CN115938600A - Mental health state prediction method and system based on correlation analysis - Google Patents
Mental health state prediction method and system based on correlation analysis Download PDFInfo
- Publication number
- CN115938600A CN115938600A CN202211582269.6A CN202211582269A CN115938600A CN 115938600 A CN115938600 A CN 115938600A CN 202211582269 A CN202211582269 A CN 202211582269A CN 115938600 A CN115938600 A CN 115938600A
- Authority
- CN
- China
- Prior art keywords
- psychological
- data
- health state
- evaluation
- data set
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention belongs to the field of psychological health assessment, and provides a psychological health state prediction method and a psychological health state prediction system based on correlation analysis, wherein the method comprises the steps of carrying out privacy calculation based on federal learning to obtain psychological assessment original data; preprocessing the original psychological evaluation data to obtain preprocessed psychological evaluation 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 to obtain a strong association rule table meeting the minimum support degree; and selecting the feature dimension with strong correlation with other factors to construct psychological features according to the strong correlation rule table, and predicting the psychological health state by using a trained psychological health state prediction model. By changing the storage mode and the scanning mode of the data set, the strong association rule can be obtained only by scanning the data set once, so that the storage space of the database is saved, and the algorithm mining efficiency of the association rule is improved.
Description
Technical Field
The invention belongs to the technical field of mental health assessment, and particularly relates to a mental health state prediction method and system based on correlation 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 of people in the society are frequently rare because of the psychological problems, and gradually become the key social problems, and because most people have insufficient cognition on the psychological diseases and the willingness of active diagnosis is low, people who have the psychological problems can be found and intervened in time. At present, a plurality of 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, the evaluation result is limited in the psychological description and exposition of the current evaluator, and the prediction and intervention capability cannot be achieved. That is, the conventional method only describes the results obtained by the scale evaluation, but cannot predict the future mental health of the testee.
In the aspect of data acquisition, data in the traditional method is derived from behavior data of a certain specific group, the acquisition and use of the data infringes personal privacy to a certain extent, and the problems of risk of exposing privacy data and the like exist; in the aspect of data analysis, the FP-growth association rule algorithm based on association analysis is widely used, the traditional FP-growth algorithm compresses data records by constructing a tree structure, a frequent item set needs to be scanned for twice data records, the FP tree constructed by traversing the algorithm is based on a memory, a larger memory space is occupied, and the operation efficiency needs to be improved; in data application, most of the current network psychological assessment systems rely on scale assessment for evaluation, and the traditional scale assessment results can assess whether the current individual has certain psychological symptoms and the severity thereof, but do not consider and are difficult to predict the psychological health state of the individual.
Disclosure of Invention
In order to solve the problems, the invention provides a mental health state prediction method and system based on correlation analysis.
According to some embodiments, a first aspect of the present invention provides a mental health state prediction method based on correlation analysis, which adopts the following technical solutions:
a mental health state prediction method based on correlation analysis comprises the following steps:
carrying out privacy calculation based on federal learning to obtain psychological assessment original data;
preprocessing the original psychological assessment 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 to obtain a strong association rule table meeting the minimum support degree;
and selecting the feature dimension with strong correlation with other factors to construct psychological features according to the strong correlation rule table, and predicting the psychological health state by using a trained psychological health state prediction model.
According to some embodiments, the second aspect of the present invention provides a mental health state prediction system based on correlation analysis, which adopts the following technical solutions:
a mental health state 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 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;
the association analysis module is configured to construct a frequent tree based on the data set to perform association analysis, and obtain a strong association rule table meeting the minimum support degree;
and the psychological evaluation module is configured to select the characteristic dimension with stronger association with other factors according to the strong association rule table to construct psychological characteristics, and predict the psychological health state by using the trained psychological health state prediction model.
According to some embodiments, a third aspect of the invention provides a computer-readable storage medium.
A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of a method for predicting a mental health state based on associative analysis as set forth in the first aspect above.
According to some embodiments, a fourth aspect of the invention provides a computer apparatus.
A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of a method for mental health state prediction based on associative analysis as described in the first aspect.
Compared with the prior art, the invention has the beneficial effects that:
according to the method, the original psychological evaluation data of the testers are analyzed to predict the psychological health states of the testers, the psychological evaluation data are subjected to privacy calculation through a horizontal federal learning technology, and the psychological evaluation data are credibly calculated and used on the premise of protecting personal privacy; by utilizing the FP-growth improved algorithm and changing the storage mode and the scanning mode of the data set, the strong association rule can be obtained by scanning the data set once, so that the storage space of the database is saved, and the mining efficiency of the association rule algorithm is improved.
The invention provides a novel mental health state analysis and prediction method. By analyzing the association rule of the psychological evaluation data, a psychological health state prediction model based on XG-Boost is constructed, the probability of abnormality of the psychological health state of an individual to be evaluated is output, and whether the individual has a psychological health problem is accurately judged.
According to the invention, based on a large amount of psychological assessment data generated by online assessment, a psychological health state prediction model is constructed by means of the structural characteristics of association rule mining technology, the prior prediction of the psychological state of a tester is realized, a psychologist is assisted to perform psychological intervention on a psychologically abnormal tester in time, and a new thought is provided for psychological researchers.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the invention and not to limit the invention.
FIG. 1 is a flowchart of a mental health status prediction method based on correlation analysis according to an embodiment of the present invention;
FIG. 2 is a diagram of the psychological assessment system privacy computing logic architecture in an embodiment of the present invention;
FIG. 3 is an improved FP-growth algorithm mining process in an embodiment of the present invention.
Detailed Description
The invention is further described with reference to the following figures and examples.
It is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
The embodiments and features of the embodiments of the present invention may be combined with each other without conflict.
Example one
As shown in fig. 1, the present embodiment provides a mental health state prediction method based on association analysis, and the present embodiment is illustrated by applying the method to a server, it is understood that the method may also be applied to a terminal, and may also be applied to a system including a terminal and a server, and implemented by interaction between the terminal and the server. The server may be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server providing basic cloud computing services such as a cloud service, a cloud database, cloud computing, a cloud function, cloud storage, a network server, cloud communication, middleware service, a domain name service, a security service CDN, a big data and artificial intelligence platform, and the like. The terminal may be, but is not limited to, a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, and the like. The terminal and the server may be directly or indirectly connected through wired or wireless communication, and the application is not limited herein. In this embodiment, the method includes the steps of:
carrying out privacy calculation based on federal learning to obtain psychological assessment original data;
preprocessing the original psychological evaluation data to obtain preprocessed psychological evaluation 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 to obtain a strong association rule table meeting the minimum support degree;
and selecting the feature dimension with strong correlation with other factors to construct psychological features according to the strong correlation rule table, and predicting the psychological health state by using a trained psychological health state prediction model.
Specifically, the privacy calculation is performed based on federal learning to obtain the psychological assessment raw data, and the method comprises the following steps:
obtaining initial psychological health evaluation results of the testers and uploading the results to a server through an evaluation system, and giving a preliminary psychological health evaluation global model of the testers;
the evaluation system locally and independently calculates model parameters by using different tester sample sets, encrypts parameter information and sends the parameter information to a server;
the server uses a weighted average algorithm based on homomorphic encryption to perform safe aggregation on the encrypted parameter information, updates the overall mental health evaluation model of the tester, and returns the aggregated parameter information to the evaluation system in an encryption mode;
the evaluation system decrypts the encrypted and aggregated parameter information, updates a local model by using the decrypted parameter information, and enters the next round of training, so that the loop is iterated until the loss function is converged;
and the server aggregates the parameter information of the last round of local calculation, outputs a model result and generates psychological evaluation original data after system privacy calculation.
Specifically, the preprocessing the original psychological assessment data to obtain the preprocessed psychological assessment data includes:
deleting non-key information in the original psychological evaluation data, reserving a field with strong correlation with the psychological health state, and reducing the data storage capacity;
deleting or filling missing values and deleting redundant data;
and respectively carrying out unified coding on the Chinese field and the factor score to obtain preprocessed psychological evaluation data.
The method comprises the following steps of scanning to create a two-dimensional storage matrix based on preprocessed psychological assessment data, and grouping the two-dimensional storage matrix to obtain a data set, wherein the method specifically comprises the following steps:
acquiring a psychological assessment experiment data set D and a minimum support count n based on the preprocessed psychological assessment data;
starting to scan the psychological assessment experiment data set D for the first time, if the factor score is more than or equal to 2 minutes, recording 1 in the two-dimensional matrix of the psychological assessment data, and otherwise, recording 0;
adding sum columns to the last column of the two-dimensional matrix of the psychological assessment data for counting the number of records 1 in the row;
deleting data which do not meet the minimum support degree in the psychological assessment data two-dimensional matrix to obtain a two-dimensional storage matrix which meets 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.
Performing the group scanning again based on the two-dimensional storage matrix satisfying the minimum support degree n to obtain a data set, including:
scanning on the basis of a two-dimensional storage matrix meeting the minimum support degree n, scanning n1 columns, continuing downward scanning because the value corresponding to m1n1 is 1 until the value corresponding to m rows in the columns is 0, finishing scanning, and establishing a group (s 1, s2, s 3.);
continuing to scan the next column, if the column has a value which is not 0, continuing to scan the position which is not 0 until the column is finished, and continuing to scan the next column; if the average values of the rows are all 0, automatically scanning the next row until the two-dimensional storage matrix meeting the minimum support degree n is scanned for the second time;
and finishing scanning grouping to obtain a grouped data set.
The method for constructing the frequent tree based on the data set to perform association analysis to obtain the strong association rule table meeting the minimum support degree comprises the following steps:
establishing a root node based on the data set, inserting the root node into the FP-Tree, and judging the relationship between the node and the node to be inserted if the node is not empty in subsequent traversal so as to complete the construction of a psychological evaluation factor FP-Tree;
and (3) carrying out frequent pattern mining on the FP-Tree by calling an FP-growth function to obtain a strong association rule between different dimensions meeting the minimum confidence coefficient and the minimum support degree in a psychological assessment experiment data set D, and outputting a strong association rule table containing a front item, a back item, the support degree and the confidence coefficient.
Selecting feature dimensions with strong correlation with other factors according to the strong correlation rule table to construct psychological features, and predicting the psychological health state by using a trained psychological health state prediction model, wherein the method comprises the following steps:
according to the obtained association rule table, performing data dimension reduction by filtering low variance features, screening features which have obvious difference with the mental health state through variance homogeneous test, obtaining feature dimensions with strong association with other factors, and constructing psychological features;
and based on the psychological characteristics, predicting the psychological health state by using a trained psychological health state prediction model based on XG-Boost.
As shown in fig. 1, the method of this embodiment specifically includes:
firstly, obtaining evaluation result original data of a symptom self-evaluation scale SCL90 scale after privacy calculation by means of a transverse federal learning technology; secondly, performing data preprocessing operations such as data selection, data cleaning, data conversion, data integration and the like on the original data to construct an experimental data set; thirdly, mining a frequent item set of the data set by utilizing an FP-growth improved algorithm to obtain strong association rules among all dimensions of the evaluation scale; and finally, selecting and constructing a psychological dimension characteristic vector according to the correlation result, constructing a mental health state prediction model based on the XG-boost, and outputting a mental health state prediction result.
Psychological assessment system privacy calculation logic architecture diagram as shown in fig. 2, based on federal learning data privacy calculation:
in order to solve the problem of mental health and safety, the method adopts the federal learning technology to carry out privacy protection. Taking students as an example, firstly, finishing initial student mental health assessment by a psychological assessment student A, a student B to a student N through an assessment system, uploading the initial student mental health assessment to a server, and setting a preliminary global model; then, the evaluation system respectively uses different student sample sets (SA }, { SB } \8230; { Sn } to locally and independently calculate model parameters (such as the evaluation result score of a certain factor, the evaluation average score of a certain factor, and the like), encrypts the parameter information and then sends the parameter information to a server; at the moment, the system server can use algorithms such as weighted average (gradient average, model average) based on homomorphic encryption to carry out safe aggregation on the sent parameter information, update the global model for student mental health assessment, and return the aggregated parameter information to the system in an encryption mode; the system decrypts the received new parameter information, updates the local model by using the decrypted parameter result, enters the next round of training, and iterates in the way until the loss function is converged; and the server aggregates the parameter information of the last round of local calculation, outputs a model result and generates psychological evaluation original data after system privacy calculation.
Data preprocessing:
the psychological assessment data acquired by the horizontal federal learning technology has abnormal conditions of deficiency, redundancy, format confusion and the like, and the method processes the psychological assessment data through the steps of data selection, data cleaning, data conversion and the like. Firstly, deleting non-key information such as sex, age and the like in original data, reserving fields with strong correlation with mental health states such as somatization, compulsive symptom and the like, and reducing data storage capacity; secondly, deleting or filling the missing values, and deleting redundant data; then, in order to unify the input format of the subsequent model training data, the Chinese field and the factor score are respectively and uniformly coded.
Improved association rule mining:
in order to excavate the strong association relationship among factors in the mental health scale, the FP-Tree constructed by the traditional FP-growth association rule algorithm in each traversal is based on a memory, occupies a larger memory space, and causes low operation efficiency. According to the data mining method, aiming at the data in the array form of the psychological evaluation result, the data storage mode and the data scanning mode in the mining process are changed, so that the scanning times are effectively reduced, the scanning efficiency is improved, and the storage space is reduced. The improved mining process is shown in fig. 3.
(1) Building a data set
Firstly, inputting a psychological assessment experiment data set D and a minimum support count n, scanning for the first time and storing into a two-dimensional matrix A, referring to the scoring standard of each assessment result of a symptom self-assessment scale SCL90 scale, if the score of the factor is more than or equal to 2, judging that the abnormal condition of the psychological state exists, recording 1 in the two-dimensional matrix A of the psychological assessment data, and otherwise, recording 0; adding sum columns to the last column of the psychological evaluation data matrix A for counting the number of the items with problems in the row, namely recording the number of 1, and sequencing the matrix A in a descending order according to the sum columns; determining that the minimum support count is n, and deleting the data which do not meet the minimum support in the psychological assessment data matrix A to obtain a two-dimensional storage matrix A1 meeting the minimum support n; and scanning the newly generated two-dimensional storage matrix A1, and grouping according to the experimental grouping requirement to obtain a grouped data set S.
It should be noted that the psychological assessment experimental data set D is specifically the preprocessed psychological assessment data; the minimum support count is used for screening data which do not meet the conditions, namely data which meet the conditions are reserved, data which do not meet the conditions are deleted, and the minimum support count belongs to one step in the FP-growth algorithm flow.
Grouping requirements: a two-dimensional memory matrix A1 (m × n) is scanned. For example, n1 columns are scanned, because the value corresponding to m1n1 is 1, downward scanning is continued until the value corresponding to m rows in the column is 0, the scanning is finished, a group is established (s 1, s2, s 3.), and the next column scanning is continued; if the column has a value other than 0, continuing to scan the position other than 0 until the column is finished, and continuing to scan the next column; and if the average values of the rows are all 0, automatically scanning the next row until the second scanning of the two-dimensional storage matrix A1 is finished. And finishing scanning grouping to obtain a grouped data set S.
(2) Construction of FP-Tree
And establishing a root node based on the packet data set S obtained in the step, inserting the root node into the FP-Tree, and judging the relationship between the node and the node to be inserted if the node is not empty in subsequent traversal. Judging whether the two nodes are in a parent-child relationship, if so, directly inserting, and continuously traversing the next node; if not, continuing to judge whether a common ancestor exists, if so, inserting the node, otherwise, jumping out of the loop. If the condition is not met, a new branch is established, and the construction of the psychological assessment factor FP-Tree is completed.
(3) Mining of frequent items
And (4) carrying out frequent pattern mining on the formed FP-Tree by calling a FP-growth function. Firstly, upward mining is carried out in sequence from leaf nodes, namely, the item with the minimum support degree in the frequent item set obtained for the first time is constructed, a conditional mode base of the item is constructed, a prefix path set of a suffix mode to be mined in the FP-Tree is found, and a sub-database is formed. Then, mining is carried out on the FP-Tree, a condition FP-Tree is constructed, and frequent patterns are discovered on the FP-Tree in a continuous recursion mode. And merging all subtrees to obtain a frequent item set among the factors, storing the frequent item set into an associated rule base to obtain a strong association rule among different dimensions meeting the minimum confidence coefficient and the minimum support degree in a psychological assessment experiment data set D, and outputting a strong association rule table containing antecedents, consequent items, support degrees and confidence degrees.
Model construction:
(1) Feature selection and construction
According to the obtained association rule table, selecting a feature dimension with strong association with other factors, constructing a psychological feature, wherein the feature with small variance and low or negative correlation with the psychological health state influences the accuracy of model prediction, so that data dimension reduction is carried out by filtering low variance features, and features with significant difference (P < 0.05) with the psychological health state are screened by a variance homogeneous test (F test). Meanwhile, the diagnosis result of mental health state in the data set is used as a label.
(2) When the prediction model is constructed and trained
And constructing a prediction model based on XG-Boost to predict the mental health state. In the model training process, the constructed psychological state characteristics and the psychological state labels are used as the input of the model, the data set division strategy is to randomly select 80% as a training set, and the rest 20% as a testing set. And randomly selecting 20% in the training set as a verification set again, improving the robustness of the model based on 7-fold cross validation, and selecting the optimal parameter combination to construct the model according to different evaluation indexes such as accuracy, recall rate, F1 score and the like in the aspect of model evaluation.
(3) Model output
The floating point number between the [0,1] output by the model is used as the probability of whether the mental health problem exists, and after the same data set is compared, the model is superior to the XG-Boost prediction model based on the traditional FP-growth association rule algorithm.
Example two
The embodiment provides a mental health state prediction system based on correlation analysis, which comprises:
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 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;
the association analysis module is configured to construct a frequent tree based on the data set to perform association analysis, and obtain a strong association rule table meeting the minimum support degree;
and the psychological evaluation module is configured to select the characteristic dimension with strong correlation with other factors to construct psychological characteristics according to the strong correlation rule table, and predict the psychological health state by using the trained psychological health state prediction model.
The modules are the same as the corresponding steps in the implementation examples and application scenarios, but are not limited to the disclosure of the first embodiment. It should be noted that the modules described above as part of a system may be implemented in a computer system such as a set of computer-executable instructions.
In the foregoing embodiments, the description of each embodiment has an emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions in other embodiments.
The proposed system can be implemented in other ways. For example, the above-described system embodiments are merely illustrative, and for example, the division of the above-described modules is merely a logical division, and in actual implementation, there may be other divisions, for example, multiple modules may be combined or integrated into another system, or some features may be omitted, or not executed.
EXAMPLE III
The present embodiment provides a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the steps in a mental health state prediction method based on correlation analysis as described in the first embodiment above.
Example four
The present embodiment provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the computer program to implement the steps of the method for predicting mental health state based on correlation analysis as described in the first embodiment.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.
Although the embodiments of the present invention have been described with reference to the accompanying drawings, it is not intended to limit the scope of the present invention, and it should be understood by those skilled in the art that various modifications and variations can be made without inventive efforts by those skilled in the art based on the technical solution of the present invention.
Claims (10)
1. A mental health state prediction method based on correlation analysis is characterized by comprising the following steps:
carrying out privacy calculation based on federal learning to obtain psychological assessment original data;
preprocessing the original psychological evaluation data to obtain preprocessed psychological evaluation 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 to obtain a strong association rule table meeting the minimum support degree;
and selecting the feature dimension with strong correlation with other factors to construct psychological features according to the strong correlation rule table, and predicting the psychological health state by using a trained psychological health state prediction model.
2. The mental health state prediction method based on associative analysis according to claim 1, wherein the privacy calculation based on federal learning to obtain the psychological evaluation raw data comprises:
obtaining initial psychological health evaluation results of the testers which are uploaded to a server by a plurality of testers through an evaluation system, and giving a primary psychological health evaluation global model of the testers;
the evaluation system locally and independently calculates model parameters by using different tester sample sets, encrypts parameter information and sends the parameter information to a server;
the server uses a weighted average algorithm based on homomorphic encryption to perform safe aggregation on the encrypted parameter information, updates the overall mental health evaluation model of the tester, and returns the aggregated parameter information to the evaluation system in an encryption mode;
the evaluation system decrypts the encrypted and aggregated parameter information, updates the local model by using the decrypted parameter information, and enters the next round of training, and the loop is iterated until the loss function is converged;
and the server aggregates the parameter information of the last round of local calculation, outputs a model result and generates psychological evaluation original data after system privacy calculation.
3. The mental health state prediction method based on correlation analysis according to claim 1, wherein the preprocessing of the raw mental evaluation data to obtain the preprocessed mental evaluation data comprises:
deleting non-key information in the original psychological evaluation data, reserving a field with strong correlation with the psychological health state, and reducing the data storage capacity;
deleting or filling missing values and deleting redundant data;
and respectively carrying out unified coding on the Chinese field and the factor score to obtain preprocessed psychological evaluation data.
4. The mental health state prediction method based on correlation analysis according to claim 1, wherein the pre-processed mental evaluation data is scanned to create a two-dimensional storage matrix, and the two-dimensional storage matrix is grouped to obtain a data set, specifically:
acquiring a psychological assessment experiment data set D and a minimum support count n based on the preprocessed psychological assessment data;
starting to scan the psychological evaluation experiment data set D for the first time, if the score of the factor is more than or equal to 2, recording 1 in the two-dimensional matrix of the psychological evaluation data, and otherwise, recording 0;
adding sum columns to the last column of the two-dimensional matrix of the psychological assessment data for counting the number of records 1 in the row;
deleting data which do not meet the minimum support degree in the psychological evaluation data two-dimensional matrix to obtain a two-dimensional storage matrix which meets 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.
5. The mental health state prediction method based on correlation analysis according to claim 4, wherein the performing of the group scan again based on the two-dimensional storage matrix satisfying the minimum support degree n to obtain the data set comprises:
scanning on the basis of a two-dimensional storage matrix meeting the minimum support degree n, scanning n1 columns, continuing downward scanning because the value corresponding to m1n1 is 1 until the value corresponding to m rows in the columns is 0, finishing scanning, and establishing a group (s 1, s2, s 3.);
continuing to scan the next column, if the column has a value which is not 0, continuing to scan the position which is not 0 until the column is finished, and continuing to scan the next column; if the average values of the rows are all 0, automatically scanning the next row until the two-dimensional storage matrix meeting the minimum support degree n is scanned for the second time;
and finishing scanning grouping to obtain a grouped data set.
6. The mental health status prediction method based on associative analysis according to claim 1, wherein constructing frequent trees based on data sets for associative analysis to obtain a strong association rule table satisfying minimum support comprises:
establishing a root node based on the data set, inserting the root node into the FP-Tree, and judging the relationship between the node and the node to be inserted if the node is not empty in subsequent traversal so as to complete the construction of a psychological evaluation factor FP-Tree;
and (3) carrying out frequent pattern mining on the FP-Tree by calling an FP-growth function to obtain a strong association rule between different dimensions meeting the minimum confidence coefficient and the minimum support degree in a psychological assessment experiment data set D, and outputting a strong association rule table containing a front item, a back item, the support degree and the confidence coefficient.
7. The mental health state prediction method based on correlation analysis according to claim 1, wherein the selecting the dimension with the characteristic with stronger correlation with other factors according to the strong correlation rule table to construct the mental feature and using the trained mental health state prediction model to predict the mental health state comprises:
according to the obtained association rule table, performing data dimension reduction by filtering low variance features, screening features which have obvious difference with the mental health state through variance homogeneous test, obtaining feature dimensions with strong association with other factors, and constructing psychological features;
and based on the psychological characteristics, predicting the psychological health state by using a trained psychological health state prediction model based on XG-Boost.
8. A mental health state 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 association analysis, and obtain a strong association rule table meeting the minimum support degree;
and the psychological evaluation module is configured to select the characteristic dimension with strong correlation with other factors to construct psychological characteristics according to the strong correlation rule table, and predict the psychological health state by using the trained psychological health state prediction model.
9. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of a method for mental health state prediction based on correlation analysis according to any one of claims 1-7.
10. 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 mental health state prediction based on associative analysis according to any one of claims 1 to 7 when executing the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211582269.6A CN115938600A (en) | 2022-12-09 | 2022-12-09 | Mental health state prediction method and system based on correlation analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211582269.6A CN115938600A (en) | 2022-12-09 | 2022-12-09 | Mental health state prediction method and system based on correlation analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115938600A true CN115938600A (en) | 2023-04-07 |
Family
ID=86700429
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211582269.6A Pending CN115938600A (en) | 2022-12-09 | 2022-12-09 | Mental health state prediction method and system based on correlation analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115938600A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116861301A (en) * | 2023-09-04 | 2023-10-10 | 山东爱福地生物股份有限公司 | Management method and system for biomass fuel data produced by straw |
-
2022
- 2022-12-09 CN CN202211582269.6A patent/CN115938600A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116861301A (en) * | 2023-09-04 | 2023-10-10 | 山东爱福地生物股份有限公司 | Management method and system for biomass fuel data produced by straw |
| CN116861301B (en) * | 2023-09-04 | 2023-11-24 | 山东爱福地生物股份有限公司 | Management method and system for biomass fuel data produced by straw |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112365171B (en) | Knowledge graph-based risk prediction method, device, equipment and storage medium | |
| CN109034861A (en) | Customer churn prediction technique and device based on mobile terminal log behavioral data | |
| Hafiz et al. | Bankruptcy prediction of construction businesses: towards a big data analytics approach | |
| CN112818032B (en) | Data screening method and data analysis server for serving big data mining analysis | |
| CN111506723A (en) | Question-answer response method, device, equipment and storage medium | |
| CN111611488B (en) | Information recommendation method and device based on artificial intelligence and electronic equipment | |
| CN111666494B (en) | Clustering decision model generation method, clustering processing method, device, equipment and medium | |
| CN111966904A (en) | Information recommendation method based on multi-user portrait model and related device | |
| CN111310918B (en) | Data processing method, device, computer equipment and storage medium | |
| CN115938600A (en) | Mental health state prediction method and system based on correlation analysis | |
| CN116386899A (en) | Graph learning-based medicine disease association relation prediction method and related equipment | |
| CN116340793A (en) | Data processing method, device, equipment and readable storage medium | |
| CN113345564B (en) | Early prediction method and device for patient hospitalization duration based on graph neural network | |
| CN113268370A (en) | Root cause alarm analysis method, system, equipment and storage medium | |
| CN112200684B (en) | Method, system and storage medium for detecting medical insurance fraud | |
| CN111737319B (en) | User cluster prediction method, device, computer equipment and storage medium | |
| CN114266352A (en) | Model training result optimization method and device, storage medium and equipment | |
| CN113360772A (en) | Interpretable recommendation model training method and device | |
| Pinheiro et al. | A bi‐objective procedure to deliver actionable knowledge in sport services | |
| CN117390455B (en) | Data processing method and device, storage medium and electronic equipment | |
| CN115514580B (en) | Method and device for detecting source-tracing intrusion of self-encoder | |
| CN113742495B (en) | Rating feature weight determining method and device based on prediction model and electronic equipment | |
| EP4372593A1 (en) | Method and system for anonymizsing data | |
| Nguyen et al. | Career path prediction using XGBoost Model and students’ academic results | |
| CN116756387A (en) | Time-awareness-based graph neural network session recommendation method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |