CN119007998A

CN119007998A - Mental health monitoring and early warning method and system based on big data

Info

Publication number: CN119007998A
Application number: CN202411047830.XA
Authority: CN
Inventors: 肖建锋
Original assignee: Beijing Ruige Soul Technology Co ltd
Current assignee: Beijing Ruige Soul Technology Co ltd
Priority date: 2024-08-01
Filing date: 2024-08-01
Publication date: 2024-11-22

Abstract

The present invention discloses a method and system for monitoring and early warning of mental health based on big data, wherein the method comprises: collecting user data through multiple collection channels to generate multi-source data. Preprocessing the multi-source data to generate a preprocessed multi-source data set. Analyzing the preprocessed multi-source data to identify mental health risk factors and abnormal behavior patterns. Constructing a mental health early warning model based on the mental health risk factors and the abnormal behavior patterns. Real-time monitoring of the user's mental health status data based on the mental health early warning model, wherein when the user's mental health status data is monitored to be abnormal, an early warning message is issued. Based on the early warning message, an assessment of the mental health status is performed to generate an assessment result. Thus, the present invention can comprehensively analyze multi-source data, monitor and accurately judge the individual's mental health status in real time, provide a personalized early warning plan, and improve the effect of early warning intervention.

Description

Mental health monitoring and early warning method and system based on big data

Technical Field

The embodiment of the invention relates to the technical field of psychological health early warning, in particular to a psychological health monitoring early warning method and system based on big data.

Background

In the current high-pressure social environment, psychological health problems are increasingly prominent, and particularly young groups such as college students face multiple pressures, so that various psychological problems are easy to occur. Traditional mental health monitoring and early warning methods are limited by manpower, time and territory, and are difficult to discover and intervene effectively in time. The rapid development of big data technology provides new possibilities for mental health monitoring and early warning.

Disclosure of Invention

The embodiment of the invention provides a psychological health monitoring and early warning method and system based on big data, which can comprehensively analyze multi-source data, monitor and accurately judge the psychological health state of an individual in real time, provide a personalized early warning intervention scheme and improve the early warning intervention effect.

In order to achieve the above object, in a first aspect, the present invention provides a mental health monitoring and early warning method based on big data, including:

collecting user data through various collecting channels to generate multi-source data;

preprocessing the multi-source data to generate a preprocessed multi-source data set;

Analyzing based on the preprocessing multi-source data, and identifying and obtaining psychological health risk factors and abnormal behavior patterns;

Constructing a mental health early warning model based on the mental health risk factors and the abnormal behavior mode;

Monitoring psychological health state data of a user in real time based on the psychological health early warning model, wherein when the psychological health state data of the user is monitored to be abnormal, early warning information is sent out;

Based on the early warning information, carrying out evaluation on the psychological health state and generating an evaluation result;

based on the evaluation result, a personalized psychological health early warning intervention scheme is generated and pushed.

In one embodiment of the present invention, the collecting the user data through the plurality of collecting channels, and generating the multi-source data includes:

collecting user data through the first collecting channel, the second collecting channel and the third collecting channel to obtain first user data, second user data and third user data;

the multi-source data is generated based on the first user data, the second user data, and the third user data.

In one embodiment of the present invention, the preprocessing the multi-source data to generate a preprocessed multi-source data set includes:

Performing data cleaning, data standardization and normalization and data integration processing on the multi-source data to generate integrated multi-source data;

Labeling the integrated multi-source data to generate normal psychological state data and abnormal psychological state data;

the pre-processed multi-source dataset is generated based on the normal mental state data and the abnormal mental state data.

In an embodiment of the present invention, the analyzing based on the preprocessed multi-source data, and identifying the mental health risk factors and the abnormal behavior patterns includes:

based on the preprocessing multisource data set, psychological characteristics, behavioral characteristics and social characteristics are obtained;

And identifying and obtaining the psychological health risk factors and the abnormal behavior patterns by adopting a preset method based on the psychological characteristics, the behavior characteristics and the social characteristics.

In an embodiment of the invention, the monitoring the mental health status data of the user in real time based on the mental health early-warning model, wherein when the monitoring that the mental health status data of the user is abnormal, sending early-warning information includes:

monitoring psychological health state data of a user in real time based on the psychological health early warning model, and generating a monitoring result;

And when the monitoring result exceeds a preset threshold value, judging that an abnormal condition occurs, and sending out early warning information.

In an embodiment of the present invention, the performing the evaluation of the mental health state based on the early warning information, and generating the evaluation result includes:

Based on the early warning information, identifying early warning factors and severity of early warning;

And acquiring multi-dimensional information, and evaluating the mental health state based on the multi-dimensional information, the early warning factors and the early warning severity degree to generate an evaluation result.

In a second aspect, the present invention provides a mental health monitoring and early warning system based on big data, comprising: the system comprises a first generation module, a second generation module, an identification module, a construction module, an emission module, a third generation module and a fourth generation module. The first generation module is used for collecting user data through various collection channels and generating multi-source data. The second generation module is used for preprocessing the multi-source data and generating a preprocessed multi-source data set. The identification module is used for analyzing based on the preprocessing multi-source data and identifying and obtaining psychological health risk factors and abnormal behavior patterns. The construction module is used for constructing a psychological health early warning model based on the psychological health risk factors and the abnormal behavior mode. The sending module is used for monitoring the psychological health state data of the user in real time based on the psychological health early warning model, wherein when the psychological health state data of the user is monitored to be abnormal, early warning information is sent. The third generation module is used for evaluating the psychological health state based on the early warning information and generating an evaluation result. The fourth generation module is used for generating and pushing a personalized psychological health early warning intervention scheme based on the evaluation result.

In one embodiment of the present invention, the first generating module includes: the obtaining unit and the first generating unit. The obtaining unit is used for collecting the user data through the first collecting channel, the second collecting channel and the third collecting channel to obtain the first user data, the second user data and the third user data. The first generation unit is configured to generate the multi-source data based on the first user data, the second user data, and the third user data.

In a third aspect, the present invention provides an electronic device, comprising:

At least one processor; and

A memory communicatively coupled to the at least one processor;

wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the big data based mental health monitoring and pre-warning method as described above.

In a fourth aspect, the present invention provides a computer readable storage medium comprising a computer program and instructions which, when executed on a computer, cause the computer to perform a big data based mental health monitoring and early warning method as described above.

Compared with the prior art, the mental health monitoring and early warning method and system based on big data have the following beneficial effects:

1. the multi-source data can be comprehensively analyzed, so that the psychological health state of an individual can be accurately judged, and the probability of false alarm and false alarm is reduced;

2. the psychological health state of the individual can be monitored in real time, abnormal conditions can be found in time, and the possibility is provided for early intervention;

3. According to the specific conditions and the data analysis results of the individuals, a personalized psychological health early warning intervention scheme is formulated, and the early warning intervention effect is improved.

Drawings

Fig. 1 is a flow chart of a mental health monitoring and early warning method based on big data in a first embodiment of the invention;

fig. 2 is a schematic structural diagram of a mental health monitoring and early warning system based on big data in a second embodiment of the present invention;

fig. 3 is a schematic structural diagram of an electronic device according to a third embodiment of the present invention.

Detailed Description

Embodiments of the present invention will be described in further detail below with reference to the drawings and examples. It should be understood that the particular embodiments described herein are illustrative only and are not limiting of embodiments of the invention. It should be further noted that, for convenience of description, only some, but not all of the structures related to the embodiments of the present invention are shown in the drawings.

For ease of understanding, the main implementation concept of the embodiments of the present invention will be briefly described first.

In the current society, with the acceleration of life rhythm and the aggravation of social competition, psychological health problems are becoming important factors affecting the quality of life and social stability of individuals. Especially young groups such as college students are more prone to psychological problems such as anxiety and depression when facing multiple pressures such as academic, employment and interpersonal relationship. However, the traditional mental health monitoring and early warning method mainly depends on means such as questionnaires, face-to-face consultation and the like, is limited by manpower, time and region, and is difficult to realize real-time monitoring and effective intervention on a large number of individuals.

In recent years, the rapid development of big data technology provides new ideas and methods for mental health monitoring and early warning. Big data technology can process and analyze massive and diversified data, and dig valuable information and modes from the massive and diversified data. In the field of mental health, the psychological state and the change trend of an individual can be more comprehensively known by collecting and analyzing multi-source data such as behavioral data, social data and physiological data of the individual, so that more accurate psychological health monitoring and early warning are realized.

However, the current mental health monitoring and early warning method based on big data has some challenges such as low data quality, insufficient accuracy of data processing and analysis methods, insufficient individuation of early warning models and the like.

The inventor provides a psychological health monitoring and early warning method and system based on big data through finding the defects in the prior art, and improves the accuracy and effectiveness of psychological health monitoring and early warning through means of comprehensively analyzing multi-source data, monitoring the psychological health state of an individual in real time, making personalized early warning intervention schemes and the like, so that the psychological health of the individual is protected and navigated.

Example 1

Fig. 1 is a flow chart of a mental health monitoring and early warning method based on big data in a first embodiment of the present invention, as shown in fig. 1, the first embodiment provides a mental health monitoring and early warning method based on big data, including:

step S100, user data are collected through various collecting channels, and multi-source data are generated;

Specifically, step S100 is an initial step of the mental health monitoring and early warning method, and is characterized in that user data is collected widely through various channels to form comprehensive and multidimensional multi-source data. Specifically, this step includes collecting relevant data of the user through different collection channels (which may be social media, online surveys, physiological monitoring devices, etc.), thereby generating multi-source data containing information about aspects of the user. The implementation of the step provides a rich and comprehensive data basis for the subsequent analysis of the psychological health state and the construction of an early warning model.

Step S200, preprocessing the multi-source data to generate a preprocessed multi-source data set;

Specifically, firstly, data cleaning is required to be carried out on multi-source data, and noise, abnormal values and repeated data in the multi-source data are removed, so that the purity and consistency of the data are ensured. And then, data standardization and normalization processing are carried out, data with different sources and different scales are converted into a unified numerical range, and dimensional differences among the data are eliminated, so that subsequent analysis and comparison are more fair and objective. And then, carrying out data integration processing, organizing and integrating the cleaned and standardized data according to a specific structure and format to form an integrated multi-source data set, and providing convenience for subsequent feature extraction and model construction. Finally, marking the integrated multi-source data, and dividing the multi-source data into normal psychological state data and abnormal psychological state data according to the characteristics and the labels of the data so as to facilitate the identification of the subsequent psychological health risk factors and abnormal behavioral patterns. Through the series of preprocessing operations, the quality and the usability of the multi-source data can be effectively improved, and a solid foundation is laid for the subsequent analysis of the psychological health state and the construction of an early warning model. Meanwhile, the step is an indispensable ring in the whole psychological health monitoring and early warning method, ensures the accuracy and effectiveness of subsequent analysis, improves the sensitivity and specificity of an early warning model, and protects the psychological health of individuals.

Step S300, analyzing based on the preprocessing multi-source data, and identifying and obtaining psychological health risk factors and abnormal behavior patterns;

Specifically, step S300 is to perform deep analysis based on the preprocessed multi-source data set to identify key links of the mental health risk factors and the abnormal behavior patterns. The core of this step is to extract valuable features and patterns from the multi-source data, revealing risk factors and abnormal behavior related to mental health. Specifically, firstly, multidimensional information such as psychological characteristics, behavioral characteristics, social characteristics and the like is acquired from the preprocessed multi-source dataset, and the information can comprehensively reflect the psychological health state and behavioral patterns of an individual. Then, the features are deeply analyzed and mined by adopting a preset method and technology, such as a machine learning algorithm, a statistical analysis and the like, so as to identify risk factors and abnormal behavior patterns related to psychological health problems. These risk factors and abnormal behavioral patterns may include, for example, mood swings, changes in social behavior, abnormalities in physiological indicators, etc. of the individual, which can serve as early warning signals that suggest mental health problems that the individual may have. Through the analysis and the recognition of the step S300, the psychological health state and the behavioral pattern of the individual can be known more accurately, and powerful support is provided for the subsequent early warning model construction and the establishment of personalized intervention schemes. Meanwhile, the step is also a core link in the whole psychological health monitoring and early warning method, is directly related to the accuracy and effectiveness of an early warning model, and has important significance for improving the psychological health level of individuals and preventing psychological diseases.

Step S400, a psychological health early warning model is constructed based on the psychological health risk factors and the abnormal behavior mode;

Specifically, step S400 is a key stage of constructing a mental health early warning model, and based on the mental health risk factors and abnormal behavior patterns extracted in step S300, the data are trained and validated by using an advanced machine learning algorithm or a statistical model, so as to establish an early warning system capable of accurately predicting the mental health risk of the individual. The core of the stage is the selection and optimization of the model, so that the model can capture the fine change in the data, and can avoid over fitting, and the prediction accuracy and generalization capability are improved. Through repeated iteration and adjustment of model parameters, the step S400 aims to construct a high-sensitivity and high-specificity early warning model so as to realize real-time monitoring and early warning of the psychological health state of an individual and provide scientific basis for timely taking intervention measures. Successful implementation of this step is significant in improving the efficiency and quality of mental health services and promoting overall well-being of individuals and society.

In step S400, the construction of the mental health early warning model is a core link of the whole mental health monitoring early warning method. This step establishes a model capable of accurately predicting the psychological health risk of the individual by applying an advanced machine learning algorithm or statistical model based on the psychological health risk factors and abnormal behavior patterns identified in step S300, specifically:

Selecting a proper machine learning algorithm is a key for building a mental health early warning model, and the embodiment adopts a long-short-term memory network (LSTM) in a deep learning model to build the early warning model, and before building the model, the preprocessed multi-source data set (including mental health risk factors and abnormal behavior patterns) needs to be ensured to be complete and high-quality. The data should be divided into training sets, validation sets and test sets for use during the model training, validation and test phases.

Feature selection is an important step in model construction that aims to select the features from the preprocessed dataset that have the most impact on the predicted target. In the mental health pre-warning model, these features may include psychological features (e.g., emotional stability, stress level), behavioral features (e.g., social activity, lifestyle), physiological features (e.g., heart rate, sleep quality), etc. of the user. By means of feature selection, complexity of the model can be reduced, and generalization capability of the model is improved.

And constructing an early warning model by using the LSTM network. LSTM is particularly suited for processing time series data because it is capable of capturing long-term dependencies. Parameters such as the input size, the number of hidden units, the number of layers, etc. of the LSTM layer are defined. And adding necessary layers, such as a full connection layer (Dense layer), for outputting a final early warning result. The LSTM model is trained using the training set data. The error between the predicted and actual values is calculated by forward propagation and then the network weights are adjusted using a back propagation algorithm to minimize the error. Loss values and accuracy during training are monitored to prevent overfitting. The generalization ability of the model is evaluated using the validation set data. And adjusting model parameters, optimizer settings or feature selection according to the verification result to improve the model performance. The performance of the final model is evaluated using the test set data. If the performance of the model meets the requirements, the model is deployed into practical application for real-time early warning. For example, an LSTM-based mental health alert model needs to be constructed for monitoring the mental health of a user in real time and giving an alert when a potential risk occurs. Collecting psychological test results (such as depression scale scores), social media activity records (such as posting frequency and emotion expression), physiological indexes (such as heart rate variability and sleep quality) monitored by wearable equipment and the like of a user. Statistical features (e.g., average score, standard deviation) are extracted from the psychological test results. Text features and time series features (e.g., emotional word frequency, posting time interval) are extracted from the social media activity record. Time series characteristics (such as daily trend of heart rate variability) are extracted from the physiological index. Defining an LSTM model, enabling an input layer to accept time sequence characteristics (such as time sequence of physiological indexes), and capturing long-term dependency through a plurality of LSTM layers. And adding a full-connection layer (Dense layer) as an output layer, and outputting early warning results (such as normal risk, mild risk, moderate risk and the like). And training the LSTM model by using the marked training set data. The generalization ability of the model is monitored using the validation set data and model parameters are adjusted to optimize performance. And (5) evaluating indexes such as accuracy, recall rate, F1 score and the like of the final model by using the test set data. If the performance of the model meets the requirements, the model is deployed into a mental health monitoring system, the mental health state of the user is monitored in real time, and early warning information is sent out when the potential risk is detected. Through the steps, an effective psychological health early warning model can be constructed by utilizing the LSTM network, and timely psychological health support is provided for users.

Step S500, monitoring psychological health state data of a user in real time based on the psychological health early warning model, wherein when the psychological health state data of the user is monitored to be abnormal, early warning information is sent out;

Specifically, step S500 is a final step in the mental health monitoring and early warning method, and is also a key step for converting the early analysis and modeling result into practical application. This stage focuses mainly on the generation and transmission of early warning signals, and the formulation and implementation of personalized intervention schemes. Specifically, in step S500, the psychological health status of the individual is evaluated in real time according to the prediction result of the early warning model, and an early warning signal is generated in time when the potential risk is identified. These signals will be communicated to the individual and its designated support or professional by appropriate means (e.g., cell phone application notifications, e-mail, etc.) so that the necessary intervention can be taken in time. Meanwhile, step S500 further creates a personalized intervention scheme according to the specific risk factors and abnormal behavior patterns of the individual. These regimens may include, for example, psychological counseling, cognitive behavioral therapy, relaxation training, and the like, in various forms of mental health services aimed at helping individuals effectively address mental health problems and improving mental toughness. Through implementation of the step S500, the psychological health monitoring and early warning method not only realizes real-time monitoring and early warning of the psychological health state of the individual, but also provides targeted intervention support, so that a complete psychological health service system is formed. The effective execution of the step has important significance for promoting the psychological health of individuals and preventing the occurrence and development of psychological diseases.

Step S600, based on the early warning information, carrying out evaluation of mental health state and generating an evaluation result;

Specifically, step S600 is a process of further evaluating the mental health state of the user based on the early warning information. The purpose of this step is to more accurately understand the current mental health of the user, so as to provide more targeted advice for subsequent early warning intervention schemes. Specifically, the early warning factor and severity of the early warning that resulted in the early warning are first identified. Early warning factors may include, for example, certain behavioral patterns, emotional states, social activities, etc. of the user, which may indicate that there is a potential risk of the mental health of the user. The severity of the warning is then used to assess the magnitude of this risk to determine if immediate intervention is required. Next, multi-dimensional information is obtained, which may include, for example, the user's personal context, historical mental health records, current living environment, and the like. Such information is critical to comprehensively assessing a user's mental health, as they may provide more context and details about the user's mental health. Finally, based on the multidimensional information, the early warning factors and the early warning severity, the psychological health state is estimated, and an estimation result is generated. The evaluation result is used as an important basis for the subsequent establishment of personalized mental health early warning intervention schemes. Through this step, the mental health of the user can be more accurately known, and more effective help and support can be provided for the user.

Step S700, based on the evaluation result, generating and pushing a personalized psychological health early warning intervention scheme;

Specifically, step S700 is a process of generating and pushing a personalized mental health early warning intervention scheme based on the evaluation result. The core of the step is to customize a set of proper early warning intervention scheme for the user according to the psychological health state assessment result obtained in the previous step. The regimen may include, for example, suggesting the user for a particular psychological consultation or treatment, recommending a suitable mental health resource or application, or providing a specific strategy to address the current mental health challenge. Through personalized intervention schemes, the unique requirements of users can be better met, and the effectiveness and success rate of early warning intervention are improved. In addition, step S700 may involve further communication with the user to ensure that the user understands and accepts the proposed intervention scheme and provides the necessary support and guidance to assist the user in implementing the scheme smoothly. Ultimately, the goal of this step is to help the user effectively address mental health challenges, raise the mental health level of the user, and prevent potential mental problems. Through personalized early warning intervention scheme, more careful and effective mental health service can be provided for the user, and the overall welfare and life quality of the user are promoted.

In this embodiment, the step S100 includes:

step S101, user data are acquired through a first collecting channel, a second collecting channel and a third collecting channel, and first user data, second user data and third user data are obtained;

Specifically, through specialized psychological assessment tools and methods, a comprehensive and detailed examination of a user's psychological condition is performed to determine whether a potential psychological problem or risk exists. This step is the basis for the development and implementation of subsequent interventions, which is crucial to ensure the pertinence and effectiveness of the interventions. Through scientific and objective evaluation, timely and accurate psychological health feedback can be provided for individuals, and a solid foundation is laid for smooth proceeding of subsequent steps. The first collecting channel may be, for example, a psychological measuring tool (such as a questionnaire, a scale, etc.), the second collecting channel may be, for example, a wearable device (such as a smart bracelet, a heart rate monitor, etc.), and the second collecting channel may be, for example, social media (such as a microblog, a micro letter, a tremble sound, etc.). And the psychological state assessment is carried out periodically or aperiodically by using a heart measuring tool, so that the timeliness and the accuracy of the data are ensured. The method comprises the steps of capturing public information of a user from social media, such as posting content, interaction conditions and the like, so as to analyze the emotional state and social behavior mode of the user. The wearable device can collect physiological data of a user, such as heart rate, blood pressure, sleep quality and the like.

Step S102, generating the multi-source data based on the first user data, the second user data and the third user data;

specifically, based on the first user data, the second user data, and the third user data obtained through the first collecting channel, the second collecting channel, and the third collecting channel, respectively, these data are integrated to generate multi-source data containing multi-dimensional information such as psychological measurement, wearing equipment, social media, and the like.

In this embodiment, the step S200 includes:

Step S201, performing data cleaning, data standardization and normalization and data integration processing on the multi-source data to generate integrated multi-source data;

Specifically, after the multi-source data is generated, a preprocessing stage is entered, and the core task of the preprocessing stage is to carefully process and sort the raw data so as to ensure the validity and accuracy of subsequent analysis. Specifically, this step first performs a data cleansing work, identifies and eliminates noise, outliers, and repeated recordings in the data, thereby cleansing the data set. And then, data standardization and normalization processing are carried out, and the step aims at eliminating deviation possibly introduced by dimension and scale differences among different data sources, so that all the features are in the same numerical range, and the subsequent unified analysis is convenient. And then, carrying out data integration processing, and organizing and arranging the multi-source data subjected to cleaning and standardization according to a specific logic structure and format to form an integrated multi-source data set with clear structure and convenient access. Through the series of pretreatment measures, a solid data base is laid for the identification of the subsequent psychological health risk factors and abnormal behavior patterns.

In one embodiment, there is a multi-source data set containing mental health data of college students from different sources, such as psychometric questionnaires (e.g., SDS depression self-assessment), social media activity records (e.g., microblog posting frequency, interaction conditions), physiological indicators monitored by wearable devices (e.g., heart rate variability, sleep quality), etc. Noise, outliers and repeated recordings exist in the original data, the dimensions and scales of different data sources are inconsistent, and preprocessing is needed to generate high-quality integrated multi-source data.

All data sources are checked, obviously unreasonable data points are identified and deleted, such as abnormally high or low heart rate variability values (beyond human physiological limits), or social media posting times are obviously wrong (e.g., time stamps are in the future). For extreme scores in psychological measurement questionnaires (e.g., SDS scores that are extremely high or low), winsorizing treatments (i.e., setting upper and lower limits, with the excess being replaced by upper and lower limits) are performed to reduce the impact of the extreme values on the analysis. Duplicate rows in the dataset are checked and deleted, ensuring the uniqueness of each record.

Since the score ranges and meanings of different questionnaires may be different, a normalization process is required. For example, the SDS fraction is converted to a Z fraction, which has a mean value of 0 and a standard deviation of 1. And (3) carrying out normalization processing on count data such as posting frequency, interaction times and the like, so that the value range of the count data is scaled to the [0,1] interval. For example, a min-max normalization method is used. And (3) carrying out normalization processing on physiological indexes such as heart rate variability, sleep quality and the like, and ensuring that different data sources are analyzed on the same scale.

Ensuring that the timestamp formats of all data sources are uniform and adjusting the time interval for subsequent time series analysis. For example, all data are integrated by day or hour. And integrating the cleaned, standardized and normalized data according to the user ID and the time stamp to form a multidimensional data set containing psychological characteristics, behavioral characteristics and social characteristics. For example, daily data may include SDS score, posting frequency, average heart rate variability, etc. for each user. For missing values in the integrated dataset, a suitable filling strategy, such as mean filling, median filling, or predictive filling using a machine learning model, is selected based on the data characteristics and the missing proportions.

Through the steps, an integrated multi-source data set is generated, and the data set is high in quality and uniform in format and is suitable for the subsequent identification of psychological health risk factors and abnormal behavior patterns. The integrated data set not only reduces the influence of noise and abnormal values, but also eliminates dimension and dimension difference among different data sources through standardization and normalization processing, and provides a solid foundation for subsequent machine learning analysis and model construction.

Step S202, labeling the integrated multi-source data to generate normal psychological state data and abnormal psychological state data;

Specifically, in step S202, labeling the integrated multi-source data is an important link in the psychological health monitoring and early warning process. The labeling aims at clearly distinguishing the normal psychological state and the abnormal psychological state in the data and providing clear standard and basis for subsequent risk factor identification and abnormal behavior pattern analysis. Labeling is to label each record or data point in the integrated multi-source data as "normal" or "abnormal". And comprehensively judging the psychological state of the user according to the psychological characteristics, the behavioral performance, the physiological indexes and other dimensions of the user. For example, users exhibiting positive emotions, social activity and stable physiological indicators may be marked as normal psychological states; conversely, users exhibiting negative emotions, social withdrawal, and abnormal physiological indicators may be marked as abnormal psychological states. In the labeling process, an automatic auxiliary tool can be adopted to improve the accuracy and efficiency of labeling. These tools can train based on existing mental health datasets, learn the characteristic differences of normal and abnormal mental states, and automatically classify new data. The normal psychological state data and the abnormal psychological state data generated through the labeling process provide clear classification standards for subsequent analysis and model construction. The method is not only beneficial to more accurately identifying the psychological health risk factors and abnormal behavior patterns, but also can improve the prediction performance and reliability of the early warning model, and provides more personalized and effective psychological health early warning intervention schemes for individuals.

In a specific embodiment, in the process of constructing the cardiac health monitoring and early warning system, after the data cleaning, standardization and normalization and integration in step S201, a high-quality integrated multi-source data set is obtained. The integrated multi-source data is then annotated to generate normal mental state data and abnormal mental state data, specifically:

and formulating a definite labeling standard according to professional knowledge and historical data in the psychological health field. These criteria should include, for example, score thresholds for psychometric questionnaires, abnormal performance of social media activities, changes in physiological metrics monitored by the wearable device, and so forth. For example, SDS depression self-score above a certain threshold (e.g. 53 minutes) may be set as an abnormal psychological state, with heart rate variability continuously above the normal range (e.g. SDNN below 50 ms) as a physiological abnormality indicator.

Automated labeling techniques are introduced to improve efficiency. For example, existing annotation data is trained using a machine learning model, allowing the model to learn annotation criteria and automatically predict unlabeled data. And dividing the integrated multi-source data set into two parts of normal psychological state data and abnormal psychological state data according to the labeling result. These two pieces of data will be used for subsequent mental health risk factor identification, abnormal behavioral pattern analysis, and construction of pre-warning models (as shown in table 1).

TABLE 1

User ID	Date of day	SDS fraction	Posting frequency	Heart rate variability	Labeling results
						U001	2024-07-01	45	5	65	Normal state
U002	2024-07-01	60	1	40	Abnormality of
						U003	2024-07-02	52	3	55	Normal state
…	…	…	…	…	…

Through the steps, the integrated multi-source data set is successfully marked, and the normal psychological state data and the abnormal psychological state data are generated, so that the data can provide powerful support for the subsequent psychological health monitoring and early warning work.

Step S203, generating the preprocessing multisource data set based on the normal psychological state data and the abnormal psychological state data;

In particular, these explicitly classified data are integrated into a preprocessing flow to form a preprocessed multi-source dataset that is ultimately used for analysis. The process not only ensures the consistency and accuracy of the data, but also provides a solid foundation for the subsequent identification and model construction of psychological health risk factors. Through this process, noise and redundancy in the data is further reduced, and the structure and format of the data are unified, so that the data set is more suitable for processing by a machine learning algorithm. The generation of the preprocessing multisource data set marks the completion of the data preparation stage, and paves the way for the development of the follow-up intelligent analysis and early warning model. Successful implementation of this step is critical to improving the accuracy and efficiency of the early warning system.

In one embodiment, after data cleaning, normalization and normalization, integration processing and labeling are completed, the labeled integrated multisource data is further consolidated to generate a preprocessed multisource dataset that is ultimately used for analysis. This step is the final step in the data preprocessing flow, and is critical to ensure the validity and accuracy of the subsequent analysis, specifically:

And ordering and integrating all marked data according to key fields such as user IDs, time stamps and the like, so as to ensure the time sequence and consistency of the data. The missing and outliers in the data are checked to ensure that these values have been properly processed according to the method in step S201, resulting in the pre-processed multi-source dataset.

In this embodiment, the step S300 includes:

Step S301, based on the preprocessing multi-source data set, psychological characteristics, behavioral characteristics and social characteristics are obtained;

Specifically, the preprocessed multi-source dataset is deeply mined to extract key psychological, behavioral and social features therefrom. Psychological characteristics may include the emotional state, stress level, cognitive function, etc. of the user, which are typically monitored by psychological measurement tools or physiological indicators. The behavior characteristics cover the daily activity mode, living habit and stress coping modes of the user, and the information can be extracted from data sources such as wearing equipment records of the user, mobile phone use habit and the like. Social features focus on the interaction of users on social media, the size and composition of social networks, and the performance of online communities, which are mainly derived from social media platforms. By comprehensively utilizing various data analysis and mining technologies, the system can comprehensively capture the psychological, behavioral and social states of the user, and provides a rich information basis for subsequent risk assessment and abnormal behavior identification.

Step S302, based on the psychological characteristics, the behavioral characteristics and the social characteristics, adopting a preset method to identify and obtain the psychological health risk factors and the abnormal behavioral patterns;

Specifically, a model is built by using a preset analysis model (the model may be, for example, a long-short-term memory network (LSTM) in the deep learning model in step S500), and the specific method is similar to step S500, and details thereof are omitted herein, so as to perform deep analysis on the extracted psychological characteristics, behavioral characteristics and social characteristics. The analytical model is used to identify potential mental health risk factors and abnormal behavioral patterns. The system evaluates whether the mental state of the user deviates from the normal range by comparing the user characteristics with a known risk index and abnormal behavior pattern library. For example, a sustained high-pressure state, an increase in social isolation behavior, or abnormal mood swings, etc. may be identified as risk factors or abnormal patterns. This process depends not only on the accuracy and integrity of the data, but also on the accuracy and adaptability of the algorithm. Through comprehensive analysis and intelligent judgment, the system can timely discover and report psychological health problems of users, and provides scientific basis for subsequent intervention measures.

For example, a data set containing mental health data of college students is collected through various channels, including psychological measurement tools (such as depression scales and anxiety scales), social media activity records (such as microblog posting frequency and emotion tendencies), physiological indexes monitored by wearing equipment (such as heart rate variability and sleep quality), and the like. These data have been pre-processed to produce a pre-processed multi-source dataset. Step S302 extracts psychological characteristics such as emotional stability and stress level from the psychological measurement tool, extracts behavioral characteristics such as posting frequency, interaction times and emotional tendency from the social media data, and analyzes social characteristics such as social network structure, number of friends and interaction frequency of the user. The extracted features are sorted in time series, ensuring that each feature is time-varying sequence data. The data set is divided into a training set, a validation set and a test set for subsequent training and evaluation of the model. And constructing a psychological health risk factor and abnormal behavior pattern recognition model by using the LSTM network. The LSTM model structure may include a plurality of LSTM layers for capturing long-term dependencies in the sequence data, followed by a fully connected layer (Dense layer) for final classification or regression prediction. The LSTM model is trained using the training set data. The error between the predicted and actual values is calculated by forward propagation and then the network weights are adjusted using a back propagation algorithm to minimize the error. During training, training loss and verification loss are monitored, overfitting is prevented, early stop (early stop) is performed or learning rate is adjusted if necessary. The generalization ability of the model is evaluated using the validation set, and the model parameters or structure are adjusted to improve performance. Feature selection, network architecture or super parameters such as the number of LSTM layers, the number of hidden units, learning rate, etc. are optimized to further improve model accuracy. And predicting the test set data by using the trained LSTM model, and evaluating the performance of the model in practical application. If the model performance meets the requirements, the model performance is applied to the real-time monitoring of the mental health state data of the user, and mental health risk factors and abnormal behavior patterns are identified.

For example, data of a specific user, is analyzed by the LSTM model: psychological measurements of the user over the past month (e.g., weekly depression scale scores), social media activity records (number of posts per day and emotional tendency), physiological indicators monitored by the wearable device (daily heart rate variability). The data of the user is input into a trained LSTM model, and the model outputs the current psychological health state and potential risk factors of the user. The model predictions show that the user's recent depression scale score continues to rise, social media activity decreases, and heart rate variability increases. Based on these features, the model identifies that the user is at risk for moderate depression and indicates that social withdrawal and abnormal physiological response may be risk factors.

In this embodiment, the step S500 includes:

Step S501, monitoring psychological health state data of a user in real time based on the psychological health early warning model, and generating a monitoring result;

Specifically, the system captures information on the emotion change, behavior pattern, social interaction and the like of the user in real time by integrating various data sources, such as sensor data on the user equipment, social media interaction records and possibly psychological assessment test results. This information is then input into a pre-trained mental health pre-warning model for analysis. The model utilizes a complex algorithm to carry out multidimensional assessment on the psychological state of the user, and compares the current data with the preset normal range or historical baseline data so as to identify any possible abnormal or risk signs. Finally, the system generates an exhaustive monitoring result report that not only outlines the current mental health status of the user, but may include, for example, risk level assessment, abnormal index list, and potential cause analysis, providing powerful support for subsequent interventions.

Step S502, when the monitoring result exceeds a preset threshold value, judging that an abnormal condition occurs, and sending out early warning information;

Specifically, when the monitored mental health state data of the user exceeds a preset safety threshold, an early warning mechanism is immediately started, and the abnormal condition of the current mental state of the user is judged. This threshold is set based on a large amount of historical data and psychological expertise, aimed at accurately distinguishing normal fluctuations from psychological health risks that really need attention. Once the occurrence of an abnormal situation is determined, the system rapidly generates early warning information which thoroughly describes the monitored abnormal indexes, risk levels and possible influence ranges, and ensures that a receiver can rapidly understand and react. The early warning information is sent to related parties in real time through communication channels (such as mobile phone short messages, emails or in-application notifications) preset by the user, including the user himself, guardianship, psychological consultants or medical institutions, so as to quickly start intervention measures and protect psychological health and safety of the user. And if the monitoring result does not exceed the preset threshold value, continuing to monitor.

In a particular embodiment, data for a user is automatically collected from a plurality of sources on a periodic basis (e.g., daily or hourly). These data sources may include mental health applications used by the user, social media accounts, and wearable devices. The data collected includes, but is not limited to, emotional self-assessment of the user, frequency of interaction with friends, emotional tendency in posting content, sleep quality reports, heart rate variability, etc. The collected raw data need to be subjected to pretreatment steps such as cleaning, standardization, formatting and the like so as to ensure that the raw data meet the input requirements of a mental health early warning model. For example, text content on social media is converted to emotion scores, data of a wearable device is converted to standardized heart rate variability metrics, and so on. The preprocessed data is input into the mental health pre-warning model in real time. The model analyzes the time sequence data by utilizing the LSTM network and captures the change trend of the psychological health state of the user. The model simultaneously considers data of multiple dimensions, such as emotion stability, social activity, physiological indexes and the like, so as to comprehensively evaluate the psychological health state of the user. And outputting the current psychological health state assessment result of the user by the model. This may be a specific score, rank, or probability distribution indicating the likelihood that the user is in a different physical health state. For example, the model outputs a score between 0 and 100, with higher scores indicating better mental health; or output a classification label such as "normal", "mild stress", "moderate risk of depression", etc. In addition to overall assessment, the model may also provide specific risk factor cues, such as "recent social activity is reduced," there may be a feeling of autism, "or" heart rate variability is increased, possibly under greater stress. The monitoring result is displayed on the platform in real time for the user and guardian or psychological consultant to check. If the monitoring result shows that the mental health state of the user is abnormal or potentially dangerous, the platform automatically triggers an early warning mechanism and sends a notification to related personnel.

For example, at some point in the day, the platform has collected the user's most recent data: the emotion self-evaluation score is lower, social media activities are reduced, learning pressure self-evaluation is improved, and meanwhile heart rate variability is obviously improved due to the fact that data of the wearable device are displayed. These data are input into the mental health pre-warning model in real time. And outputting a monitoring result after model analysis, displaying that the user is in a 'moderate depression risk' state, and particularly indicating that 'recent social activity reduction and physiological index abnormality' can be main risk factors. The platform immediately sends an early warning notice to the user and recommends him to seek professional psychological consultation assistance. At the same time, notifications are also sent to the user's guardian or designated emergency contact.

In this embodiment, the step S600 includes:

Step S601, based on the early warning information, identifying early warning factors and severity of early warning;

Specifically, early warning information is deeply analyzed, and specific factors causing early warning, namely early warning factors, are accurately positioned through data mining and pattern recognition technologies. These early warning factors may include, for example, abnormal patterns of behavior of the user (e.g., social withdrawal, overuse of social media), mood swings (e.g., sustained falls, increased anxiety), or physiological indicators abnormalities (e.g., decreased sleep quality, heart rate changes), etc. Meanwhile, the system can evaluate the severity of the early warning, and the potential influence degree of the early warning on the mental health of the user is determined through quantitative analysis, so that an important reference is provided for subsequent intervention measures. The process combines big data analysis and expert knowledge, ensures the accuracy of early warning factors and the rationality of early warning severity, and lays a solid foundation for the subsequent evaluation of mental health and the establishment of intervention schemes.

Step S602, obtaining multi-dimensional information, and carrying out evaluation of mental health state based on the multi-dimensional information, the early warning factors and the early warning severity degree to generate an evaluation result;

Specifically, the multidimensional information related to the user is comprehensively collected, and the information includes, but is not limited to, personal background information (such as age, gender, occupation, and the like) of the user, past mental health records, life habits, family and social environments, and the like. By integrating the multidimensional information, the system can construct a more complete and three-dimensional user image, and provides rich data support for assessing the psychological health state. And then, the collected multidimensional information is combined with the early warning factors and the early warning severity degree identified before by using an advanced algorithm and model to carry out deep analysis and comparison. The process not only considers the direct influence of early warning factors, but also evaluates the potential chain reaction of the early warning factors to the overall psychological health state of the user. For example, if the early warning factor is a persistent sleep disorder, the system will further discuss its possible negative impact on the user's daytime function, mood adjustment capability, or social relationship. Finally, based on the above comprehensive evaluation, an exhaustive evaluation result is generated. This result not only explicitly indicates the current mental health status of the user, but also details the specific reasons for this status, the potential impact of early warning factors and the possible future trends. Meanwhile, the evaluation result can also comprise targeted suggestions or intervention measures, a personalized mental health improvement scheme is provided for the user, and the overall improvement of the mental health level of the user is promoted.

In an embodiment, before step S601, the method further includes, confirming validity of the early warning information, if valid, executing step S601, and if invalid, not continuing to execute step S601;

in particular, confirming the validity of the pre-warning information is a vital link, as it directly relates to the accuracy and timeliness of the subsequent intervention. The validity confirmation of the early warning information can avoid resource waste and unnecessary panic caused by false alarm or invalid early warning, and the specific method is as follows:

By comparing user behavioral and mental state information collected by different data sources (e.g., social media data, wearable data, mental test data, etc.), it is checked whether there is an abnormal manifestation of consistency. If the data sources are abnormal, the effectiveness of the early warning information is increased. And analyzing the change trend in the historical data of the user, and confirming whether the current early warning information represents a continuous or sudden abnormal mode, but not an occasional event. And the rule base is used for verifying whether the early warning information accords with the known psychological health risk factors and abnormal behavior patterns based on the psychological principles and expert experience. For example, if the alert information indicates that the user's sleep quality is drastically reduced while the heart rate is continuously increasing and the change is continued for several days, this may be a valid alert signal according to rules in the mental health field. In addition, before the early warning information is sent, the current psychological state of the user can be confirmed through interaction of a lightweight questionnaire or instant message with the user, and the current psychological state can be used as a reference for effectiveness of the early warning information. The user feedback can be combined with a threshold value or rule preset by the system, so that the accuracy of early warning is improved. The machine learning model is utilized to classify and predict early warning information in real time, and the model can learn which early warning signals are valid and which are false positives based on a large amount of historical data. The AI-assisted verification mechanism can be continuously self-optimized, and the accuracy of early warning information is improved. For example, when the activity of social media of a college student user is suddenly and greatly reduced, and meanwhile, the wearing equipment records and displays that the night sleep quality is lower than the normal level for a period of time, the heart rate is slightly increased. The system confirms the validity of the early warning information according to the following steps: the system firstly compares the interaction frequency of the small sheets on the social media with the history record to confirm whether the activity reduction is obvious and continuous. Meanwhile, heart rate and sleep quality data recorded by the wearable equipment are checked to confirm whether the two indexes also show abnormal trends. The system applies a preset rule base to find that the sleep quality of the small pieces is reduced and the heart rate is increased for more than one week, and the sleep quality is consistent with the preset rule of sleep disorder caused by long-term psychological pressure accumulation. The system may also ask a brief questionnaire to small Zhang Fasong by applying a message asking if it recently felt increased stress or had other discomfort, such as small Zhang Huifu confirming that it was really anxious and difficult to fall asleep recently. The machine learning model predicts that the matching degree of the current state of the small sheet and the category of the medium psychological stress is as high as 90 percent according to the historical data. By combining the above information, it is possible to confirm that the early warning information about the sheetlet is valid, and then step S601 is performed to further identify the early warning factor and the early warning severity, and generate the evaluation result. Therefore, the system not only can accurately identify the psychological health risk of the user, but also can timely take effective intervention measures.

For example, in the actual real-time monitoring process, early warning information is sent to a user to prompt the user to have moderate depression risk. The platform first reviews and analyzes specific monitoring data that results in the issuance of pre-warnings, including the user's mood self-score, social media activity frequency, learning stress self-score, heart rate variability, etc. And determining the accuracy and reliability of the early warning information, and eliminating possible false alarm factors. Through analysis, the platform recognizes several major early warning factors: the recent emotion self-evaluation score of the user is continuously low and has large fluctuation; social media activity frequency is significantly reduced, and interaction with friends is reduced; the learning pressure self-evaluation result shows that the pressure level continuously rises; the wearable device monitoring data shows an increase in heart rate variability, indicating a physiological stress response. And according to the current mental health state data and the historical comparison of the user, the platform evaluates that the depression risk is at a moderate level. Further analysis, if the emotional state of the user is continuously worsened, or a more obvious abnormality occurs in the physiological index, the early warning severity may be increased. In addition to monitoring data in real time, the platform also collects personal background information of the user, such as age, gender, profession, home environment, etc. The user's historical mental health records are reviewed to see if he has similar psychological problems or intervention experiences in the past. Communicate with the user and learn the factors which may affect mental health, such as recent life events, interpersonal relationship changes, and the like. And the platform comprehensively evaluates the psychological health state of the user by combining the real-time monitoring data, the personal background information, the history record and the communication result. Interactions and potential impacts between early warning factors are analyzed, such as how learning stress affects a user's emotional and social behavior. The user's current mental toughness, coping mechanisms, and effectiveness of the social support system are evaluated. Based on the comprehensive assessment, the platform generates detailed assessment reports indicating major psychological problems (e.g., moderate depression risk), early warning factors, early warning severity, and possible trends in development currently existing for the user. Suggested interventions for the personal situation of the user, such as seeking professional psychological consultation, participating in mental health support teams, adjusting learning pace and social activities, etc. may also be included in the report. The method is not only helpful for timely finding problems and providing early warning support, but also provides scientific basis for subsequent personalized intervention measures.

Example two

Fig. 2 is a schematic structural diagram of a mental health monitoring and early warning system based on big data in a second embodiment of the present invention, and as shown in fig. 2, the second embodiment provides a mental health monitoring and early warning system based on big data, including: a first generation module 201, a second generation module 202, an identification module 203, a construction module 204, an issue module 205, a third generation module 206, and a fourth generation module 207. The first generation module 201 is configured to collect user data through multiple collection channels to generate multi-source data. The second generating module 202 is configured to perform preprocessing on the multi-source data to generate a preprocessed multi-source data set. The identifying module 203 is configured to analyze the preprocessed multi-source data, and identify and obtain a mental health risk factor and an abnormal behavior pattern. The construction module 204 is configured to construct a mental health early warning model based on the mental health risk factors and the abnormal behavioral patterns. The sending module 205 is configured to monitor the mental health status data of the user in real time based on the mental health early warning model, where when the mental health status data of the user is monitored to be abnormal, sending early warning information. The third generating module 206 is configured to perform an evaluation of mental health status based on the early warning information, and generate an evaluation result. The fourth generating module 207 is configured to generate and push a personalized mental health early warning intervention scheme based on the evaluation result.

In this embodiment, the first generating module 201 includes: the obtaining unit and the first generating unit. The obtaining unit is used for collecting the user data through the first collecting channel, the second collecting channel and the third collecting channel to obtain the first user data, the second user data and the third user data. The first generation unit is configured to generate the multi-source data based on the first user data, the second user data, and the third user data.

In this embodiment, the second generating module 202 includes: a second generation unit, a third generation unit and a fourth generation unit. The second generating unit is used for performing data cleaning, data standardization and normalization and data integration processing on the multi-source data to generate integrated multi-source data. The third generating unit is used for labeling the integrated multi-source data and generating normal psychological state data and abnormal psychological state data. The fourth generation unit is configured to generate the preprocessed multi-source dataset based on the normal mental state data and the abnormal mental state data.

In this embodiment, the identification module 203 includes: an acquisition unit and a first identification unit. The acquisition unit is used for acquiring psychological characteristics, behavioral characteristics and social characteristics based on the preprocessing multi-source data set. The first identifying unit is used for identifying and obtaining the psychological health risk factors and the abnormal behavior patterns by adopting a preset method based on the psychological characteristics, the behavior characteristics and the social characteristics.

In this embodiment, the issue module 205 includes: and a fifth generation unit and an emission unit. The fifth generation unit is used for monitoring the mental health state data of the user in real time based on the mental health early warning model and generating a monitoring result. And the sending unit is used for judging that an abnormal condition occurs when the monitoring result exceeds a preset threshold value and sending out early warning information.

In this embodiment, the third generating module 206 includes: a second identification unit and a sixth generation unit. The second identification unit is used for identifying early warning factors and severity of early warning based on the early warning information. The sixth generation unit is configured to obtain multidimensional information, and perform evaluation of mental health status based on the multidimensional information, the early warning factors and the early warning severity degree, so as to generate an evaluation result.

Various changes and specific examples of the big data-based mental health monitoring and early warning method provided in the first embodiment are applicable to the big data-based mental health monitoring and early warning system provided in the present embodiment, and by the foregoing detailed description of the big data-based mental health monitoring and early warning method, a person skilled in the art can clearly know the implementation of the big data-based mental health monitoring and early warning system in the present embodiment, so that the details of the description are omitted herein for brevity.

Example III

Fig. 3 is a schematic structural diagram of an electronic device in a third embodiment of the present invention, and as shown in fig. 3, the third embodiment further provides an electronic device 300, which may include: a processor 301 and a memory 302.

A memory 302 for storing a program; the memory 302 may include volatile memory (english: volatilememory), such as random-access memory (english: RAM), such as static random-access memory (english: SRAM), double data rate synchronous dynamic random-access memory (english: double Data Rate Synchronous Dynamic Random Access Memory, DDR SDRAM), and the like; the memory may also include a non-volatile memory (English) such as a flash memory (English). The memory 302 is used to store computer programs (e.g., application programs, functional modules, etc. that implement the methods described above), computer instructions, etc., which may be stored in one or more of the memories 302 in a partitioned manner. And the above-described computer programs, computer instructions, data, etc. may be invoked by the processor 301.

The computer programs, computer instructions, etc., described above may be stored in one or more of the memories 302 in partitions. And the above-described computer programs, computer instructions, etc. may be invoked by the processor 301.

A processor 301 for executing a computer program stored in a memory 302 to implement the steps of the method according to the above-mentioned embodiment.

Reference may be made in particular to the description of the embodiments of the method described above.

The processor 301 and the memory 302 may be separate structures or may be integrated structures integrated together. When the processor 301 and the memory 302 are separate structures, the memory 302 and the processor 301 may be coupled by a bus 303.

The electronic device in this embodiment may execute the technical scheme in the above method, and the specific implementation process and the technical principle are the same, which are not described herein again.

Example IV

The fourth embodiment also provides a computer readable storage medium, which comprises a computer program and instructions, wherein the computer program or instructions, when run on a computer, cause the computer to execute the big data based mental health monitoring and early warning method according to any embodiment of the present invention.

The computer readable storage medium includes: a usb disk, a removable hard disk, a ROM, a RAM, a magnetic disk, or an optical disk, etc.

The present embodiment also provides a computer program product comprising: a computer program stored in a readable storage medium, from which at least one processor of an electronic device can read, the at least one processor executing the computer program causing the electronic device to perform the solution provided by any one of the embodiments described above.

It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present disclosure may be performed in parallel, sequentially, or in a different order, provided that the desired results of the technical solution of the present disclosure are achieved, and are not limited herein.

In a word, the psychological health monitoring and early warning method and system based on big data can comprehensively analyze multi-source data, judge the psychological health state of an individual more accurately, and reduce the probability of false alarm and false alarm; the psychological health state of the individual can be monitored in real time, abnormal conditions can be found in time, and the possibility is provided for early intervention; personalized psychological health early warning intervention schemes can be formulated according to specific conditions and data analysis results of individuals, and early warning intervention effects are improved.

Note that the above is only a preferred embodiment of the present invention and the technical principle applied. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, while the invention has been described in connection with the above embodiments, the invention is not limited to the embodiments, but may be embodied in many other equivalent forms without departing from the spirit or scope of the invention, which is set forth in the following claims.

Claims

1. The psychological health monitoring and early warning method based on big data is characterized by comprising the following steps of:

2. The big data based mental health monitoring and early warning method according to claim 1, wherein the collecting the user data through a plurality of collecting channels, and generating the multi-source data comprises:

3. The big data based mental health monitoring and early warning method of claim 1, wherein the preprocessing the multi-source data to generate a preprocessed multi-source data set comprises:

4. The big data-based mental health monitoring and early warning method according to claim 1, wherein the analyzing based on the preprocessed multi-source data, and identifying the mental health risk factors and the abnormal behavior patterns comprise:

5. The big data-based mental health monitoring and early warning method according to claim 1, wherein the mental health monitoring and early warning model monitors mental health state data of the user in real time, and wherein when the mental health state data of the user is monitored to be abnormal, sending early warning information comprises:

6. The big data-based mental health monitoring and early warning method according to claim 1, wherein the performing the evaluation of the mental health state based on the early warning information, generating the evaluation result includes:

7. The psychological health monitoring and early warning system based on big data is characterized by comprising:

the first generation module is used for collecting user data through various collection channels and generating multi-source data;

the second generation module is used for preprocessing the multi-source data and generating a preprocessed multi-source data set;

The identification module is used for analyzing based on the preprocessing multi-source data and identifying and obtaining psychological health risk factors and abnormal behavior patterns;

the construction module is used for constructing a psychological health early warning model based on the psychological health risk factors and the abnormal behavior mode;

The sending module is used for monitoring the psychological health state data of the user in real time based on the psychological health early warning model, wherein when the psychological health state data of the user is monitored to be abnormal, early warning information is sent out;

The third generation module is used for carrying out evaluation of the mental health state based on the early warning information and generating an evaluation result;

and the fourth generation module is used for generating and pushing a personalized psychological health early warning intervention scheme based on the evaluation result.

8. The big data based mental health monitoring and early warning system of claim 7, wherein the first generation module comprises:

The acquisition unit is used for acquiring the user data through the first collection channel, the second collection channel and the third collection channel to acquire first user data, second user data and third user data;

A first generation unit configured to generate the multi-source data based on the first user data, the second user data, and the third user data.

9. An electronic device, comprising:

At least one processor; and

A memory communicatively coupled to the at least one processor;

Wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the big data based mental health monitoring and pre-warning method of any of claims 1-6.

10. A computer readable storage medium comprising a computer program and instructions which, when run on a computer, cause the computer to perform the big data based mental health monitoring and pre-warning method according to any of claims 1-6.