CN113921113A - Positive psychology computerized training method and system - Google Patents
Positive psychology computerized training method and system Download PDFInfo
- Publication number
- CN113921113A CN113921113A CN202110899292.7A CN202110899292A CN113921113A CN 113921113 A CN113921113 A CN 113921113A CN 202110899292 A CN202110899292 A CN 202110899292A CN 113921113 A CN113921113 A CN 113921113A
- Authority
- CN
- China
- Prior art keywords
- training
- user
- current
- data
- value
- 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
- 238000012549 training Methods 0.000 title claims abstract description 284
- 238000000034 method Methods 0.000 title claims abstract description 31
- 230000008451 emotion Effects 0.000 claims abstract description 58
- 230000001149 cognitive effect Effects 0.000 claims abstract description 36
- 238000011156 evaluation Methods 0.000 claims abstract description 28
- 238000004458 analytical method Methods 0.000 claims description 33
- 230000008901 benefit Effects 0.000 claims description 26
- 238000012360 testing method Methods 0.000 claims description 25
- 230000002996 emotional effect Effects 0.000 claims description 13
- 238000000605 extraction Methods 0.000 claims description 13
- 230000008909 emotion recognition Effects 0.000 claims description 9
- 230000036651 mood Effects 0.000 claims description 9
- 230000003203 everyday effect Effects 0.000 claims description 8
- 230000002354 daily effect Effects 0.000 claims description 7
- 230000006996 mental state Effects 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 7
- 208000024891 symptom Diseases 0.000 claims description 7
- 230000003796 beauty Effects 0.000 claims description 6
- 230000006855 networking Effects 0.000 claims description 6
- 230000006870 function Effects 0.000 claims description 5
- 230000000694 effects Effects 0.000 claims description 4
- 238000012417 linear regression Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 claims description 4
- 230000009471 action Effects 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000008859 change Effects 0.000 claims description 3
- 230000019771 cognition Effects 0.000 claims description 3
- 238000010219 correlation analysis Methods 0.000 claims description 3
- 230000003993 interaction Effects 0.000 claims description 3
- 238000000611 regression analysis Methods 0.000 claims description 3
- 230000007704 transition Effects 0.000 claims description 3
- 230000003340 mental effect Effects 0.000 claims description 2
- 238000001514 detection method Methods 0.000 abstract description 8
- 201000010099 disease Diseases 0.000 abstract description 4
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 abstract description 4
- 238000012544 monitoring process Methods 0.000 abstract description 3
- 206010012374 Depressed mood Diseases 0.000 description 3
- 230000006735 deficit Effects 0.000 description 3
- 238000013467 fragmentation Methods 0.000 description 2
- 238000006062 fragmentation reaction Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000009323 psychological health Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 1
- 230000000994 depressogenic effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000004630 mental health Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
- G16H20/70—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to mental therapies, e.g. psychological therapy or autogenous training
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
- G06F18/24155—Bayesian classification
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Child & Adolescent Psychology (AREA)
- Probability & Statistics with Applications (AREA)
- Developmental Disabilities (AREA)
- Hospice & Palliative Care (AREA)
- Psychiatry (AREA)
- Psychology (AREA)
- Social Psychology (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Electrically Operated Instructional Devices (AREA)
Abstract
The application relates to the field of cognitive training, in particular to a positive psychology computerized training method and system. The method comprises the following steps: collecting relevant data of user disease characteristics, cognitive scale evaluation results and real-time emotion monitoring conditions, determining psychological needs of a user, determining current training rating of the user, designating a training scheme by combining the psychological needs of the user, carrying out training effectiveness detection after training is finished, carrying out training at the current level again according to the evaluation conditions if the training does not pass the evaluation, and unlocking the training at the new level by the user if the training at the new level passes the evaluation and reviewing the training at the original level.
Description
Technical Field
The application relates to the field of cognitive training, in particular to a positive psychology computerized training method and system.
Background
The current technologies similar to the scheme are training courses, self-help exercises and love meditation audios under the same theory, and generally have the following defects:
(1) fragmentation, self-help exercise and love meditation audio are generally characterized by fragmentation. People can use these resources to improve mood instantly, but cannot do long-term improvement training through them. The active psychological training course is relatively complete, but it is not reusable, and is based on theoretical explanation, and the exercises contained therein still belong to one of self-help contact.
(2) The paradigms are few and single, and self-help practice and love meditation audios usually mainly adopt one or two paradigms, and after the paradigms are used for a plurality of times, the training can be abandoned due to the boring and single paradigms. Positive psychological training courses are relatively numerous, but are often few in number and not sufficient to sustain long-term training or use.
(3) The active psychological resources are mostly acquired autonomously or are guided and suggested in the psychological consultation process due to the lack of adaptability. The resources obtained in this case often do not correspond to the current situation of the user. For example, with self-help exercise, thank-tales are common and effective training for many people, but writing can be a stressful activity for users with severe depressed mood.
Disclosure of Invention
It is an object of the present application to provide an active psychology computerized training method.
It is a further object of the present application to provide a positive psychology computerized training system.
The positive psychology computerized training method according to the invention comprises the following steps:
s1: determining a user's condition characteristics;
s2: performing cognitive tests and emotion tests, including an emotion type scale, sentence positive emotion analysis, face recognition data, analyzing the current emotion condition of the user, and grading;
s3: the user selects the training requirement of the user as a training target, a personalized training scheme is designated for the user, the training difficulty of the user is gradually upgraded by using training related to the rating, when the user is excellent in the current rating training and reaches the crowd mean value, and after a certain number of days, the effectiveness evaluation of the user is carried out, and the next-stage training can be carried out until the target set by the user is completed.
The positive psychology computerized training method according to the present invention, wherein an initial training level is obtained for the inputted data of the condition, the cognitive test and the emotional test by the following analysis, a basic scheme corresponding to the training is obtained,
(1) generating single-dimensional variable information according to the Euclidean space distance by adopting distance-based clustering analysis;
(2) and (3) processing by adopting a Bayesian regression model, wherein the clustered single-dimensional data is used as an initial state, the prior distribution data is generated according to a normal distribution model as a prior parameter, a Markov Chain Monte Carlo (MCMC) is used for combining with the transition probability to generate a sample of probability posterior distribution, and an output predicted value, namely an initial training level, is generated according to parameter estimation and the solved Bayesian linear regression model.
According to the active psychology computerized training method, after a user finishes training at the current level, self-supervision learning is carried out according to collected observation data and based on multi-modal feature data related to sound channels, image channels and character data, wherein input various different data types are mapped to the same space for calculation based on typical correlation analysis, corresponding specific modes are selected according to current training content for data enhancement, cross-modal data comparison learning is carried out, local area data and global actions are combined, a distribution form of active cognition and emotion states is reconstructed by using an disentangled feature representation model, user data information is generated by restoring learned features, and a proper subsequent training scheme is selected and updated to promote target evaluation and prediction.
The positive psychology computerized training method according to the invention, wherein the training scheme comprises the following values:
(1) the stay time scheme T of each category of training content comprises T1, T2, T3 and T4 value ranges, T1, T2, T3 and T4 respectively correspond to four stages of training, namely thanksgiving-treasure series training, training positive mental state series training, advantage exploration and growth series training, hope searching and resource series training,
(2) reading time, setting training duration Y, increasing the Y value by 1 every day, setting the initial Y value to be 0, sequentially presenting training contents according to a training scheme according to the Y value, wherein,
if Y is within t1, any exercise in the thanksgiving series of exercises is presented, a context is presented, the user is guided to recall and think, the user's selection among exercises is recorded,
if Y is within t2, any exercise in training positive mental series is presented, pictures or situations are presented, the user is required to write his own feelings and score his own experiences, the user's scores and writing feelings are saved,
if Y is within t3, any exercise in the advantage exploration training is presented, the cards are presented, the user is guided to think about the advantages of the user, or the situation is presented, the user is guided to think about the advantages of other people, the card selection of the user and the analysis of the selection result of the card selection by the system are presented and saved,
if Y is in the range of t4, any exercise in the training series is presented, a guide phrase or a guide mini-game is presented, the user is encouraged to think about the future of the user's beauty and the advantages of the user, and the characters selected and recorded by the user are saved.
The positive psychology computerized training method according to the present invention, wherein, every time the user completes the training of the current day, the user writes and speaks a piece of description of the current emotion and the latest state, and performs text extraction analysis thereon, the result is denoted as Rt2, the audio data is analyzed by the speech emotion recognition algorithm, the result is denoted as Rv2, Rt2 and Rv2 are presented in the log record of the current day,
the current training time Y value is increased every day according to the current time acquired through networking, when the Y value reaches the maximum value of the range t to which the Y value belongs currently, a cognitive scale and emotion assessment are called to obtain the score Rc2 and the emotion scale Re2 of the current cognitive scale of the user, and the comparison with the preset system is carried outWhen the conditions that Rc2 is not more than or equal to Ac and Re2 is not more than or equal to Ae are not met, the Y value is reset to the minimum value of the current range, the proportion of Ac to Rc2 and the proportion of Ae to Re2 are compared, and the larger one is taken as a criterion, the proportion is multiplied by the current training class i (i belongs to [1:4 ] in the original training residence time scheme T) in the original training residence time scheme T]) Corresponding dwell range tiAnd recalculating the stay time of the current category, and resetting the training time. Within this range, the training difficulty is unchanged.
According to the active psychology computerized training method of the invention, each time a user performs training of a current level, the level permanently opens the right to the user, the user can choose to return to an initial login scene, reselect symptoms and assessment, perform clustering and regression analysis by the system by using Bayesian estimation, and reset a training scheme.
The positive psychology computerized training system according to the invention comprises:
the evaluation unit is used for collecting personal cognitive condition and emotional condition data of the user and generating and adjusting a training scheme;
the output unit is used for presenting the system content in a picture and character mode;
the observation unit is used for observing the emotion change of the user through voice recognition and text recognition;
the interaction unit interacts with the user and shows an animation effect after the user finishes setting a target;
a storage unit: storing the training scheme, preset training content and daily updated training log of the user, storing the positive emotion adjustment model formula of the user, storing the selection of the user in the training process and new materials generated, generating a personalized information base of the user,
wherein the above-mentioned unit enables positive psychology computerized training by performing the following steps:
s1: acquiring user information, realizing login, and presenting symptom options by an output unit;
s2: personal cognitive and emotional condition data of a user is collected, wherein,
the user's voice describes the current emotion and the recent state, the audio enters the observation unit for processing, the audio is analyzed by the voice emotion recognition algorithm,
the evaluation unit performs the following functions:
invoking cognitive test, evaluating the contents including active resources, integral life quality and life satisfaction, collecting and recording evaluation data,
invoking emotion test, presenting instant emotion test table, collecting and recording evaluation data,
the user describes the current emotion and the recent state through the text, performs text extraction analysis and records the result;
and S3, analyzing and operating according to the input cognitive test data, emotion test data, text extraction analysis data and voice analysis data, outputting a predicted value, namely an initial training grade, setting as a starting point of a training scheme, and setting the selected requirement as an end point of the training scheme, wherein the training scheme comprises the following numerical values:
(1) the stay time scheme T of each category of training content comprises the value ranges of T1, T2, T3 and T4, T1, T2, T3 and T4 respectively correspond to thanksgiving-treasure series training of training, positive mental state series training, advantage exploration and growth series training and hope and resource series training,
(2) reading time, setting training duration Y, increasing the Y value by 1 every day, setting the initial Y value to be 0, sequentially presenting training contents according to a training scheme according to the Y value, wherein,
if Y is within t1, any exercise in the thanksgiving series training is presented, the situation is presented by the output unit, the user is guided to recall and think, the user's selection in the exercise is stored in the storage unit,
if Y is within t2, any exercise in training of the series of active moods is presented, a picture or situation is presented by the output unit, the user is asked to write a feeling and to score the experience, both the user's score and the writing feeling are stored in the storage unit,
if Y is in the range of t3, any exercise in the advantage exploration and growth series training is presented, the card is presented by the output unit, the user is guided to think about the advantage of the card, or the situation is presented, the user is guided to think about the advantage of other people, the card selection of the user and the analysis of the selection result of the card by the system are presented and stored in the storage unit,
if Y is in the range of t4, the user presents the wish to be found and any exercise in the series training of resources, the output unit presents the guide words or guide mini-games, the user is encouraged to think about the future of his own beauty and the advantages of his own existing, and the characters selected and recorded by the user are stored in the storage unit;
wherein,
after the user finishes the training of the current day, the user writes and describes the current emotion and the latest state, the system performs text extraction and analysis, the audio enters the observation unit for processing, the audio is analyzed by a speech emotion recognition algorithm, the system acquires time, a calendar is started, the text extraction and analysis result and the audio analysis result are presented in the log record of the current day,
the current trained time Y value is increased daily according to the current time obtained by networking. Calling a cognitive scale and emotion assessment every time the Y value reaches the maximum value of the range T to which the Y value belongs, obtaining the score Rc2 and the emotion scale Re2 of the current cognitive scale of the user, comparing with the passing values Ac and Ae preset by the system, adding one to the Y value when Rc2 is larger than or equal to Ac and Re2 is larger than or equal to Ae, entering the next training range the next day, resetting the Y value to the minimum value of the current range when the conditions that Rc2 is larger than or equal to Ac and Re2 is larger than or equal to Ae are not met, comparing the ratios of Ac to Rc2, Ae to Re2, taking the larger one as a reference, and multiplying the larger one by the current training category i (i belongs to [1: 4) in the original training stay time scheme T]) Corresponding dwell range tiAnd recalculating the dwell time of the current category, resetting the training time, and keeping the training difficulty unchanged within the range.
After the user finishes training at the current level, the level permanently opens the authority to the user, and the system carries out self-supervision learning based on multi-modal characteristic data such as sound channels, character data and the like according to the collected training footprint data;
s4: and obtaining the number of days S needing training according to the training scheme and the training duration Y, wherein when S =0, the training scheme is ended, the initial page can be returned, and the training scheme can be evaluated and formulated again.
The active psychology computerized training method and system of the invention can push proper training for the user according to the disease characteristics and emotional state of the user, and ensure the usability of the training in a progressive way. The training content relates to a training paradigm of four aspects of depressed mood, positive mood, self-cognition, optimistic expectation. Each aspect comprises a plurality of paradigms, and the paradigms correspond to respective question banks, so that the training content in the scheme is rich and diverse.
Drawings
Fig. 1 is a flow chart of the active psychology computerized training method of the present application.
Detailed Description
The positive psychology computerized training method according to the present application comprises the following steps:
collecting relevant data of user disease characteristics, cognitive scale evaluation results and real-time emotion monitoring conditions;
determining the psychological needs of the user according to the obtained related data;
determining the current training rating of the user according to the difference between the comprehensive score of the user in the four aspects of depressed mood, positive mood, self-cognition and optimistic expectation and the population distribution mean value, and designating a training scheme by combining the psychological needs of the user;
and (3) effectiveness detection, wherein the current training effectiveness detection is carried out after the training is finished, the effectiveness detection comprises cognitive level detection, corresponding emotion scale detection, emotion condition detection of three dimensions of text analysis and face emotion recognition, if the emotion condition detection does not pass the evaluation, the current training is carried out again according to the evaluation condition, and if the user passes the evaluation, the user can unlock the training of the new level and can review the training of the original level.
The technical solutions of the present application are described in detail below with reference to the detailed description and the accompanying drawings.
As shown in fig. 1, the active psychology computerized training method according to the present application comprises the following steps:
s1: when a user logs in for the first time, the mental health state of the user is selected firstly, namely the disease characteristics of the user are divided into three categories: depressed people, psychological sub-health people and psychological health people.
S2: and after selecting the state, performing cognitive test and emotion test, wherein the user participates in the cognitive condition scale evaluation presented by the system, determines the current cognitive level, performs real-time emotion monitoring, completes emotion class scale, positive emotion (text) analysis and positive emotion (voice) analysis so as to analyze the current emotional condition and perform rating. Grade 1 is the most severe impairment of positive emotions, and grade 4 is the least severe impairment of positive emotions.
S3: after rating, the user selects the training requirement of the user as a training target, the system generates a personalized training scheme, and the user can start to use the training associated with the rating of the user. Accordingly, training is divided into four stages. Starting with the level 1 thanksgiving series of training, difficulty and training pressures have increased. The level 4 search hope and resource series training grade is the highest, and the method is generally suitable for users with slight positive emotion impairment or with certain positive emotion ability restored after training. And when the user is excellent in performance in the current rating training, the average value of the crowd is reached, the effectiveness evaluation of the current rating is carried out after a certain number of days, and the next-stage training can be unlocked after the effectiveness evaluation of the current rating is passed. And unlocking one by one until the target set by the user is finished.
According to the active psychology computerized training method, for input data of symptoms, cognitive tests and emotional tests, firstly, distance-based clustering analysis is adopted, single-dimensional variable information is generated according to Euclidean spatial distance, and then a Bayesian regression model is adopted: and generating prior distribution data by taking the clustered single-dimensional data as an initial state and taking a normal distribution model as a prior parameter, and generating a sample of probability posterior distribution by using Markov Chain Monte Carlo (MCMC) in combination with transition probability. And generating an output predicted value, namely an initial training grade, according to the parameter estimation and the solved Bayes linear regression model, and acquiring a basic scheme corresponding to the training. Wherein,
the Bayesian linear regression model is as follows:
outputting a training scheme according to the posterior distribution probability:
according to the active psychology computerized training method, after the user finishes training at the current level, the system can perform self-supervision learning according to collected observation data and multi-modal characteristic data such as sound channels, image channels, character data and the like. Specifically, multiple input different data types are mapped to the same space for calculation based on typical correlation analysis (CCA), corresponding specific modalities are selected according to current training contents for data enhancement, cross-modality data are compared and learned, local area data and global actions are combined, a distribution form of positive cognition and emotional states is reconstructed by using a Disentanglement Feature Representation Model (DFRM), user data information is generated through learned feature reduction, and a proper subsequent training scheme is selected and updated to promote target evaluation and prediction.
The positive psychology computerized training method according to the invention, wherein the training scheme comprises the following values:
(1) the stay time scheme T of each category of training content comprises T1, T2, T3 and T4 value ranges, T1, T2, T3 and T4 respectively correspond to four stages G, C, P and F of training, namely a thanksgiving-treasure series training, training a positive mental state series training, a superiority exploration and growth series training, searching hope and resource series training,
(2) reading time, setting training duration Y, increasing the Y value by 1 every day, setting the initial Y value to be 0, sequentially presenting training contents according to a training scheme according to the Y value, wherein,
if Y is within the range of t1, any exercise in the thanksgiving series of exercises is presented, the situation is presented by the output unit, the user is guided to recall and think, the selection of the user in the exercise is recorded and is stored in the storage unit as the exercise footprint of the user,
if Y is within t2, any exercise in training of training positive mental state series is presented, picture or situation is presented by the output unit, user is required to write own feeling and score own experience, user's scoring and writing feeling are saved and stored in the storage unit as user's training footprint,
if Y is within t3, any exercise in the advantage exploration training is presented, the card is presented by the output unit, the user is guided to think about the own advantage, or the situation is presented, the user is guided to think about the advantages of others, the card selection of the user and the analysis of the selection result by the system are presented and stored as the training footprint of the user in the storage unit,
if Y is in the range of t4, any exercise in the series of growth exercises is presented, the output unit presents a guide word or guides a mini-game, so that the user is encouraged to think about the future of the user's beauty and the advantages of the user, and the characters selected and recorded by the user are stored as the training footprint of the user and are stored in the storage unit;
wherein,
every time the user finishes the training on the current day, the user writes and speaks a section to describe the current emotion and the latest state, the system starts a text recognition function, performs text extraction and analysis on the text, records the result as Rt2, the audio enters an observation unit for processing, is analyzed into a result Rv2 by a speech emotion recognition algorithm, acquires the time by the system, starts a calendar, presents Rt2 and Rv2 in the log record of the current day,
the current trained time Y value is increased daily according to the current time obtained by networking. Calling a cognitive scale and emotion assessment every time the Y value reaches the maximum value of the range T to which the Y value belongs, obtaining the score Rc2 and the emotion scale Re2 of the current cognitive scale of the user, comparing with the passing values Ac and Ae preset by the system, adding one to the Y value when Rc2 is larger than or equal to Ac and Re2 is larger than or equal to Ae, entering the next training range the next day, resetting the Y value to the minimum value of the current range when the conditions that Rc2 is larger than or equal to Ac and Re2 is larger than or equal to Ae are not met, comparing the ratios of Ac to Rc2, Ae to Re2, taking the larger one as a reference, and multiplying the larger one by the current training category i (i belongs to [1: 4) in the original training stay time scheme T]) Corresponding toDwell range tiAnd recalculating the stay time of the current category, and resetting the training time. Within this range, the training difficulty is unchanged.
After the user finishes the training of the current level, the level permanently opens the authority to the user, the system carries out self-supervision learning according to the collected training footprint data and multi-modal characteristic data such as sound channels, character data and the like,
the user can also choose to return to the initial login scene, reselect the symptoms and the evaluation, and the system adopts Bayesian estimation to perform clustering and regression analysis and reset a training scheme.
The positive psychology computerized training system according to the invention comprises:
the evaluation unit is used for collecting personal cognitive condition and emotional condition data of the user and generating and adjusting a training scheme;
the output unit is used for presenting the system content in a picture and character mode;
the observation unit is used for observing the emotion change of the user through voice recognition and text recognition;
the interaction unit interacts with a user by touching and clicking a screen, and shows an animation effect after the user finishes setting a target;
a storage unit: storing the training scheme, preset training content and daily updated training log of the user, storing the positive emotion adjustment model formula of the user, storing the selection of the user in the training process and new materials generated, generating a personalized information base of the user,
wherein the above-mentioned unit enables positive psychology computerized training by performing the following steps:
s1: acquiring user information, realizing login, presenting symptom options by an output unit, and dividing symptoms into three categories including depression, psychological sub-health and psychological health;
s2: personal cognitive and emotional condition data of a user is collected, wherein,
the user speaks a section of speech to describe the current emotion and the latest state, a recording key is presented, after the user holds down, recording is started, the audio enters the observation unit for processing, the audio is analyzed into a result Rv1 by a speech emotion recognition algorithm,
the evaluation unit collects the following data of the user:
calling a cognitive test Tc1, evaluating the content including active resource class, overall life quality class, life satisfaction degree and the like, collecting and recording evaluation data Rc1,
calling an emotion test Te1, presenting an instant emotion PANAS table or TA-SA table, collecting and recording evaluation data Re1,
writing a section of description by the user on the current emotion and the latest state, starting a text recognition function by the system, extracting and analyzing the text, and recording the result as Rt 1;
and S3, analyzing and operating according to the input Rc1, Re1, Rt1 and Rv1, outputting a predicted value, namely an initial training grade, and setting the predicted value as a starting point of a training scheme, wherein the selected requirement is a target and is a terminal point of the training scheme, the training scheme is output according to the posterior distribution probability, and comprises the following numerical values:
(1) the stay time scheme T of each category of training content comprises T1, T2, T3 and T4 value ranges, T1, T2, T3 and T4 respectively correspond to four stages G, C, P and F of training (Thyan-treasure series training, positive mental state series training, advantage exploration and growth series training, hope-seeking and resource series training),
(2) reading time, setting training duration Y, increasing the Y value by 1 every day, setting the initial Y value to be 0, sequentially presenting training contents according to a training scheme according to the Y value, wherein,
if Y is within t1, any exercise in G-class training is presented, the situation is presented by the output unit, the user is guided to recall and think, the selection of the user in the exercise is recorded and is stored in the storage unit as the training footprint of the user,
if Y is within t2, any exercise in the class C training is presented, a picture or situation is presented by the output unit, the user is asked to write his own experience and to score his own experience, both the user's score and the written experience are saved and saved in the storage unit as the user's training footprint,
if Y is within t3, any exercise in P-type training is presented, the card is presented by the output unit, the user is guided to think about own superiority, or the situation is presented, the user is guided to think about other superiority, the card selection by the user and the analysis of the selection result by the system are presented and stored as the training footprint of the user in the storage unit,
if Y is in the range of t4, any exercise in the F-type training is presented, the output unit presents a guide word or guides a mini-game, so that the user is encouraged to think about the future of the user's beauty and the advantages of the user, and the characters selected and recorded by the user are stored as the training footprint of the user and are stored in the storage unit;
wherein,
every time the user finishes training on the current day, the user writes and speaks a section to describe the current emotion and the latest state, the system starts a text recognition function, text extraction and analysis are carried out on the text, the result is recorded as Rt2, audio enters an observation unit to be processed, the audio is analyzed into a result Rv2 through a speech emotion recognition algorithm, the system acquires time, a calendar is started, and Rt2 and Rv2 are presented in a log record of the current day.
The current trained time Y value is increased daily according to the current time obtained by networking. Calling a cognitive scale and emotion assessment every time the Y value reaches the maximum value of the range T to which the Y value belongs, obtaining the score Rc2 and the emotion scale Re2 of the current cognitive scale of the user, comparing with the passing values Ac and Ae preset by the system, adding one to the Y value when Rc2 is larger than or equal to Ac and Re2 is larger than or equal to Ae, entering the next training range the next day, resetting the Y value to the minimum value of the current range when the conditions that Rc2 is larger than or equal to Ac and Re2 is larger than or equal to Ae are not met, comparing the ratios of Ac to Rc2, Ae to Re2, taking the larger one as a reference, and multiplying the larger one by the current training category i (i belongs to [1: 4) in the original training stay time scheme T]) Corresponding dwell range tiAnd recalculating the dwell time of the current category, resetting the training time, and keeping the training difficulty unchanged within the range.
After the user finishes training at the current level, the level permanently opens the authority to the user, and the system carries out self-supervision learning based on multi-modal characteristic data such as sound channels, character data and the like according to the collected training footprint data;
s4: and obtaining the number of training days S needed again according to the training scheme and the training duration Y, wherein when S =0, the training scheme is ended, and the initial page can be returned to, and the training scheme can be re-evaluated and formulated.
The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (8)
1. A positive psychology computerized training method, characterized in that it comprises the following steps:
s1: determining a user's condition characteristics;
s2: performing cognitive tests and emotion tests, including an emotion type scale, sentence positive emotion analysis, face recognition data, analyzing the current emotion condition of the user, and grading;
s3: the user selects the training requirement of the user as a training target, a personalized training scheme is formulated for the user, the training difficulty of the user is gradually upgraded by using training related to the rating, when the user is excellent in the current rating training and reaches the crowd mean value, and after a certain number of days, the effectiveness evaluation of the user is carried out, and the next-stage training can be carried out until the target set by the user is completed.
2. The proactive psychology computerized training method according to claim 1, wherein the data for the inputted condition, cognitive test and emotional test is analyzed to obtain an initial training level,
(1) generating single-dimensional variable information according to the Euclidean space distance by adopting distance-based clustering analysis;
(2) and (3) processing by adopting a Bayesian regression model, wherein the clustered single-dimensional data is used as an initial state, the prior distribution data is generated by using a normal distribution model as a prior parameter, a Markov chain Monte Carlo is used for combining with the transition probability to generate a sample of probability posterior distribution, and an output prediction value, namely an initial training level, is generated according to parameter estimation and the solved Bayesian linear regression model.
3. The active psychology computerized training method as claimed in claim 1, wherein in step S3, after the user completes the current level of training, the user performs self-supervised learning based on multi-modal feature data related to voice channel, image channel, and text data according to the collected observation data, wherein the input multiple different data types are mapped to the same space for calculation based on typical correlation analysis, corresponding specific modalities are selected according to the current training content for data enhancement, cross-modal data are compared and learned, local area data and global actions are combined, the distribution form of active cognition and emotional states is reconstructed by using the disentangled feature representation model, user data information is generated by restoring the learned features, and a suitable subsequent training scheme is selected and updated.
4. The positive psychology computerized training method of claim 1, wherein the training regimen includes the following values:
(1) the stay time scheme T of each category of training content comprises the value ranges of T1, T2, T3 and T4, T1, T2, T3 and T4 which respectively correspond to thanksgiving treasure series training, positive mental state series training, advantage exploration and growth series training, hope searching and resource series training,
(2) reading time, setting training duration Y, increasing the Y value by 1 every day, setting the initial Y value to be 0, sequentially presenting training contents according to a training scheme according to the Y value, wherein,
if Y is within t1, provide any exercises in the thanksgiving series of exercises, present context, guide the user to recall and think, record the user's selections in the exercises,
if Y is within t2, providing any exercise in training positive mental series, presenting pictures or situations, asking the user to write and score his experience, saving the user's score and writing experience,
if Y is within t3, any exercise in the benefit exploration training is provided, the cards are presented, the user is guided to think about the advantages of the user, or the situation is presented, the user is guided to think about the advantages of other people, the card selection of the user and the analysis of the selection result by the system are presented and saved,
if Y is in the range of t4, any exercise in the training series is provided, a guide phrase is presented or a small game is guided, the user is encouraged to think about the future of the user's beauty and the advantages of the user, and the characters selected and recorded by the user are saved.
5. The proactive psychology computerized training method of claim 4, wherein a value of a current trained time Y is increased daily according to a current trained time obtained through networking, a cognitive scale and a mood test are called each time the value of Y reaches a maximum value of a T range to which the current training belongs, a user current cognitive scale score Rc2 and a mood scale Re2 are obtained, and passing values Ac and Ae preset by a comparison system are obtained, wherein when Rc2 is larger than or equal to Ac and Re2 is larger than or equal to Ae, the value of Y is increased by one the next day, when a condition that Rc2 is larger than or equal to Ac and Re2 is not larger than or equal to Ae is not satisfied, the value of Y is reset to a minimum value of the current range, ratios of Ac to Rc2, Ae to Re2 are compared, and a stay range T corresponding to a current class i in an original training stay time scheme T is multiplied by a larger ratioiWherein i ∈ [1:4 ]]And recalculating the dwell time of the current category, resetting the training time, and keeping the training difficulty unchanged within the range.
6. The active psychology computerized training method of claim 1, wherein each time the user completes the current day of training, the user composes and speaks a dialog describing the current mood and recent state, from which a text extraction analysis is performed, the audio data is analyzed by a speech mood recognition algorithm, and the text extraction analysis and the audio analysis data are presented in a log record of the current day.
7. The active psychology computerized training method of claim 1, wherein each time a user completes a current level of training, the level permanently opens the user with permission, the user may choose to return to an initial login scenario, reselect a medical condition and evaluate, perform a clustering and regression analysis by the system using bayesian estimation, and re-configure a training scenario.
8. A positive psychology computerized training system, the system comprising:
the evaluation unit is used for collecting personal cognitive condition and emotional condition data of the user and generating and adjusting a training scheme;
the output unit is used for presenting the system content in a picture and character mode;
the observation unit is used for observing the emotion change of the user through voice recognition and text recognition;
the interaction unit interacts with the user and shows an animation effect after the user finishes setting a target;
a storage unit: storing the training scheme, preset training content and daily updated training log of the user, storing the positive emotion adjustment model formula of the user, storing the selection of the user in the training process and new materials generated, generating a personalized information base of the user,
wherein the above-mentioned unit enables positive psychology computerized training by performing the following steps:
s1: acquiring user information, realizing login, and presenting symptom options by an output unit;
s2: personal cognitive and emotional condition data of a user is collected, wherein,
the user voice describes the current emotion and the recent state, the audio enters the observation unit to be processed, the audio is analyzed by a voice emotion recognition algorithm, and the following functions are executed through the evaluation unit:
invoking cognitive test, evaluating the contents including active resources, integral life quality and life satisfaction, collecting and recording evaluation data,
invoking emotion test, presenting instant emotion test table, collecting and recording evaluation data,
the user describes the current emotion and the recent state through the text, performs text extraction analysis and records the result;
and S3, analyzing and operating according to the input cognitive test data, emotion test data, text extraction analysis data and voice analysis data, outputting a predicted value, namely an initial training grade, setting the predicted value as a starting point of a training scheme, and setting the selected requirement as an end point of the training scheme, wherein the training scheme comprises the following numerical values:
(1) the stay time scheme T of each category of training content comprises the value ranges of T1, T2, T3 and T4, T1, T2, T3 and T4 respectively correspond to thanksgiving-treasure series training of training, positive mental state series training, advantage exploration and growth series training and hope and resource series training,
(2) reading time, setting training duration Y, increasing the Y value by 1 every day, setting the initial Y value to be 0, sequentially presenting training contents according to a training scheme according to the Y value, wherein,
if Y is within t1, any exercise in the thanksgiving series training is presented, the situation is presented by the output unit, the user is guided to recall and think, the user's selection in the exercise is stored in the storage unit,
if Y is within t2, any exercise in training of the series of active moods is presented, a picture or situation is presented by the output unit, the user is asked to write a feeling and to score the experience, both the user's score and the writing feeling are stored in the storage unit,
if Y is in the range of t3, any exercise in the advantage exploration and growth series training is presented, the card is presented by the output unit, the user is guided to think about the advantage of the card, or the situation is presented, the user is guided to think about the advantage of other people, the card selection of the user and the analysis of the selection result of the card by the system are presented and stored in the storage unit,
if Y is in the range of t4, the user presents the wish to be found and any exercise in the series training of resources, the output unit presents the guide words or guide mini-games, the user is encouraged to think about the future of his own beauty and the advantages of his own existing, and the characters selected and recorded by the user are stored in the storage unit;
wherein,
after the user finishes the training of the current day, the user writes and describes the current emotion and the latest state, the system performs text extraction and analysis, the audio enters the observation unit for processing, the audio is analyzed by a speech emotion recognition algorithm, the system acquires time, a calendar is started, the text extraction and analysis result and the audio analysis result are presented in the log record of the current day,
the current training time Y value is increased every day according to the current training time acquired through networking, when the Y value reaches the maximum value of the range T to which the current training time belongs, a cognitive scale and emotion assessment are called to obtain the score Rc2 and the emotion scale Re2 of the current cognitive scale of the user, passing values Ac and Ae preset by a comparison system are compared, when the Rc2 is larger than or equal to Ac and the Re2 is larger than or equal to Ae, the Y value is increased by one, the training in the next range can be carried out next day, when the conditions that the Rc2 is larger than or equal to Ac and the Re2 is larger than or equal to Ae are not met, the Y value is reset to the minimum value of the current range, the proportions of Ac and Rc2, the Ae and the Re2 are compared, and the larger proportion is multiplied by the staying range T corresponding to the current training category i in the original training staying time scheme TiRecalculating the current class dwell time, where i e [1:4 ]]The training time is reset, within the range, the training difficulty is not changed,
after the user finishes training at the current level, the level permanently opens the authority to the user, and the system carries out self-supervision learning based on multi-modal characteristic data such as sound channels, character data and the like according to the collected training footprint data;
s4: and obtaining the number of days S needing training according to the training scheme and the training duration Y, wherein when S =0, the training scheme is ended, the initial page is returned, and the training scheme is evaluated and formulated again.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110899292.7A CN113921113A (en) | 2021-08-06 | 2021-08-06 | Positive psychology computerized training method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110899292.7A CN113921113A (en) | 2021-08-06 | 2021-08-06 | Positive psychology computerized training method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113921113A true CN113921113A (en) | 2022-01-11 |
Family
ID=79233085
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110899292.7A Pending CN113921113A (en) | 2021-08-06 | 2021-08-06 | Positive psychology computerized training method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113921113A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113990452A (en) * | 2021-12-27 | 2022-01-28 | 广东数业智能科技有限公司 | Evaluation method and system based on psychological literacy and readable storage medium |
| CN115641943A (en) * | 2022-12-22 | 2023-01-24 | 北京智精灵科技有限公司 | TMS technology-based paranoid mental disorder training method and system |
| CN116327132A (en) * | 2023-05-26 | 2023-06-27 | 首都医科大学宣武医院 | Cognitive assessment method and cognitive intervention method based on contextual memory |
| CN116785553A (en) * | 2023-08-25 | 2023-09-22 | 北京智精灵科技有限公司 | Cognitive rehabilitation system and method based on interface type emotion interaction |
-
2021
- 2021-08-06 CN CN202110899292.7A patent/CN113921113A/en active Pending
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113990452A (en) * | 2021-12-27 | 2022-01-28 | 广东数业智能科技有限公司 | Evaluation method and system based on psychological literacy and readable storage medium |
| CN113990452B (en) * | 2021-12-27 | 2022-05-17 | 广东数业智能科技有限公司 | Evaluation method and system based on psychological literacy and readable storage medium |
| CN115641943A (en) * | 2022-12-22 | 2023-01-24 | 北京智精灵科技有限公司 | TMS technology-based paranoid mental disorder training method and system |
| CN115641943B (en) * | 2022-12-22 | 2023-05-12 | 北京智精灵科技有限公司 | TMS technology-based delusional mental disorder training system |
| CN116327132A (en) * | 2023-05-26 | 2023-06-27 | 首都医科大学宣武医院 | Cognitive assessment method and cognitive intervention method based on contextual memory |
| CN116785553A (en) * | 2023-08-25 | 2023-09-22 | 北京智精灵科技有限公司 | Cognitive rehabilitation system and method based on interface type emotion interaction |
| CN116785553B (en) * | 2023-08-25 | 2023-12-26 | 北京智精灵科技有限公司 | Cognitive rehabilitation system and method based on interface type emotion interaction |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yazdavar et al. | Semi-supervised approach to monitoring clinical depressive symptoms in social media | |
| CN113921113A (en) | Positive psychology computerized training method and system | |
| Stern et al. | Using grounded theory methodology to study women's culturally based decisions about health | |
| Bargsted et al. | Social cleavages and political dealignment in contemporary Chile, 1995–2009 | |
| Sanchez-Cortes et al. | Identifying emergent leadership in small groups using nonverbal communicative cues | |
| Lowry | Network TV news framing of good vs. bad economic news under democrat and republican presidents: A lexical analysis of political bias | |
| Neuendorf et al. | Quantitative content analysis and the measurement of collective identity | |
| Speer | Transcending thenatural'/contrived'distinction: a rejoinder to ten Have, Lynch and Potter | |
| Iliev et al. | Political rhetoric through the lens of non‐parametric statistics: are our legislators that different? | |
| Erhard et al. | Diverse effects of mass media on concerns about immigration: new evidence from Germany, 2001–2016 | |
| Elvir et al. | Remembering a conversation–a conversational memory architecture for embodied conversational agents | |
| Kim et al. | Who voted in 2016? Using fuzzy forests to understand voter turnout | |
| Zhao et al. | Political irony as self-censorship practice? Examining dissidents’ use of Weibo in the 2017 Hong Kong Chief Executive Election | |
| CN113573128B (en) | Audio processing method, device, terminal and storage medium | |
| Zárate | Dimensions of pandering perceptions among Hispanic Americans and their effect on political trust | |
| Pollock et al. | Coda/s/and intervocalic/d/elision in Andalusia: The formation of Susana Diaz's regional identity in political discourse | |
| Taylor et al. | Encouraging attention and exploration in a hybrid recommender system for libraries of unfamiliar music | |
| Pentland et al. | Behavioral indicators of dominance in an adversarial group negotiation game | |
| Kailash | The more things change, the more they stay the same in India: The Bahujan and the paradox of the “Democratic Upsurge” | |
| Mostafa et al. | Sentiment analysis of spanish words of arabic origin related to islam: a social network analysis | |
| Barron-Lutzross | Speech stereotypes of female sexuality | |
| Candia et al. | The rising of collective forgetting and cultural selectivity in inventors and physicists communities | |
| Çalışkan et al. | Quantitative analysis of spoken discourse using memoirs of old-time moviegoers | |
| Lemay et al. | Framing gay men’s sexual relationships: A rapid review of the literature from 2011–2018 | |
| Beyleveld | Constructing a leadership model: derived from a South African business leader's life-story |
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 |