key reference value and the second key reference value, an optimal key threshold in the preset key threshold table is matched.

10. The encrypted communication system for intervention of non-suicidal self-harm behaviors in depressed adolescents according to claim 1, characterized in that: The implementation method of the policy execution module includes: dividing the positive fluctuation domain and the negative fluctuation domain according to the data fluctuation trend of each node in the key deviation sequence; The convergence frequency of key deviation in the forward fluctuation domain and the diffusion frequency of key deviation in the reverse fluctuation domain are extracted, and the two are dynamically reconciled according to the communication weight of the behavior node to generate the configuration parameters of the communication intervention execution plan.

Images