Intelligent pressure emotion management method and system based on multi-dimensional behavior style feature evaluation

Abstract

The invention discloses an intelligent pressure emotion management method and system based on multi-dimensional behavior style feature evaluation, and relates to the field of pressure emotion management. Firstly, body and behavior information of a new user is collected and preprocessed, and a pressure data set is generated. And through similarity analysis with historical user data, a group most similar to the new user is found out. On the basis of the pressure release schemes of the similar groups, scheme vectors are constructed, users are divided into different similar groups according to the schemes, and proportion distribution of each group is recorded; thirdly, social attributes of the new users are analyzed, and the matching degree between the new users and all the groups is calculated; and finally, performing weighted summation on the matching degree and the group proportion distribution to obtain a conformity coefficient, and selecting the most suitable pressure release scheme of the new user according to the coefficient. According to the method, the user behavior style can be evaluated in multiple dimensions, a personalized pressure release scheme is recommended, and the management efficiency and the user satisfaction degree are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of stress emotion management, and more specifically, to an intelligent stress emotion management method and system based on multi-dimensional behavioral style feature evaluation. Background Art

[0002] The intelligent stress and emotion management method and system based on multi-dimensional behavioral style characteristics assessment is an innovative method that integrates psychology, behavior, artificial intelligence and other technologies. It aims to evaluate the behavioral style characteristics of individuals through multi-dimensional means and manage and regulate stress and emotions through intelligent means. This type of system has broad application prospects in modern society, especially in the fields of workplace management, mental health intervention, education, and emotional management under high-pressure environments.

[0003] The existing technology is unable to evaluate the user's behavior style in multiple dimensions, there are large errors in the user's evaluation process, and it is impossible to recommend personalized solutions to relieve stress.

[0004] In order to solve the above defects, a technical solution is now provided. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present invention provide an intelligent stress emotion management method and system based on multi-dimensional behavior style feature evaluation to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions: The intelligent stress and emotion management method based on multi-dimensional behavioral style characteristic assessment includes the following steps: Step S1: collecting the physical information data, behavioral information data, and electronic medical record data of the new user, and preprocessing them to generate a preprocessed stress data set; Step S2: Based on the pre-processed pressure data set of the new user, similarity matching is performed with historical users to determine a set of historical users with high similarity; Step S3: Determine the user's stress relief plan, divide the stress relief plans of users with high similarity into similar groups by cluster analysis, and record the proportion distribution of the groups; Step S4: Determine the social attributes of the user, calculate the matching degree between the new user and the divided users; calculate the matching coefficient by weighted summation based on the matching degree between the new user and the divided users and the group proportion distribution, and determine the output of the pressure release plan with the largest matching coefficient.

[0007] In a preferred embodiment, step S1 includes the following process: Step S11: The collected body information data, behavior information data, and electronic medical record data form an original pressure data set R; the body information data includes heart rate and skin electricity; the behavior information data includes expression, posture, and movement; and the electronic medical record data includes the patient's historical medical records and treatment plans; Step S12: performing denoising processing on the original pressure data set R, filtering out noise signals in the data by a denoising algorithm, and generating a denoised pressure data set; Step S13: performing data cleaning on the denoised pressure data set, removing missing values, abnormal values ​​and duplicate values ​​in the denoised pressure data set, and generating a cleaned pressure data set; Step S14: performing format conversion on the cleaning pressure data set, converting the behavior information data from different sources into structured data, and generating a pressure data set after format conversion; Step S15: standardizing the pressure data set after format conversion, normalizing all numerical data, mapping the data to a unified scale range, and generating a standardized pressure data set; Specifically, use the following formula to convert the data Normalized to the range [0,1]: ;in is the normalized value, and The data The minimum and maximum values ​​of the data is the data in the dataset R.

[0008] In a preferred embodiment, in step S2, the standardized pressure data set of each user is feature vectorized, and the pressure state vector is constructed for the historical user to construct the following vector: ; Represents the value of the mth feature of the i-th user; construct a feature vector for the new user: .

[0009] In a preferred embodiment, historical users and new users are determined to construct feature vectors and cosine similarity is used to calculate the similarity between historical users and new users: ; In the formula, k is the feature index, which ranges from 1 to m, traversing all features; determine the similarity between historical users and new users, and filter out users whose similarity is greater than the system threshold.

[0010] In a preferred embodiment, in step S3, similarity groups are divided according to different stress release schemes of users with high similarity, and different stress release scheme feature groups are identified, which is accomplished by using a K-means clustering method, specifically including the following steps: Step S31: Initialize the cluster center, select K cluster centers, and the K value can be determined by methods such as the elbow rule; each center ck is an n-dimensional vector: ck=(ck1,ck2,...,ckn); where k represents the kth cluster center, k=1,2,...,K; Step S32: randomly selecting K samples from the user pressure release scheme vector data set as initial cluster centers, which are the center points of the initial assumed clusters; Step S33: Assign data points to the nearest cluster center. For each user di, calculate its Euclidean distance to all K cluster centers: ; The distance between the i-th user and the k-th cluster center; is the mth feature of the i-th data, is the mth feature of the cluster center; assign the user to the cluster with the nearest center; Step S34: Update the cluster center, calculate the mean vector of each cluster to update the cluster center, which can be calculated by the following formula: ; Indicates that it currently belongs to the cluster The number of users; Step S35: If the change in the center point after the update is compared with the previous center point is less than a certain set threshold value ϵ (i.e., the convergence condition), the algorithm ends.

[0011] In a preferred embodiment, the number of users in the cluster where each user is located is recorded, and the group ratio distribution of each user is equal to the number of users in the cluster where the user is located divided by the total number of users in all clusters.

[0012] In a preferred embodiment, in step S4, the social attributes of the user are determined, and the matching degree between the new user and the divided users is calculated by cosine similarity based on the social attributes of the user; the matching degree between the new user and the divided users and the group proportion distribution are comprehensively weighted and summed to calculate the matching coefficient: F=αPD+βZB; wherein F represents the matching coefficient, PD represents the matching degree between the new user and the divided users, and ZB represents the group proportion distribution; α and β represent the matching degree between the new user and the divided users and the weights of the group proportion distribution, respectively, and the pressure release plan for the user with the largest matching coefficient is determined and output as the recommended plan.

[0013] In a preferred embodiment, it includes: a data collection module, a preliminary data processing module, a screening module, a recommendation module, and a storage module; The data collection module is used to collect the user's physical information data, behavioral information data, electronic medical record data, and user social attributes; transmit the user's physical information data, behavioral information data, and electronic medical record data to the data preliminary processing module, and transmit the user's social attributes to the recommendation module; The data preliminary processing module is used to process the user's physical information data, behavioral information data, and electronic medical record data transmitted by the data acquisition module, and perform denoising, cleaning, and normalization on the user's physical information data, behavioral information data, and electronic medical record data, and merge them into a user stress data set; and transmit the user stress data set to the screening module; The screening module is used to calculate the similarity between historical users and new users based on the pressure data sets of new users and historical users, and screen out a user set whose similarity is greater than a threshold; and transmit the screened user set to the recommendation module; The recommendation module is used to perform cluster analysis on the selected user sets according to different pressure release schemes, and then determine the group proportion distribution; calculate the matching degree between the new user and the divided users through cosine similarity according to the user social attributes; calculate the matching coefficient by weighted summation based on the matching degree between the new user and the divided users and the group proportion distribution, and determine the pressure release scheme of the user with the largest matching coefficient as the recommended scheme output; The storage module is used to store data during system operation.

[0014] Technical effects and advantages of the present invention: The present invention first collects the physical and behavioral information data of the new user, and pre-processes it to generate a stress data set. Then, the data set is used to perform similarity analysis with the historical user data to find the historical user group that is most similar to the new user. Then, based on the stress release solutions of these similar historical users, a solution vector is constructed, and these users are divided into different similarity groups according to their stress release solutions, and the proportion distribution of each group is recorded. Next, by analyzing the social attributes of the new user, the matching degree between the new user and each divided group is calculated. Finally, the matching degree between the new user and the group and the group proportion distribution are weighted and summed to calculate a compliance coefficient, and the stress release solution with the highest compliance is selected based on this coefficient. The user's behavior style can be evaluated in multiple dimensions, and the recommended stress release solution is more suitable for the user. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings; Figure 1 It is a flow chart of the intelligent stress emotion management method based on multi-dimensional behavior style feature assessment of the present invention; Figure 2It is a schematic diagram of the structure of the intelligent stress and emotion management system based on multi-dimensional behavioral style feature evaluation of the present invention. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.

[0017] First, collect the physical and behavioral information data of new users and preprocess them to generate a stress data set. Then, use this data set to perform similarity analysis with historical user data to find the historical user group that is most similar to the new user. Then, based on the stress release solutions of these similar historical users, construct a solution vector, and divide these users into different similarity groups according to their stress release solutions, and record the proportion distribution of each group. Next, by analyzing the social attributes of new users, calculate the matching degree between the new user and each divided group. Finally, perform a weighted summation of the matching degree between the new user and the group and the group proportion distribution, calculate a compliance coefficient, and select the stress release solution with the highest compliance based on this coefficient.

[0018] Example 1: Figure 1 As shown, the intelligent stress emotion management method based on multi-dimensional behavioral style feature assessment of the present invention includes the following steps: Step S1: collecting the physical information data, behavioral information data, and electronic medical record data of the new user, and preprocessing them to generate a preprocessed stress data set; Step S2: Based on the pre-processed pressure data set of the new user, similarity matching is performed with historical users to determine a set of historical users with high similarity; Step S3: Determine the user's stress relief plan, divide the stress relief plans of users with high similarity into similar groups by cluster analysis, and record the proportion distribution of the groups; Step S4: Determine the social attributes of the user, calculate the matching degree between the new user and the divided users; calculate the matching coefficient by weighted summation based on the matching degree between the new user and the divided users and the group proportion distribution, and determine the output of the pressure release plan with the largest matching coefficient.

[0019] Specifically, in step S1, the user carries a watch with him / her to monitor the user's physical information at all times, including heart rate, skin electricity, etc. Heart rate refers to the number of heartbeats per unit time, usually expressed in beats per minute (bpm). It reflects the intensity of cardiac activity, which is controlled by the autonomic nervous system, especially the balance between the sympathetic nerves (responsible for the "fight or flight" response) and the parasympathetic nerves (responsible for the "rest and digest" response). Increased sympathetic nerve activity: Under emotional conditions such as stress, anxiety, and tension, sympathetic nerve activity increases and heart rate usually increases. This is because under the action of the sympathetic nerves, the "stress response" in the body is activated, preparing the body to respond quickly to external threats. Decreased parasympathetic nerve activity: On the contrary, under the action of the parasympathetic nerves, the heart rate will decrease and the body will be in a relaxed state. When stressed or anxious, the inhibitory effect of the parasympathetic nerves weakens, resulting in an increased heart rate.

[0020] Galvanic skin response (EDA), also known as skin conductance response (SCR), refers to the change in the conductivity of the skin surface, usually obtained by measuring the resistance of the skin. This response is directly related to the activity of the body's sweat glands, which is mainly controlled by the sympathetic nerves. Sympathetic nerve activation: When under stress, anxiety or tension, sympathetic nerve activity increases, resulting in more sweat secretion, which increases the conductivity of the skin. Therefore, the galvanic skin response will increase significantly. This is due to the stress response caused by anxiety or stress, which leads to increased sweat gland secretion and increased skin conductivity. Resting state: When the individual is relaxed, the sympathetic nerve activity is low and the galvanic skin response is also low, which is manifested as lower skin conductivity.

[0021] The user's behavior information, including expressions, postures, movements, etc., is constantly monitored through monitoring probes. Facial expressions are one of the most direct ways to reflect emotions. Some stress-related facial changes often manifest as: Tight jaw or facial muscles: People who are stressed may unconsciously tighten their jaw or lips, especially in the chin area. The way a person stands or sits can indicate their stress level, often in an unnatural or defensive posture. Gestures and other body movements are also important signs of stress reactions: Clenched fists: Intense stress is often reflected in clenched fists, which indicates an unconscious confrontational posture. Tapping objects: Tapping on the table, tapping a pen, or making other repetitive movements are often signs of anxiety or tension. Excessive hand movements: If a person frequently waves their arms or makes exaggerated gestures while talking, it may be a nervous reaction to their stress.

[0022] Obtain electronic medical record data through historical information, including the patient's historical medical records and treatment plans. Analyze whether the user has other mental illnesses, thereby indirectly judging the user's psychological stress.

[0023] Further, step S1 includes the following process: Step S11: The collected body information data, behavior information data, and electronic medical record data form an original pressure data set R; the body information data includes heart rate and skin electricity; the behavior information data includes expression, posture, and movement; and the electronic medical record data includes the patient's historical medical records and treatment plans; Step S12: performing denoising processing on the original pressure data set R, filtering out noise signals in the data by a denoising algorithm, and generating a denoised pressure data set; Step S13: performing data cleaning on the denoised pressure data set, removing missing values, abnormal values ​​and duplicate values ​​in the denoised pressure data set, and generating a cleaned pressure data set; Step S14: performing format conversion on the cleaning pressure data set, converting the behavior information data from different sources into structured data, and generating a pressure data set after format conversion; Step S15: Standardize the pressure data set after format conversion, normalize all numerical data, map the data to a unified scale range, and generate a standardized pressure data set.

[0024] Specifically, use the following formula to convert the data Normalized to the range [0,1]: ;in is the normalized value, and The data The minimum and maximum values ​​of the data is the data in the dataset R.

[0025] In step S2, the standardized stress data set of each user is feature vectorized, and the stress state vector of the historical user is constructed as follows: ; Represents the value of the mth feature of the i-th user; construct a feature vector for the new user: .

[0026] Determine historical users and new users to construct feature vectors and use cosine similarity to calculate the similarity between historical users and new users: ; Where k is the feature index, which ranges from 1 to m, traversing all features, so that the cosine similarity formula can calculate the similarity between multi-dimensional feature vectors.

[0027] Determine the similarity between historical users and new users, and filter out users whose similarity is greater than the system threshold. Different users correspond to different pressure release plans.

[0028] In step S3, the stress relief scheme of each user is feature vectorized; the user stress relief scheme vector is determined: ; Represents the pressure release plan for the i-th user, Represents the value of the nth feature of the i-th user; The user pressure relief program mainly includes the following aspects: Cognitive regulation: such as helping individuals adjust their emotional responses through mindfulness meditation and emotional reshaping; behavioral intervention: such as relaxation exercises, exercise, social support and other activities to relieve stress; physiological relaxation training: such as deep breathing, meditation, yoga, etc. Different users may choose different ways to release stress in combination, so that they can release their stress faster and better, and continue to devote themselves to their work, study and life.

[0029] By dividing similarity groups based on the different stress release solutions of users with high similarity, and identifying different characteristic groups of stress release solutions, the K-means clustering method can be used to accomplish this. K-means is an unsupervised learning algorithm that aims to divide data into K clusters. Its goal is to make the sample similarity within each cluster as high as possible, while the sample differences between different clusters are as large as possible. The main steps of this process include selecting the initial center point, assigning samples to corresponding clusters based on distance, and updating the cluster center point until the algorithm converges. The specific steps include: Step S31: Initialize the cluster center, select K cluster centers, and the K value can be determined by methods such as the elbow rule; each center ck is an n-dimensional vector: ck=(ck1,ck2,...,ckn); where k represents the kth cluster center, k=1,2,...,K; Step S32: randomly selecting K samples from the user pressure release scheme vector data set as initial cluster centers, which are the center points of the initial assumed clusters; Step S33: Assign data points to the nearest cluster center. For each user di, calculate its Euclidean distance to all K cluster centers: ; The distance between the i-th user and the k-th cluster center; is the mth feature of the i-th data, is the mth feature of the cluster center; assign the user to the cluster with the nearest center; Step S34: Update the cluster center, calculate the mean vector of each cluster to update the cluster center, which can be calculated by the following formula: ; Indicates that it currently belongs to the cluster The number of users.

[0030] Step S35: If the change in the center point after the update is compared with the previous center point is less than a certain set threshold value ϵ (i.e., the convergence condition), the algorithm ends.

[0031] The number of users in each user's cluster is recorded. The group proportion distribution of each user is equal to the number of users in the user's cluster divided by the total number of users in all clusters.

[0032] In step S4, determine the user's social attributes, including age, occupation, etc.; enter the system according to the user's own; calculate the matching degree between the new user and the divided users through cosine similarity according to the user's social attributes; the calculation method is consistent with the above, and will not be repeated here. The weighted sum of the matching degree between the new user and the divided users and the group proportion distribution is used to calculate the matching coefficient: F=αPD+βZB; where F represents the matching coefficient, PD represents the matching degree between the new user and the divided users, and ZB represents the group proportion distribution; α and β represent the matching degree between the new user and the divided users, and the weights of the group proportion distribution, respectively, and the appropriate weights are selected independently according to the actual situation. Determine the pressure release plan for the user with the largest matching coefficient and output it as the recommended plan.

[0033] Embodiment 2: An intelligent stress and emotion management system based on multi-dimensional behavioral style feature assessment includes the following modules: a data acquisition module, a preliminary data processing module, a screening module, a recommendation module, and a storage module; The data collection module is used to collect the user's physical information data, behavioral information data, electronic medical record data, and user social attributes; transmit the user's physical information data, behavioral information data, and electronic medical record data to the data preliminary processing module, and transmit the user's social attributes to the recommendation module; The data preliminary processing module is used to process the user's physical information data, behavioral information data, and electronic medical record data transmitted by the data acquisition module, and perform denoising, cleaning, and normalization on the user's physical information data, behavioral information data, and electronic medical record data, and merge them into a user stress data set; and transmit the user stress data set to the screening module; The screening module is used to calculate the similarity between historical users and new users based on the pressure data sets of new users and historical users, and screen out a user set whose similarity is greater than a threshold; and transmit the screened user set to the recommendation module; The recommendation module is used to perform cluster analysis on the selected user sets according to different pressure release schemes, and then determine the group proportion distribution; calculate the matching degree between the new user and the divided users through cosine similarity according to the user social attributes; calculate the matching coefficient by weighted summation based on the matching degree between the new user and the divided users and the group proportion distribution, and determine the pressure release scheme of the user with the largest matching coefficient as the recommended scheme output; The storage module is used to store data during system operation.

[0034] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0035] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0036] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0037] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0038] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An intelligent stress and emotion management method based on multi-dimensional behavioral style characteristic assessment, characterized by: The following steps are involved: Step S1: collecting the physical information data, behavioral information data, and electronic medical record data of the new user, and preprocessing them to generate a preprocessed stress data set; Step S2: Based on the pre-processed pressure data set of the new user, similarity matching is performed with historical users to determine a set of historical users with high similarity; Step S3: Determine the user's stress relief plan, divide the stress relief plans of users with high similarity into similar groups by cluster analysis, and record the proportion distribution of the groups; Step S4: Determine the social attributes of the user, calculate the matching degree between the new user and the divided users; perform weighted summation to calculate the matching coefficient by comprehensively considering the matching degree between the new user and the divided users and the group proportion distribution, and determine the output of the pressure release plan with the largest matching coefficient; In step S4, the social attributes of the user are determined, and the matching degree between the new user and the divided users is calculated by cosine similarity based on the social attributes of the user; the matching degree between the new user and the divided users and the group proportion distribution are comprehensively weighted and summed to calculate the matching coefficient: F=αPD+βZB; wherein F represents the matching coefficient, PD represents the matching degree between the new user and the divided users, and ZB represents the group proportion distribution; α and β represent the matching degree between the new user and the divided users and the weights of the group proportion distribution, respectively, and the pressure release plan for the user with the largest matching coefficient is determined and output as the recommended plan.

2. The intelligent stress and emotion management method based on multi-dimensional behavioral style feature assessment according to claim 1 is characterized by: Step S1 includes the following process: Step S11: The collected body information data, behavior information data, and electronic medical record data form an original pressure data set R; Among them, physical information data includes heart rate and skin electricity; behavioral information data includes expression, posture, and movement; electronic medical record data includes the patient's historical medical records and treatment plans; Step S12: performing denoising processing on the original pressure data set R, filtering out noise signals in the data by a denoising algorithm, and generating a denoised pressure data set; Step S13: performing data cleaning on the denoised pressure data set, removing missing values, abnormal values ​​and duplicate values ​​in the denoised pressure data set, and generating a cleaned pressure data set; Step S14: performing format conversion on the cleaning pressure data set, converting the behavior information data from different sources into structured data, and generating a pressure data set after format conversion; Step S15: standardizing the pressure data set after format conversion, normalizing all numerical data, mapping the data to a unified scale range, and generating a standardized pressure data set; Specifically, use the following formula to convert the data Normalized to the range [0,1]: ;in is the normalized value, and Data The minimum and maximum values ​​of the data is the data in the dataset R.

3. The intelligent stress and emotion management method based on multi-dimensional behavioral style feature assessment according to claim 1 is characterized by: In step S2, the standardized stress data set of each user is feature vectorized, and the stress state vector of the historical user is constructed as follows: ; Represents the value of the mth feature of the i-th user; construct a feature vector for the new user: .

4. The intelligent stress and emotion management method based on multi-dimensional behavioral style feature assessment according to claim 3 is characterized by: Determine historical users and new users to construct feature vectors and use cosine similarity to calculate the similarity between historical users and new users: ; In the formula, k is the feature index, which ranges from 1 to m, traversing all features; determine the similarity between historical users and new users, and filter out users whose similarity is greater than the system threshold.

5. The intelligent stress and emotion management method based on multi-dimensional behavioral style feature assessment according to claim 1 is characterized by: In step S3, the similarity groups are divided according to the different stress release schemes of users with high similarity, and different stress release scheme feature groups are identified. This is accomplished by using the K-means clustering method, which specifically includes the following steps: Step S31: Initialize the cluster center, select K cluster centers, and the K value can be determined by methods such as the elbow rule; each center ck is an n-dimensional vector: ck=(ck1,ck2,...,ckn); where k represents the kth cluster center, k=1,2,...,K; Step S32: randomly selecting K samples from the user pressure release scheme vector data set as initial cluster centers, which are the center points of the initial assumed clusters; Step S33: Assign data points to the nearest cluster center. For each user di, calculate its Euclidean distance to all K cluster centers: ; The distance between the i-th user and the k-th cluster center; is the mth feature of the i-th data, is the mth feature of the cluster center; assign the user to the cluster with the nearest center; Step S34: Update the cluster center, calculate the mean vector of each cluster to update the cluster center, which can be calculated by the following formula: ; Indicates that it currently belongs to the cluster The number of users; Step S35: If the change in the center point after the update is compared with the previous center point is less than a certain set threshold value ϵ, the algorithm ends.

6. The intelligent stress and emotion management method based on multi-dimensional behavioral style feature assessment according to claim 5 is characterized by: The number of users in each user's cluster is recorded. The group proportion distribution of each user is equal to the number of users in the user's cluster divided by the total number of users in all clusters.

7. An intelligent stress and emotion management system based on multi-dimensional behavioral style feature evaluation, used to implement the intelligent stress and emotion management method based on multi-dimensional behavioral style feature evaluation as described in any one of claims 1 to 6, characterized in that: include: Data collection module, data preliminary processing module, screening module, recommendation module, storage module; The data collection module is used to collect the user's physical information data, behavioral information data, electronic medical record data, and user social attributes; transmit the user's physical information data, behavioral information data, and electronic medical record data to the data preliminary processing module, and transmit the user's social attributes to the recommendation module; The data preliminary processing module is used to process the user's physical information data, behavioral information data, and electronic medical record data transmitted by the data acquisition module, and perform denoising, cleaning, and normalization on the user's physical information data, behavioral information data, and electronic medical record data, and merge them into a user stress data set; and transmit the user stress data set to the screening module; The screening module is used to calculate the similarity between historical users and new users based on the pressure data sets of new users and historical users, and screen out the user set whose similarity is greater than a threshold; Transmit the filtered user set to the recommendation module; The recommendation module is used to perform cluster analysis on the selected user sets according to different pressure release schemes, and then determine the group proportion distribution; calculate the matching degree between the new user and the divided users through cosine similarity according to the user social attributes; calculate the matching coefficient by weighted summation based on the matching degree between the new user and the divided users and the group proportion distribution, and determine the pressure release scheme of the user with the largest matching coefficient as the recommended scheme output; The storage module is used to store data during system operation.

Images