CN-115938600-B - Psychological health state prediction method and system based on association analysis

CN115938600BCN 115938600 BCN115938600 BCN 115938600BCN-115938600-B

Abstract

The invention belongs to the field of psychological health assessment, and provides a psychological health state prediction method and system based on association analysis, wherein the method comprises the steps of performing privacy calculation based on federal learning to obtain psychological assessment original data; preprocessing psychological assessment raw data to obtain preprocessed psychological assessment data, scanning based on the preprocessed psychological assessment data to create a two-dimensional storage matrix, grouping the two-dimensional storage matrix to obtain a data set, constructing a frequent tree based on the data set to perform association analysis to obtain a strong association rule table meeting minimum support, selecting feature dimension construction psychological features with strong association with other factors according to the strong association rule table, and predicting psychological health states by using a trained psychological health state prediction model. By changing the data set storage mode and the scanning mode, the strong association rule can be obtained by only scanning the data set once, so that the mining efficiency of the association rule algorithm is improved while the storage space of the database is saved.

Inventors

CHEN ZHENXIANG
WANG ZHENGLI
JIANG XIAOQING
LIU WENJUAN
Wang Youmian
WANG HUCHENG
HU BIN
WANG PEICHENG

Assignees

济南大学

Dates

Publication Date: 20260508
Application Date: 20221209

Claims (8)

1. A psychological health state prediction method based on association analysis, comprising: The method comprises the steps of obtaining initial psychological health evaluation results of a plurality of testers through an evaluation system, uploading the initial psychological health evaluation results of the testers to a server, giving a preliminary psychological health evaluation global model of the testers, independently calculating model parameters locally by using sample sets of the testers respectively, encrypting parameter information and sending the parameter information to the server, carrying out safe aggregation on the encrypted parameter information by using a weighted average algorithm based on homomorphic encryption by the server, updating the psychological health evaluation global model of the testers, returning the aggregated parameter information to the evaluation system in an encryption mode, decrypting the encrypted aggregated parameter information, updating the local model by using the decrypted parameter information, entering the next round of training, and carrying out iterative circulation until a loss function converges, and carrying out aggregation on the parameter information calculated locally by the last round by the server, outputting the model result, and generating the psychological evaluation initial data after privacy calculation of the system; preprocessing the psychological assessment original data to obtain preprocessed psychological assessment data; The method comprises the steps of obtaining a psychological assessment experiment data set D and a minimum support degree count of n based on the preprocessed psychological assessment data, starting scanning for the first time on the psychological assessment experiment data set D, recording 1 in the psychological assessment data two-dimensional matrix if the factor score is more than or equal to 2, otherwise recording 0, adding a sum column in the last column of the psychological assessment data two-dimensional matrix for counting the number of the records 1, deleting the data which do not meet the minimum support degree in the psychological assessment data two-dimensional matrix to obtain a two-dimensional storage matrix meeting the minimum support degree n, and performing grouping scanning again based on the two-dimensional storage matrix meeting the minimum support degree n to obtain a data set; Constructing a frequent tree based on the data set to perform association analysis, so as to obtain a strong association rule table meeting the minimum support degree; and selecting feature dimensions with strong relevance to other factors to construct psychological features according to the strong relevance rule table, and predicting the psychological health state by using a trained psychological health state prediction model.
2. The method for predicting mental health based on association analysis as set forth in claim 1, wherein the preprocessing of the raw mental assessment data to obtain preprocessed mental assessment data includes: deleting non-key information in the psychological assessment original data, reserving a field with strong correlation with the psychological health state, and reducing the data storage quantity; deleting or filling the missing value, and deleting redundant data; And uniformly coding the Chinese fields and the factor scores respectively to obtain the preprocessed psychological assessment data.
3. The method for predicting mental health based on associative analysis according to claim 1, wherein said performing a group scan again based on the two-dimensional memory matrix satisfying the minimum support n to obtain a data set comprises: scanning based on the two-dimensional storage matrix meeting the minimum support degree n, scanning n1 columns, and continuing downward scanning until the value corresponding to m rows of the column is 0 because the value corresponding to m1n1 is 1, finishing scanning, and establishing groups (s 1, s2, s3..); Continuing to scan the next column, if the column has a value other than 0, continuing to scan the non-0 position until the column is finished, and continuing to scan the next column, if the column average value is 0, automatically scanning the next column until the second scanning of the two-dimensional storage matrix meeting the minimum support degree n is finished; and finishing scanning grouping to obtain a grouped data set.
4. The method for predicting mental health based on association analysis according to claim 1, wherein said constructing a frequent tree based on a dataset for association analysis results in a strong association rule table satisfying a minimum support degree, comprising: based on the data set, a root node is established and inserted into the FP-Tree, if the subsequent traversal is not empty, the relationship between the node and the node to be inserted is judged, and the construction of the psychological assessment factor FP-Tree is completed; And carrying out frequent pattern mining on the FP-Tree by calling the FP-growth function to obtain strong association rules among different dimensions meeting the minimum confidence and the minimum support in the psychological assessment experimental data set D, and outputting a strong association rule table containing the front item, the rear item, the support and the confidence.
5. The method for predicting mental health based on association analysis as set forth in claim 1, wherein selecting feature dimensions with strong association with other factors to construct mental features according to the strong association rule table, and predicting mental health using the trained mental health prediction model, comprises: According to the obtained association rule table, performing data dimension reduction by filtering low variance features, screening features with obvious differences from mental health states by variance homogeneity test, obtaining feature dimensions with strong association with other factors, and constructing mental features; based on psychological characteristics, the trained psychological health state prediction model based on XG-Boost is utilized to predict the psychological health state.
6. A mental health prediction system based on association analysis, comprising: The system comprises a data acquisition module, a server, an evaluation system, a judgment system and a system generation module, wherein the data acquisition module is configured to carry out privacy calculation based on federal learning to obtain psychological evaluation original data, the data acquisition module is configured to acquire a plurality of testers, finish initial psychological health evaluation results of the testers through the evaluation system, upload the initial psychological health evaluation results of the testers to the server, give a preliminary psychological health evaluation global model of the testers, respectively independently calculate model parameters locally by using different tester sample sets, encrypt parameter information and send the parameter information to the server, carry out safe aggregation on the encrypted parameter information by using a weighted average algorithm based on homomorphic encryption, update the psychological health evaluation global model of the testers, return the aggregated parameter information to the evaluation system in an encryption mode, decrypt the encrypted and aggregated parameter information, update the local model by using the decrypted parameter information, enter the next round of training, iterate until a loss function converges; The data preprocessing module is configured to preprocess the psychological assessment original data to obtain preprocessed psychological assessment data; The data grouping module is configured to scan based on the preprocessed psychological assessment data to create a two-dimensional storage matrix and group the two-dimensional storage matrix to obtain a data set, and comprises the steps of acquiring a psychological assessment experiment data set D and a minimum support count of n based on the preprocessed psychological assessment data, starting scanning for the first time on the psychological assessment experiment data set D, recording 1 in the psychological assessment data two-dimensional matrix if the factor score is more than or equal to 2, otherwise recording 0, adding a sum column to the last column of the psychological assessment data two-dimensional matrix for counting the number of the records 1, deleting the data which does not meet the minimum support in the psychological assessment data two-dimensional matrix to obtain a two-dimensional storage matrix which meets the minimum support n, and performing grouping scanning again based on the two-dimensional storage matrix which meets the minimum support n to obtain the data set; The association analysis module is configured to construct a frequent tree based on the data set to perform association analysis so as to obtain a strong association rule table meeting the minimum support degree; And the psychological assessment module is configured to select feature dimensions with strong relevance to other factors to construct psychological features according to the strong relevance rule table, and predict the psychological health state by using the trained psychological health state prediction model.
7. A computer readable storage medium, on which a computer program is stored, characterized in that the program, when being executed by a processor, implements the steps of a method for predicting mental health based on correlation analysis as claimed in any one of claims 1-5.
8. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of a method for predicting mental health based on associative analysis as claimed in any one of claims 1 to 5 when the program is executed.

Description

Psychological health state prediction method and system based on association analysis Technical Field The invention belongs to the technical field of psychological health assessment, and particularly relates to a psychological health state prediction method and system based on association analysis. Background The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art. At present, the psychological problems gradually become social important problems due to frequent occurrence of light events caused by psychological problems of people in the society, and people with psychological problems are found and intervened in time because most people have insufficient cognition on psychological diseases and have low initiative to visit. At present, more scale evaluation modes are used, hysteresis exists in the aspects of data collection, data statistics and the like, deviation exists between the hysteresis and the actual situation, and the evaluation result is limited to the psychological description and the psychological description of the current evaluator, so that the prediction and intervention capability cannot be achieved. That is, the existing method only explains the result obtained by the scale evaluation, but cannot predict the future mental health of the tester. In the aspect of data acquisition, the traditional method is used for data from behavioral data of a certain specific group, personal privacy is violated to a certain extent by data acquisition and use, the risk of privacy data exposure and the like are solved, in the aspect of data analysis, the traditional FP-growth association rule algorithm is widely used, the data records are compressed by constructing a tree structure, the frequent item sets are mined and need to be scanned for two times, the structured FP tree is based on memory, a large memory space is occupied, the operation efficiency is required to be improved, in the data application, most of network psychological assessment systems at present rely on scale assessment to evaluate whether a current individual has a psychological symptom and the severity thereof or not, but the psychological health state of the individual is not considered and is difficult to predict by the traditional scale assessment result. Disclosure of Invention In order to solve the problems, the invention provides a psychological health state prediction method and a psychological health state prediction system based on association analysis, which ensure the accuracy of psychological assessment data and provide users with big data analysis on the premise of ensuring privacy, by changing the data set storage mode and the scanning mode, the strong association rule can be obtained by only scanning the data set once, so that the mining efficiency of the association rule algorithm is improved while the storage space of the database is saved. According to some embodiments, the first aspect of the present invention provides a psychological health status prediction method based on association analysis, which adopts the following technical scheme: a psychological health state prediction method based on association analysis, comprising: Performing privacy calculation based on federal learning to obtain psychological assessment original data; preprocessing the psychological assessment original data to obtain preprocessed psychological assessment data; Scanning based on the preprocessed psychological assessment data to create a two-dimensional storage matrix, and grouping the two-dimensional storage matrix to obtain a data set; Constructing a frequent tree based on the data set to perform association analysis, so as to obtain a strong association rule table meeting the minimum support degree; and selecting feature dimensions with strong relevance to other factors to construct psychological features according to the strong relevance rule table, and predicting the psychological health state by using a trained psychological health state prediction model. According to some embodiments, a second aspect of the present invention provides a psychological health status prediction system based on association analysis, which adopts the following technical scheme: A mental health prediction system based on associative analysis, comprising: the data acquisition module is configured to perform privacy calculation based on federal learning to obtain psychological assessment original data; The data preprocessing module is configured to preprocess the psychological assessment original data to obtain preprocessed psychological assessment data; the data grouping module is configured to scan and create a two-dimensional storage matrix based on the preprocessed psychological assessment data, and group the two-dimensional storage matrix to obtain a data set; The association analysis module is configured to construct a frequent tree based on the data set to perform