CN117711572A - Psychological adjustment information determining method and device, storage medium and computer equipment - Google Patents

Psychological adjustment information determining method and device, storage medium and computer equipment Download PDF

Info

Publication number
CN117711572A
CN117711572A CN202311720130.8A CN202311720130A CN117711572A CN 117711572 A CN117711572 A CN 117711572A CN 202311720130 A CN202311720130 A CN 202311720130A CN 117711572 A CN117711572 A CN 117711572A
Authority
CN
China
Prior art keywords
psychological
adjustment information
historical
consultation
data
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
Application number
CN202311720130.8A
Other languages
Chinese (zh)
Inventor
唐蕊
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ping An Chuangke Technology Beijing Co ltd
Original Assignee
Ping An Chuangke Technology Beijing Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Ping An Chuangke Technology Beijing Co ltd filed Critical Ping An Chuangke Technology Beijing Co ltd
Priority to CN202311720130.8A priority Critical patent/CN117711572A/en
Publication of CN117711572A publication Critical patent/CN117711572A/en
Pending legal-status Critical Current

Links

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

  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

The invention discloses a method, a device, a storage medium and computer equipment for determining psychological adjustment information, which relate to the technical field of information and the technical field of digital medical treatment and mainly can improve the determining efficiency and the determining accuracy of the psychological adjustment information. The method comprises the following steps: responding to the psychological consultation signal of the psychological consultation object, and judging whether the psychological consultation object is a re-diagnosis object or not; if the psychological consultation object is a re-diagnosis object, acquiring re-diagnosis characteristic data, historical psychological state data and historical mental state data in each historical consultation process, historical psychological adjustment information, and current psychological state data and current mental state data in the current psychological consultation process; and inputting the re-diagnosis characteristic data, the historical psychological state data, the historical mental state data, the historical psychological adjustment information, the current psychological state data and the current mental state data into a preset psychological adjustment information prediction model to obtain psychological adjustment information of a re-diagnosis object aiming at the current psychological consultation process.

Description

Psychological adjustment information determining method and device, storage medium and computer equipment
Technical Field
The present invention relates to the field of information technologies, and in particular, to a method and apparatus for determining psychological adjustment information, a storage medium, and a computer device.
Background
With the rapid development of social and economic levels in China, the economic level of people is improved, people pay more attention to the health condition of people, psychological health is paid more attention, and more people begin to seek psychological consultation services to cope with the psychological health problem of people.
Currently, psychological adjustment information is often provided to patients manually by a psychiatrist. However, this manner of manually determining the psychological adjustment information may result in a low efficiency of determining the psychological adjustment information, and at the same time, may result in a situation in which the psychological adjustment task is determined to be wrong due to the influence of subjective experiences of a psychological doctor or due to uneven technical levels of the doctor, thereby resulting in a low accuracy of determining the psychological adjustment information.
Disclosure of Invention
The invention provides a method, a device, a storage medium and computer equipment for determining psychological adjustment information, which mainly aims to improve the determining efficiency and the determining accuracy of the psychological adjustment information.
According to a first aspect of the present invention, there is provided a method of determining psychological adjustment information, comprising:
responding to a psychological consultation signal of a psychological consultation object, and judging whether the psychological consultation object is a re-diagnosis object or not;
if the psychological consultation object is a re-diagnosis object, acquiring re-diagnosis characteristic data of the re-diagnosis object, historical psychological state data and historical mental state data of the re-diagnosis object in each historical consultation process, historical psychological adjustment information, and current psychological state data and current mental state data of the re-diagnosis object in the current mental consultation process;
and inputting the re-diagnosis characteristic data, the historical psychological state data, the historical mental state data, the historical psychological adjustment information, the current psychological state data and the current mental state data into a preset psychological adjustment information prediction model to perform adjustment information prediction, so as to obtain psychological adjustment information of the re-diagnosis object aiming at the current heart consultation process.
Optionally, the inputting the review feature data, the historical mental state adjustment information, the current mental state data and the current mental state data into a preset mental adjustment information prediction model for adjustment information prediction to obtain mental adjustment information of the review object for the current mental consultation process includes:
Performing cross processing on the re-diagnosis feature data, the historical psychological state data, the historical mental state data, the historical psychological adjustment information, the current psychological state data and the current mental state data to obtain psychological cross feature vectors;
and inputting the psychological cross characteristic vector into a preset psychological adjustment information prediction model to predict adjustment information, so as to obtain the psychological adjustment information of the review object aiming at the current psychological consultation process.
Optionally, the cross processing is performed on the review feature data, the historical mental state data, the historical mental regulation information, the current mental state data and the current mental state data to obtain a mental cross feature vector, which includes:
determining a feature vector corresponding to the re-diagnosis feature data, a historical psychological state vector corresponding to the historical psychological state data, a mental state vector corresponding to the historical mental state data, an adjustment vector corresponding to the historical psychological adjustment information and a current psychological state vector corresponding to the current psychological state data;
performing feature level cross processing on the feature vector, the historical psychological state vector, the mental state vector, the adjustment vector and the current psychological state vector to obtain a feature level cross vector;
Performing element level cross processing on the feature vector, the historical psychological state vector, the mental state vector, the adjustment vector and the current psychological state vector to obtain an element level cross vector;
performing low-order cross processing on the feature vector, the historical psychological state vector, the mental state vector, the adjustment vector and the current psychological state vector to obtain a low-order cross vector;
and carrying out transformation processing on the feature level cross vector, the element level cross vector and the low-order cross vector by using a preset transformation function to obtain a psychological cross feature vector.
Optionally, the preset psychological adjustment information prediction model includes a name layer, a type layer, a sequence layer, a plurality of full connection layers and a sigmoid layer; inputting the psychological cross feature vector into a preset psychological adjustment information prediction model for adjustment information prediction to obtain psychological adjustment information of the review object aiming at the current psychological consultation process, wherein the method comprises the following steps:
inputting the psychological cross feature vector into the name layer for feature extraction to obtain a name feature vector;
inputting the psychological cross feature vector into the type layer for feature extraction to obtain a type feature vector;
Inputting the psychological cross feature vector into the sequence layer for feature extraction to obtain a sequence feature vector;
adding the name feature vector, the type feature vector and the order feature vector to obtain a fusion feature vector;
sequentially inputting the fusion feature vectors to the plurality of full-connection layers for feature extraction, and extracting a psychological adjustment feature vector output by the last full-connection layer;
and inputting the psychological adjustment feature vector to the sigmoid layer to obtain psychological adjustment information of the re-diagnosis object aiming at the current heart consultation process.
Optionally, before the review feature data, the historical mental state adjustment information, the current mental state data and the current mental state data are input into a preset mental adjustment information prediction model to perform adjustment information prediction, so as to obtain mental adjustment information of the review object aiming at the current mental consultation process, the method further includes:
constructing a preset initial psychological adjustment information prediction model, and acquiring training characteristic data of a sample training object which is cured after completing a target psychological consultation course, psychological state training data, mental state training data and corresponding actual psychological adjustment information of the sample training object in each sample consultation process, wherein the target psychological consultation course comprises a plurality of sample consultation processes;
Inputting training characteristic data of the sample training object, psychological state training data, mental state training data of the sample training object in the nth sample consultation process, historical psychological state training data and historical mental state training data of the sample training object in the previous n-1 sample consultation processes and corresponding actual mental regulation information of the historical psychological state training data into the preset initial psychological regulation information prediction model for regulation information prediction, and obtaining predicted psychological regulation information of the sample training object in the nth sample consultation process, wherein n is more than or equal to 2;
constructing a loss function corresponding to the preset initial psychological adjustment information prediction model based on the actual psychological adjustment information and the predicted psychological adjustment information in the nth sample consultation process;
and constructing a preset psychological adjustment information prediction model based on the loss function.
Optionally, after the constructing the predictive model of preset psychological adjustment information based on the loss function, the method further includes:
acquiring verification feature data of a sample verification object, historical psychological state verification data, historical mental state verification data and historical psychological adjustment verification information of the sample verification object in the previous m-1 sample historical consultation processes, and current psychological state verification data of the sample verification object in the mth psychological consultation process, wherein if the psychological consultation process of the sample verification object comprises y consultation processes, m is more than 1 and less than y-1, and m is an integer;
Inputting the verification feature data, the historical psychological state verification data, the historical psychological adjustment verification information and the current psychological state verification data into the preset psychological adjustment information prediction model for mediation information prediction to obtain psychological adjustment verification information of the sample verification object in the m-th psychological consultation process;
obtaining psychological state verification data of the sample verification object after psychological adjustment is carried out on the psychological adjustment verification information;
scoring the psychological state mediated by the sample verification object based on the psychological state verification data to obtain a first verification scoring value;
predicting the psychological adjustment information of the sample verification object in each psychological consultation process by using the preset psychological adjustment information prediction model to obtain the last psychological adjustment information of the sample verification object in the last psychological consultation process in the psychological consultation course;
acquiring last psychological state data of the sample verification object after psychological adjustment is carried out on the last psychological adjustment information;
scoring the psychological state of the sample verification object after the psychological consultation treatment course is finished based on the last psychological state data to obtain a second verification scoring value;
Determining weight coefficients of the first verification score value and the second verification score value respectively;
based on the weight coefficient, adding the first verification score value and the second verification score value to obtain a prediction effect score value corresponding to the preset psychological adjustment information prediction model;
if the prediction effect evaluation value is larger than a preset threshold value, predicting the psychological adjustment information of the psychological consultation object by using the preset psychological adjustment information prediction model;
and if the prediction effect evaluation value is smaller than or equal to a preset threshold value, optimizing and adjusting model parameters of the preset psychological adjustment information prediction model by using a preset near-end strategy optimization algorithm to obtain an adjusted preset psychological adjustment information prediction model, and predicting psychological adjustment information of the psychological consultation object by using the adjusted preset psychological adjustment information prediction model.
Optionally, after the determining whether the psychological consultation object is a review object, the method further includes:
if the psychological consultation object is a primary consultation object, primary consultation characteristic data of the primary consultation object, primary consultation psychological state data and primary consultation mental state data of the primary consultation object in the current heart consultation process are obtained;
Inputting the initial diagnosis characteristic data, the initial diagnosis mental state data and the initial diagnosis mental state data into a preset mental regulation information prediction model to predict regulation information, and obtaining mental regulation information of the initial diagnosis object aiming at the current mental consultation process.
According to a second aspect of the present invention, there is provided a determining apparatus of psychological adjustment information, comprising:
the judging unit is used for responding to the psychological consultation signal of the psychological consultation object and judging whether the psychological consultation object is a review object or not;
the system comprises an acquisition unit, a review unit and a review unit, wherein the acquisition unit is used for acquiring review characteristic data of a review object, historical psychological state data and historical mental state data of the review object in each historical consultation process, historical psychological adjustment information and current psychological state data and current mental state data of the review object in the current mental consultation process if the psychological consultation object is the review object;
the prediction unit is used for inputting the review feature data, the historical psychological state data, the historical mental state data, the historical psychological adjustment information, the current psychological state data and the current mental state data into a preset psychological adjustment information prediction model to perform adjustment information prediction, so that psychological adjustment information of the review object in the current heart consultation process is obtained.
According to a third aspect of the present invention, there is provided a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the above method of determining psychological adjustment information.
According to a fourth aspect of the present invention there is provided 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 method of determining psychological adjustment information above when executing the program.
According to the method, the device, the storage medium and the computer equipment for determining the psychological adjustment information, compared with the prior method for providing the psychological adjustment information for the patient by the psychological doctor, the method and the device for determining the psychological adjustment information determine whether the psychological consultation object is a re-diagnosis object or not by responding to the psychological consultation signal of the psychological consultation object; then if the psychological consultation object is a re-diagnosis object, acquiring re-diagnosis characteristic data of the re-diagnosis object, historical psychological state data and historical mental state data of the re-diagnosis object in each historical consultation process, historical psychological adjustment information, and current psychological state data and current mental state data of the re-diagnosis object in the current mental consultation process; and finally, inputting the re-diagnosis characteristic data, the historical psychological state data, the historical mental state data, the historical psychological adjustment information, the current psychological state data and the current mental state data into a preset psychological adjustment information prediction model to perform adjustment information prediction, and obtaining psychological adjustment information of the re-diagnosis object aiming at the current heart consultation process. Therefore, under the condition that the psychological consultation object is a re-diagnosis object, the psychological adjustment information of the re-diagnosis object is determined by comprehensively analyzing the re-diagnosis characteristic data, the historical psychological state data, the historical mental state data, the historical psychological adjustment information, the current psychological state data and the current mental state data of the re-diagnosis object by utilizing the preset psychological adjustment information prediction model, so that the problems of low determination efficiency and low determination accuracy caused by manually determining the psychological adjustment information can be avoided, and the determination efficiency and the determination accuracy of the psychological adjustment information can be improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiments of the invention and together with the description serve to explain the invention and do not constitute a limitation on the invention. In the drawings:
FIG. 1 is a flowchart of a method for determining psychological adjustment information according to an embodiment of the present invention;
FIG. 2 is a flowchart of another method for determining psychometric information according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a device for determining psychological adjustment information according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of another apparatus for determining mental regulation information according to an embodiment of the present invention;
fig. 5 shows a schematic physical structure of a computer device according to an embodiment of the present invention.
Detailed Description
The invention will be described in detail hereinafter with reference to the drawings in conjunction with embodiments. It should be noted that, in the case of no conflict, the embodiments and features in the embodiments may be combined with each other.
At present, the psychological doctor manually provides psychological adjustment information for the patient, which results in lower determining efficiency of the psychological adjustment information, and meanwhile, the psychological adjustment information is determined to be lower in accuracy due to the influence of subjective experiences of the psychological doctor or the fact that the skill level of the doctor is uneven.
In order to solve the above-mentioned problems, an embodiment of the present invention provides a method for determining psychological adjustment information, as shown in fig. 1, the method includes:
101. and responding to the psychological consultation signal of the psychological consultation object, and judging whether the psychological consultation object is a re-diagnosis object or not.
Wherein, the re-diagnosis object refers to the second inquiry or more inquiry of the psychological consultation object in the same psychological consultation course.
For the embodiment of the invention, when a psychological consultation signal of a psychological consultation object is received, firstly, whether the psychological consultation object is a re-diagnosis object is required to be judged, if the psychological consultation object is the re-diagnosis object, historical psychological state data, historical psychological mediation information of the psychological consultation object in each historical consultation process are required to be obtained, current psychological state data and current psychological state data of the psychological consultation object in the current consultation process are obtained, then the data are input into a preset psychological regulation information prediction model to predict regulation information, and psychological regulation information of the re-diagnosis object in the current psychological consultation process is obtained, so that the determination efficiency and the determination accuracy of the psychological regulation information can be improved. The embodiment of the invention is mainly suitable for determining the scene of the psychological adjustment information. The execution subject of the embodiment of the invention is a device or equipment capable of determining psychological adjustment information, and can be specifically arranged at one side of a server.
102. If the psychological consultation object is a re-diagnosis object, the re-diagnosis characteristic data of the re-diagnosis object, the historical psychological state data and the historical mental state data of the re-diagnosis object in each historical consultation process, the historical psychological adjustment information and the current psychological state data and the current mental state data of the re-diagnosis object in the current mental consultation process are obtained.
The review feature data refers to age, occupation, interest, income, hobbies and other data of a review object, the data are necessary data of the review object in a psychological consultation process, non-user privacy data, each historical consultation process refers to a previous consultation process of the review object in the psychological consultation process, the historical psychological state data refer to depression emotion, anxiety emotion, social relation, attention, memory, thinking, logical reasoning, perception, understanding, thinking mode and the like of the review object in each historical consultation process, the historical mental state data refer to actions, speech, facial expression, non-language behaviors and the like of the review object in each historical consultation process, the historical psychological regulation information refers to psychological regulation tasks given by pointers to each historical consultation process, and the psychological regulation tasks comprise: writing tasks, dialogue tasks, meditation tasks and the like, the current psychological state data refer to depression emotion, anxiety emotion, social relation, attention, memory, thinking, logical reasoning, perception, understanding and thinking modes of things and the like of a review object in the current psychological consultation process, and the current mental state data refer to actions, speech, facial expression, non-language behaviors and the like of the review object in the current consultation process.
For the embodiment of the invention, in order to accurately predict the psychological adjustment information of the psychological consultation object in the consultation process, firstly, the omnibearing data including the re-diagnosis characteristic data of the re-diagnosis object, the historical psychological state data and the historical mental state data of the re-diagnosis object in each historical consultation process, the historical psychological adjustment information and the current psychological state data and the current mental state data of the re-diagnosis object in the current mental consultation process are required to be obtained, and then the psychological adjustment information of the re-diagnosis object is determined by comprehensively analyzing the data, so that the determination accuracy of the psychological adjustment information can be improved.
103. And inputting the re-diagnosis characteristic data, the historical psychological state data, the historical mental state data, the historical psychological adjustment information, the current psychological state data and the current mental state data into a preset psychological adjustment information prediction model to perform adjustment information prediction, so as to obtain psychological adjustment information of a re-diagnosis object aiming at the current psychological consultation process.
For the embodiment of the invention, after the review feature data, the historical psychological state data, the historical mental state data, the historical psychological adjustment information, the current psychological state data and the current mental state data of the review object are obtained, the data are input into a preset psychological adjustment information prediction model for adjustment information prediction, so that the psychological adjustment information of the review object aiming at the current psychological consultation process is obtained, the model is utilized for predicting the psychological adjustment information of the psychological consultation object, the determination efficiency and the determination accuracy of the information adjustment information can be improved, and a doctor can carry out psychological adjustment on the psychological consultation object according to the psychological adjustment information, so that the psychological adjustment effect can be improved.
According to the method for determining psychological adjustment information provided by the invention, compared with the current mode that psychological adjustment information is manually provided for a patient by a psychological doctor, whether the psychological consultation object is a re-diagnosis object is judged by responding to the psychological consultation signal of the psychological consultation object; then if the psychological consultation object is a re-diagnosis object, acquiring re-diagnosis characteristic data of the re-diagnosis object, historical psychological state data and historical mental state data of the re-diagnosis object in each historical consultation process, historical psychological adjustment information, and current psychological state data and current mental state data of the re-diagnosis object in the current mental consultation process; and finally, inputting the re-diagnosis characteristic data, the historical psychological state data, the historical mental state data, the historical psychological adjustment information, the current psychological state data and the current mental state data into a preset psychological adjustment information prediction model to perform adjustment information prediction, and obtaining psychological adjustment information of the re-diagnosis object aiming at the current heart consultation process. Therefore, under the condition that the psychological consultation object is a re-diagnosis object, the psychological adjustment information of the re-diagnosis object is determined by comprehensively analyzing the re-diagnosis characteristic data, the historical psychological state data, the historical mental state data, the historical psychological adjustment information, the current psychological state data and the current mental state data of the re-diagnosis object by utilizing the preset psychological adjustment information prediction model, so that the problems of low determination efficiency and low determination accuracy caused by manually determining the psychological adjustment information can be avoided, and the determination efficiency and the determination accuracy of the psychological adjustment information can be improved.
Further, in order to better illustrate the above process of determining the psychological adjustment information, as a refinement and extension of the above embodiment, another method for determining psychological adjustment information is provided in an embodiment of the present invention, as shown in fig. 2, where the method includes:
201. and responding to the psychological consultation signal of the psychological consultation object, and judging whether the psychological consultation object is a re-diagnosis object or not.
For the embodiment of the present invention, after receiving a psychological consultation signal of a psychological consultation object, it is first required to determine whether the psychological consultation object is a review object, and if the psychological consultation object is not a review object, that is, if the psychological consultation object is a primary consultation object, another adjustment information determining method is required to determine psychological adjustment information of the primary consultation object, based on this, the method includes: if the psychological consultation object is a primary consultation object, primary consultation characteristic data of the primary consultation object, primary consultation psychological state data and primary consultation mental state data of the primary consultation object in the current heart consultation process are obtained; inputting the initial diagnosis characteristic data, the initial diagnosis mental state data and the initial diagnosis mental state data into a preset mental regulation information prediction model to predict regulation information, and obtaining mental regulation information of the initial diagnosis object aiming at the current mental consultation process.
Wherein, the initial consultation object refers to an object for psychological consultation for the first time. The initial diagnosis state data refer to depression emotion, anxiety emotion, social relationship, attention, memory, thinking, logic reasoning, perception, understanding, thinking mode and the like of the initial diagnosis object in the current psychological consultation process, and the initial diagnosis state data refer to actions, speech, facial expression, non-language behaviors and the like of the initial diagnosis object in the current consultation process.
Specifically, if the psychological consultation object is a primary consultation object, the psychological consultation operator needs to communicate with the psychological consultation object, in the communication process, the system collects primary diagnosis feature data such as age, occupation, hobbies and the like of the psychological consultation object, meanwhile, in the communication process with the primary consultation object, the system can also collect primary diagnosis psychological state data and primary diagnosis mental state data of the primary consultation object, and it is required to explain that the primary diagnosis feature data, the primary diagnosis psychological state data and the primary diagnosis mental state data of the primary consultation object can also be obtained in a answering mode. Further, after the initial diagnosis feature data, the initial diagnosis mental state data and the initial diagnosis mental state data of the initial diagnosis object are obtained, the initial diagnosis feature data, the initial diagnosis mental state data and the initial diagnosis mental state data are input into a preset mental regulation information prediction model to conduct regulation information prediction, and mental regulation information of the initial diagnosis object aiming at the current mental consultation process is obtained, so that the mental regulation information of the mental consultation object is predicted through the model, and the prediction efficiency and the prediction accuracy of the mental regulation information can be improved.
202. If the psychological consultation object is a re-diagnosis object, the re-diagnosis characteristic data of the re-diagnosis object, the historical psychological state data and the historical mental state data of the re-diagnosis object in each historical consultation process, the historical psychological adjustment information and the current psychological state data and the current mental state data of the re-diagnosis object in the current mental consultation process are obtained.
For the embodiment of the invention, if the psychological consultation object carries out psychological consultation more than once, the psychological consultation object is a review object, the current psychological state data and the current mental state data of the review object in the current psychological consultation process are obtained in a mode of answering questions or the psychological consultation person carries out talking communication with the review object, meanwhile, the historical psychological state data, the historical mental state data and the historical mental state regulation information of the review object in each historical consultation process are recorded in a system database, and then the review feature data, the historical psychological state data, the historical mental state regulation information, the current psychological state data and the current mental state data of the review object are input into a preset psychological regulation information prediction model to predict the psychological regulation information.
203. And performing cross processing on the re-diagnosis characteristic data, the historical psychological state data, the historical mental state data, the historical psychological adjustment information, the current psychological state data and the current mental state data to obtain psychological cross characteristic vectors.
For the embodiment of the present invention, since in order to extract more implicit features in the review feature data, the historical mental state data, the historical mental adjustment information, the current mental state data, and the current mental state data, the cross processing needs to be performed on the above data, based on this, step 203 specifically includes: determining a feature vector corresponding to the re-diagnosis feature data, a historical psychological state vector corresponding to the historical psychological state data, a mental state vector corresponding to the historical mental state data, an adjustment vector corresponding to the historical psychological adjustment information and a current psychological state vector corresponding to the current psychological state data; performing feature level cross processing on the feature vector, the historical psychological state vector, the mental state vector, the adjustment vector and the current psychological state vector to obtain a feature level cross vector; performing element level cross processing on the feature vector, the historical psychological state vector, the mental state vector, the adjustment vector and the current psychological state vector to obtain an element level cross vector; performing low-order cross processing on the feature vector, the historical psychological state vector, the mental state vector, the adjustment vector and the current psychological state vector to obtain a low-order cross vector; and carrying out transformation processing on the feature level cross vector, the element level cross vector and the low-order cross vector by using a preset transformation function to obtain a psychological cross feature vector.
In particular, to obtain more invisible features, it is necessary to cross-process different vectors, where the specific cross-processing scheme is, for example, if the feature vector is (a) 1 ,a 2 ,a 3 ) The historical psychological state vector is (b) 1 ,b 2 ,b 3 ) The mental vector is (c) 1 ,c 2 ,c 3 ) The adjustment vector is (d 1 ,d 2 ,d 3 ) The current mental state vector is (e 1 ,e 2 ,e 3 ) The specific cross processing comprises the following steps: characteristic level cross processing is carried out on the characteristic vector, the historical psychological state vector, the mental state vector, the regulating vector and the current psychological state vector, namely, after Hadamard products are carried out on all elements among vectors, convolution transformation is carried out under a certain weight, and the characteristic level cross vector f (w (a) 1 *b 1 *c 1 *d 1 *e 1 ,a 2 *b 2 *c 2 *d 2 *e 2 ,a 3 *b 3 *c 3 *d 3 *e 3 ) A) is provided; meanwhile, performing element level cross processing on the feature vector, the historical psychological state vector, the mental state vector, the regulating vector and the current psychological state vector, namely after each element among vectors is Hadamard product, assigning different weight values to the result after each product, and performing linear transformation to obtain an element level cross vector f (w) 1 *a 1 *b 1 *c 1 *d 1 *e 1 ,w 2 *a 2 *b 2 *c 2 *d 2 *e 2 ,w 3 *a 3 *b 3 *c 3 *d 3 *e 3 ) The method comprises the steps of carrying out a first treatment on the surface of the At the same time, the feature vector, the historical psychological state vector, the mental state vector, the adjustment vector and all elements in the current psychological state vector are combined two by two, and low-order cross processing is carried out to obtain a low-order cross vector f (w (a) 1 ,a 2 ,a 3 ,b 1 ,b 2 ,b 3, c 1 ,c 2 ,c 3 ,d 1 ,d 2 ,d 3 ,e 1 ,e 2 ,e 3 ) And combining the three crossed vectors, and performing transformation processing by using a preset transformation function to obtain the psychological cross feature vector. The preset transform function here may be set according to actual conditions, which is not limited in this embodiment. It should be noted that the above examples are illustrative only and are not limiting on the embodiments of the present invention. According to the embodiment of the invention, more implicit features are extracted by fully utilizing the relation between the data, and meanwhile, the feature level, the element and the low-order processing are considered, so that the data is more fully utilized, the prediction result obtained later is more accurate, and the requirements of actual application scenes are met.
204. And inputting the psychological cross characteristic vector into a preset psychological adjustment information prediction model to predict adjustment information, so as to obtain psychological adjustment information of the re-diagnosis object aiming at the current psychological consultation process.
For the embodiment of the present invention, in order to improve the prediction accuracy of the preset psychological adjustment information prediction model, the preset psychological adjustment information prediction model needs to be trained and constructed first, and based on this, the method includes: constructing a preset initial psychological adjustment information prediction model, and acquiring training characteristic data of a sample training object which is cured after completing a target psychological consultation course, psychological state training data, mental state training data and corresponding actual psychological adjustment information of the sample training object in each sample consultation process, wherein the target psychological consultation course comprises a plurality of sample consultation processes; inputting training characteristic data of the sample training object, psychological state training data, mental state training data of the sample training object in the nth sample consultation process, historical psychological state training data and historical mental state training data of the sample training object in the previous n-1 sample consultation processes and corresponding actual mental regulation information of the historical psychological state training data into the preset initial psychological regulation information prediction model for regulation information prediction, and obtaining predicted psychological regulation information of the sample training object in the nth sample consultation process, wherein n is more than or equal to 2; constructing a loss function corresponding to the preset initial psychological adjustment information prediction model based on the actual psychological adjustment information and the predicted psychological adjustment information in the nth sample consultation process; and constructing a preset psychological adjustment information prediction model based on the loss function.
The training feature data refers to age, occupation, interest, income, hobbies and other data of a sample training object, the data are necessary data and non-user privacy data of the sample training object in the psychological consultation process, the psychological state training data refer to depression emotion, anxiety emotion, social relation, attention, memory, thinking, logical reasoning, perception, understanding and thinking modes of things and the like of the sample training object in each sample consultation process, the mental state training data refer to actions, speech, facial expression, non-language behaviors and the like of the sample training object in each sample consultation process, the actual psychological adjustment information refers to psychological adjustment tasks given by pointers in each sample consultation process, and the psychological adjustment tasks comprise: writing class tasks, reading class tasks, social class tasks, and the like.
Specifically, at least one preset initial psychological adjustment information prediction model is firstly constructed, the structure of the constructed preset initial psychological adjustment information prediction model can be the same or different, meanwhile, training characteristic data of a sample training object which has completed the whole psychological consultation course and is cured and psychological state training data, mental state training data and corresponding actual psychological adjustment information of the sample training object in each sample consultation process are obtained, for example, if 5 psychological consultations are carried out on the whole target consultation course of the sample training object, historical psychological state training data, historical mental state training data and corresponding actual mental adjustment information of the sample training object in the previous 3 psychological consultation processes and psychological state training data and mental state training data of the sample training object in the 4 th psychological consultation process are input into the preset initial psychological adjustment information prediction model, the sample training object can output predicted psychological adjustment information aiming at the 4 th psychological adjustment process through the preset initial psychological adjustment information prediction model, for example, if the prediction information and the actual psychological adjustment information of the sample training object are carried out for the 4 th psychological consultation process together, the psychological adjustment function can be carried out according to the prediction model, the prediction model can be adjusted according to the predicted by the training object, the initial psychological adjustment function can be reduced, the final psychological adjustment loss can be obtained, the final psychological adjustment model can be obtained according to the preset initial training parameter prediction loss is obtained, and the final training loss can be adjusted according to the preset psychological adjustment model, and the final training parameter prediction loss can be adjusted, and the final training method can be obtained. It should be noted that if a plurality of preset initial psychological adjustment information prediction models are constructed, the obtained data may be divided into training data and test data, the training data is divided into a plurality of sets of training data according to the number of the preset initial psychological adjustment information prediction models, the corresponding preset initial psychological adjustment information prediction models are trained by using the plurality of sets of training data respectively, the trained preset initial psychological adjustment information prediction models are obtained, then the plurality of preset initial psychological adjustment information prediction models are tested by using the test data respectively, and the preset initial psychological adjustment information prediction model with the highest prediction accuracy is selected as the preset psychological adjustment information prediction model according to the test result.
Further, after the preset psychological adjustment information prediction model is built, in order to further improve the preset precision of the preset psychological adjustment information prediction model, model parameters of the preset psychological adjustment information prediction model need to be optimized, and based on this, the method includes: acquiring verification feature data of a sample verification object, historical psychological state verification data, historical mental state verification data and historical psychological adjustment verification information of the sample verification object in the previous m-1 sample historical consultation processes, and current psychological state verification data of the sample verification object in the mth psychological consultation process, wherein if the psychological consultation process of the sample verification object comprises y consultation processes, m is more than 1 and less than y-1, and m is an integer; inputting the verification feature data, the historical psychological state verification data, the historical psychological adjustment verification information and the current psychological state verification data into the preset psychological adjustment information prediction model for mediation information prediction to obtain psychological adjustment verification information of the sample verification object in the m-th psychological consultation process; obtaining psychological state verification data of the sample verification object after psychological adjustment is carried out on the psychological adjustment verification information; scoring the psychological state mediated by the sample verification object based on the psychological state verification data to obtain a first verification scoring value; predicting the psychological adjustment information of the sample verification object in each psychological consultation process by using the preset psychological adjustment information prediction model to obtain the last psychological adjustment information of the sample verification object in the last psychological consultation process in the psychological consultation course; acquiring last psychological state data of the sample verification object after psychological adjustment is carried out on the last psychological adjustment information; scoring the psychological state of the sample verification object after the psychological consultation treatment course is finished based on the last psychological state data to obtain a second verification scoring value; determining weight coefficients of the first verification score value and the second verification score value respectively; based on the weight coefficient, adding the first verification score value and the second verification score value to obtain a prediction effect score value corresponding to the preset psychological adjustment information prediction model; if the prediction effect evaluation value is larger than a preset threshold value, predicting the psychological adjustment information of the psychological consultation object by using the preset psychological adjustment information prediction model; and if the prediction effect evaluation value is smaller than or equal to a preset threshold value, optimizing and adjusting model parameters of the preset psychological adjustment information prediction model by using a preset near-end strategy optimization algorithm to obtain an adjusted preset psychological adjustment information prediction model, and predicting psychological adjustment information of the psychological consultation object by using the adjusted preset psychological adjustment information prediction model.
The preset threshold is set according to actual requirements. The sample verification object is a psychological consultation object participating in psychological consultation courses, the verification feature data refer to age, occupation, interests, income, hobbies and the like of the sample verification object, the data are necessary data of the sample verification object in the psychological consultation process, the historical psychological state verification data refer to depression emotion, anxiety emotion, social relation, attention, memory, thinking, logical reasoning, perception of things, understanding, thinking modes and the like of the sample verification object in each sample historical consultation process, the historical psychological state verification data refer to actions, speech, facial expressions, non-language behaviors and the like of the sample verification object in each sample historical consultation process, the historical psychological regulation information refers to psychological regulation tasks given by pointers to each sample historical consultation process, and the psychological regulation tasks comprise: writing tasks, reading tasks, social tasks and the like, and the current psychological state verification data refer to depression emotion, anxiety emotion, social relationship, attention, memory, thinking, logical reasoning, perception, understanding, thinking mode and the like of a sample verification object in the m-th psychological consultation process after m-1 psychological adjustment.
Specifically, if the course of this psychological consultation of the sample verification object includes 6 psychological consultations in total, and the sample verification object has completed 2 psychological consultations, when the sample verification object is performing 3 psychological consultations, it is necessary to obtain historical psychological state verification data, historical mental state verification data, historical psychological adjustment verification information in the previous 2 psychological consultations, and also obtain current psychological state verification data in the 3 rd psychological consultations, then the verification data is simultaneously input into a predictive model of preset psychological adjustment information to predict the psychological adjustment information, so as to obtain psychological adjustment verification information of the sample verification object in the 3 rd psychological consultations, and then the sample verification object is executed according to the psychological adjustment verification information recommended in the 3 rd psychological consultations, for example, if the psychological adjustment verification information recommended in the 3 rd psychological consultations is that a certain impurity is read in a certain period of time, the sample verification object is read according to the above requirement, then the 4 th psychological consultations are performed, in the 4 th psychological consultations, state data of the sample verification object is obtained, and according to the psychological state data of the sample verification object is predicted, so that the first psychological state of the sample verification object is evaluated to obtain a value representing a better psychological state score value of 0 (the more than 0 score can be evaluated). Further, after the sample verification object performs the 4 th psychological consultation, the sample verification object may recommend psychological adjustment information to the 4 th psychological consultation process by using a preset psychological adjustment information prediction model, then participate in the 5 th psychological consultation, after the sample verification object performs the 5 th psychological consultation, the sample verification object may recommend psychological adjustment information to the 5 th psychological consultation process by using a preset psychological adjustment information prediction model, then the sample verification object performs the psychological adjustment information, then participates in the 6 th psychological consultation, after the sample verification object performs the 6 th psychological consultation, namely, last psychological adjustment information, the sample verification object performs the last psychological adjustment information, after the execution is finished, psychological state data after the whole consultation course is finished, namely, last psychological state data is obtained, then according to the last psychological state data, the psychological state of the sample verification object finished is evaluated, a second verification score value is obtained (can be a numerical value between 0 and 5, the greater representative state is better), after the sample verification object performs the 6 th psychological consultation, if the first verification weight and the second verification score are respectively determined to be a predictive value of 0.7, and if the first verification weight is a predictive value of 2.7, and the second verification score is a predictive value of 2.7.5, and if the predictive value is respectively a value of the first verification weight is determined to be a threshold value of 2.0, and the predictive value is corresponding to a value of the predictive value of the threshold value is determined to be 2.0.7, and the threshold value is satisfied, and if the predictive value is corresponding to the value of the first verification value is required to be 2.0.2.5 is a predictive value for the corresponding value of the 2.2 for the corresponding value of the first verification score and the first verification value is corresponding to the 2 is a prediction 2, the prediction method comprises the steps that a preset psychological adjustment information prediction model can be directly used for predicting psychological adjustment information of a psychological consultation object, if a prediction effect evaluation value is smaller than or equal to a preset threshold value, it is determined that the prediction accuracy of the preset psychological adjustment information prediction model does not meet the requirement, model parameters of the preset psychological adjustment information prediction model can be optimized through a PPO (Proximal Policy Optimization near-end strategy optimization) algorithm at the moment, the preset psychological adjustment information prediction model with the preset accuracy meeting the requirement is obtained, and finally the psychological adjustment information of the psychological consultation object is predicted through the preset psychological adjustment information prediction model with the prediction accuracy meeting the requirement. Therefore, the prediction accuracy of the psychological adjustment information can be improved by periodically and continuously performing parameter adjustment on the preset psychological adjustment information prediction model.
Further, the model structure of the constructed preset psychological adjustment information prediction model comprises: in order to predict psychological adjustment information of a psychological consultation object by using a preset psychological adjustment information prediction model, the method comprises the following steps: inputting the psychological cross feature vector into the name layer for feature extraction to obtain a name feature vector; inputting the psychological cross feature vector into the type layer for feature extraction to obtain a type feature vector; inputting the psychological cross feature vector into the sequence layer for feature extraction to obtain a sequence feature vector; adding the name feature vector, the type feature vector and the order feature vector to obtain a fusion feature vector; sequentially inputting the fusion feature vectors to the plurality of full-connection layers for feature extraction, and extracting a psychological adjustment feature vector output by the last full-connection layer; and inputting the psychological adjustment feature vector to the sigmoid layer to obtain psychological adjustment information of the re-diagnosis object aiming at the current heart consultation process.
Specifically, the psychological cross feature vector can be respectively input to a name layer, a type layer and a sequence layer, the name feature vector is output through the name layer, the type feature vector is output through the type layer, the sequence feature vector is output through the sequence layer, then the name feature vector, the type feature vector and the sequence feature vector are added to obtain a fusion feature vector, the fusion feature vector is input to a plurality of fully-connected layers to perform feature extraction, the psychological adjustment feature vector output by the last fully-connected layer is extracted, and finally the psychological adjustment feature vector is input to the sigmoid layer to obtain psychological adjustment information of a review object aiming at the current heart consultation process, wherein the psychological adjustment information can comprise a plurality of psychological adjustment tasks.
It should be noted that, in the embodiment of the present invention, the review feature data, the historical mental state adjustment information, the current mental state data, and the current mental state data may also be directly input into a preset mental state adjustment information prediction model to perform adjustment information prediction, based on which, the method includes: based on the historical psychological state data, evaluating the historical psychological state of the review object to obtain a historical psychological state grading value; based on the historical attitude data, evaluating the historical attitude of the re-diagnosis object to obtain a historical attitude grading value; based on the current psychological state data, evaluating the current psychological state of the review object to obtain a current psychological state grading value; based on the current state data, evaluating the current state of the re-diagnosis object to obtain a current state grading value; the re-diagnosis feature data, the historical psychological state score value, the historical psychological adjustment information, the current psychological state score value and the current psychological state score value are respectively input into a name layer, a type layer and an order layer to obtain three feature vectors; adding the three feature vectors to obtain a comprehensive feature vector; sequentially inputting the comprehensive feature vectors to the plurality of full-connection layers to perform feature extraction, and extracting the full-connection feature vector output by the last full-connection layer; and inputting the full-connection feature vector to the sigmoid layer to obtain psychological adjustment information of the re-diagnosis object aiming at the current heart consultation process.
Specifically, the historical psychological state data, the historical mental state data, the current psychological state data and the current mental state data are respectively evaluated, 0-5 is used for evaluation, the historical psychological state score value, the historical mental state score value, the current psychological state score value and the current mental state score value are obtained, the re-diagnosis characteristic data, the historical psychological state score value, the historical mental regulation information, the current psychological state score value and the current mental state score value are input into a name layer, each psychological state, each mental state and each mental state in the name layer are subjected to vector representation corresponding to one token, the psychological state is provided with 6 tokens corresponding to 0-5 points, each score corresponds to one token vector representation, the mental state is provided with 6 tokens corresponding to 0-5 points, each score corresponds to one token vector representation, each psychological adjustment information corresponds to a token vector representation, whereby a feature vector can be output through the name layer, while the review feature data, the historical psychological state score, the historical mental state score, the historical psychological adjustment information, the current psychological state score, the current mental state score are input to the type layer, the mental state corresponds to a type in the type layer, the mental state corresponds to a type, each psychological adjustment information corresponds to a type, the feature vector output by the type layer is determined by the type, and at the same time, the review feature data, the historical psychological state score, the historical mental state score, the historical psychological adjustment information, the current psychological state score, the current mental state score are input to the order layer, each psychological consultation in the order layer corresponds to an order, such as the psychological state of the 1 st psychological consultation, the order corresponding to the mental state and the recommended mental regulation information is the same, namely, the order 1, the order corresponding to the mental state, the mental state and the recommended mental regulation information of the 2 nd psychological consultation is the same, namely, the order 2, so that one feature vector can be output through the order layer, then the feature vectors output by the name layer, the type layer and the order layer are added, the vector addition result is input to a plurality of fully connected layers, the vector output by the last fully connected layer is input to N sigmoid layers in sequence, and finally the mental regulation information can be output through the sigmoid layers.
According to the other determination method of psychological adjustment information provided by the invention, compared with the current manner of manually providing psychological adjustment information for patients by psychological doctors, the method of the invention judges whether the psychological consultation object is a review object or not by responding to psychological consultation signals of the psychological consultation object; then if the psychological consultation object is a re-diagnosis object, acquiring re-diagnosis characteristic data of the re-diagnosis object, historical psychological state data and historical mental state data of the re-diagnosis object in each historical consultation process, historical psychological adjustment information, and current psychological state data and current mental state data of the re-diagnosis object in the current mental consultation process; and finally, inputting the re-diagnosis characteristic data, the historical psychological state data, the historical mental state data, the historical psychological adjustment information, the current psychological state data and the current mental state data into a preset psychological adjustment information prediction model to perform adjustment information prediction, and obtaining psychological adjustment information of the re-diagnosis object aiming at the current heart consultation process. Therefore, under the condition that the psychological consultation object is a re-diagnosis object, the psychological adjustment information of the re-diagnosis object is determined by comprehensively analyzing the re-diagnosis characteristic data, the historical psychological state data, the historical mental state data, the historical psychological adjustment information, the current psychological state data and the current mental state data of the re-diagnosis object by utilizing the preset psychological adjustment information prediction model, so that the problems of low determination efficiency and low determination accuracy caused by manually determining the psychological adjustment information can be avoided, and the determination efficiency and the determination accuracy of the psychological adjustment information can be improved.
Further, as a specific implementation of fig. 1, an embodiment of the present invention provides a device for determining psychological adjustment information, as shown in fig. 3, where the device includes: a judging unit 31, an acquiring unit 32, and a predicting unit 33.
The judging unit 31 may be configured to judge whether the psychological consultation object is a review object in response to a psychological consultation signal of the psychological consultation object.
The obtaining unit 32 may be configured to obtain, if the psychological consultation target is a review target, review feature data of the review target, historical psychological state data and historical mental state data of the review target in each historical consultation process, historical psychological adjustment information, and current psychological state data and current mental state data of the review target in the current mental consultation process.
The prediction unit 33 may be configured to input the review feature data, the historical mental state adjustment information, the current mental state data, and the current mental state data into a preset mental adjustment information prediction model to perform adjustment information prediction, so as to obtain mental adjustment information of the review object for the current mental consultation process.
In a specific application scenario, in order to determine psychological adjustment information of the review object for the current psychological consulting process, as shown in fig. 4, the prediction unit 33 includes a cross processing module 331 and a prediction module 332.
The cross processing module 331 may be configured to cross-process the review feature data, the historical mental state data, the historical mental adjustment information, the current mental state data, and the current mental state data to obtain a mental cross feature vector.
The prediction module 332 may be configured to input the psychological cross feature vector into a preset psychological adjustment information prediction model to predict adjustment information, so as to obtain psychological adjustment information of the review object for the current psychological consultation process.
In a specific application scenario, in order to perform cross processing on the review feature data, the historical mental state data, the historical mental adjustment information, the current mental state data and the current mental state data, the cross processing module 331 includes a determining sub-module, a cross processing sub-module and a transforming sub-module.
The determining submodule can be used for determining a feature vector corresponding to the review feature data, a historical psychological state vector corresponding to the historical psychological state data, a mental state vector corresponding to the historical mental state data, an adjusting vector corresponding to the historical psychological adjusting information and a current psychological state vector corresponding to the current psychological state data.
The cross processing sub-module can be used for carrying out feature level cross processing on the feature vector, the historical psychological state vector, the mental state vector, the adjustment vector and the current psychological state vector to obtain a feature level cross vector.
The cross processing sub-module is further configured to perform element level cross processing on the feature vector, the historical mental state vector, the adjustment vector, and the current mental state vector to obtain an element level cross vector;
the cross processing sub-module is further configured to perform low-order cross processing on the feature vector, the historical psychological state vector, the mental state vector, the adjustment vector, and the current psychological state vector to obtain a low-order cross vector;
the transformation submodule can be used for transforming the feature level cross vector, the element level cross vector and the low-order cross vector by using a preset transformation function to obtain a psychological cross feature vector.
In a specific application scenario, in order to predict the psychological adjustment information, the prediction module 332 includes a feature extraction sub-module and an addition sub-module.
The feature extraction sub-module can be used for inputting the psychological cross feature vector into the name layer for feature extraction to obtain a name feature vector.
The feature extraction sub-module may be further configured to input the psychological cross feature vector to the type layer for feature extraction, to obtain a type feature vector.
The feature extraction sub-module may be further configured to input the psychological cross feature vector to the order layer for feature extraction, to obtain an order feature vector.
The adding submodule can be used for adding the name feature vector, the type feature vector and the order feature vector to obtain a fusion feature vector.
The feature extraction submodule can be used for sequentially inputting the fusion feature vector into the plurality of full-connection layers to perform feature extraction and extracting a psychological adjustment feature vector output by the last full-connection layer.
The feature extraction sub-module may be specifically configured to input the psychological adjustment feature vector to the sigmoid layer, so as to obtain psychological adjustment information of the review object for the current heart consultation process.
In a specific application scenario, in order to train and construct a preset psychological adjustment information prediction model, the device further comprises: a construction unit 34.
The construction unit 34 may be configured to construct a preset initial psychological adjustment information prediction model, and obtain training feature data of a sample training object that has completed a target psychological consultation course and is cured, psychological state training data, mental state training data and corresponding actual psychological adjustment information of the sample training object in each sample consultation process, where the target psychological consultation course includes multiple sample consultation processes.
The prediction unit 33 may be further configured to input training feature data of the sample training object, mental state training data of the sample training object during an nth sample consultation process, historical mental state training data and historical mental state training data of the sample training object during a previous n-1 sample consultation process, and corresponding actual mental regulation information of the historical mental state training data and the historical mental state training data into the preset initial mental regulation information prediction model to perform adjustment information prediction, so as to obtain predicted mental adjustment information of the sample training object for the nth sample consultation process, where n is greater than or equal to 2.
The construction unit 34 may specifically be configured to construct a loss function corresponding to the preset initial psychological adjustment information prediction model based on the actual psychological adjustment information and the predicted psychological adjustment information in the nth sample consultation process.
The construction unit 34 may specifically be configured to construct a preset psychological adjustment information prediction model based on the loss function.
In a specific application scenario, in order to verify the prediction capability of the preset psychological adjustment information prediction model, the device further includes: a scoring unit 35, a determining unit 36, an adding unit 37 and an adjusting unit 38.
The obtaining unit 32 may be configured to obtain verification feature data of a sample verification object, historical psychological state verification data, historical mental state verification data, historical psychological adjustment verification information of the sample verification object in a previous m-1 sample historical consultation process, and current psychological state verification data of the sample verification object in an mth psychological consultation process, where if the psychological consultation process of the sample verification object includes y consultation processes, m is greater than 1 and less than y-1, and m is an integer.
The prediction unit 33 may be further configured to input the verification feature data, the historical psychological state verification data, the historical psychological adjustment verification information, and the current psychological state verification data into the preset psychological adjustment information prediction model to perform mediation information prediction, so as to obtain psychological adjustment verification information of the sample verification object for the mth psychological consultation process.
The obtaining unit 32 may be further specifically configured to obtain mental state verification data after the mental adjustment of the mental adjustment verification information by the sample verification object.
The scoring unit 35 may be configured to score the mental state mediated by the sample verification object based on the mental state verification data, to obtain a first verification scoring value.
The prediction unit 33 may be further configured to predict, using the preset psychological adjustment information prediction model, psychological adjustment information of the sample verification object in each psychological consultation process that follows, so as to obtain last psychological adjustment information of the sample verification object in a last psychological consultation process in a psychological consultation course.
The obtaining unit 32 may be further configured to obtain last psychological state data after the sample verification object performs psychological adjustment on the last psychological adjustment information.
The scoring unit 35 may be further configured to score, based on the last psychological state data, a psychological state of the sample verification object after the psychological consultation course is ended, to obtain a second verification score value.
The determining unit 36 may be configured to determine weight coefficients of the first authentication score value and the second authentication score value, respectively.
The adding unit 37 may be configured to add the first verification score value and the second verification score value based on the weight coefficient, to obtain a prediction effect evaluation value corresponding to the preset psychological adjustment information prediction model.
The prediction unit 33 may be further configured to predict the psychological adjustment information of the psychological consultation object by using the preset psychological adjustment information prediction model if the prediction effect evaluation value is greater than a preset threshold.
The adjusting unit 38 may be configured to perform optimization adjustment on model parameters of the preset psychological adjustment information prediction model by using a preset near-end policy optimization algorithm if the prediction effect evaluation value is less than or equal to a preset threshold value, obtain an adjusted preset psychological adjustment information prediction model, and predict psychological adjustment information of the psychological consultant by using the adjusted preset psychological adjustment information prediction model.
In a specific application scenario, in order to predict the psychological adjustment information of the primary diagnosis object, the obtaining unit 32 may be further configured to obtain primary diagnosis feature data of the primary diagnosis object, primary diagnosis psychological state data and primary diagnosis mental state data of the primary diagnosis object in the current mental consultation process if the psychological consultation object is the primary diagnosis object.
The prediction unit 33 may be further configured to input the feature data of the first diagnosis, the mental state data of the first diagnosis, and the mental state data of the first diagnosis into a preset mental adjustment information prediction model to perform adjustment information prediction, so as to obtain mental adjustment information of the first diagnosis target for the current mental consultation process.
It should be noted that, other corresponding descriptions of each functional module related to the determining device for determining psychological adjustment information provided by the embodiment of the present invention may refer to corresponding descriptions of the method shown in fig. 1, and are not described herein again.
Based on the above method as shown in fig. 1, correspondingly, the embodiment of the present invention further provides a computer readable storage medium, on which a computer program is stored, which when being executed by a processor, implements the following steps: responding to a psychological consultation signal of a psychological consultation object, and judging whether the psychological consultation object is a re-diagnosis object or not; if the psychological consultation object is a re-diagnosis object, acquiring re-diagnosis characteristic data of the re-diagnosis object, historical psychological state data and historical mental state data of the re-diagnosis object in each historical consultation process, historical psychological adjustment information, and current psychological state data and current mental state data of the re-diagnosis object in the current mental consultation process; and inputting the re-diagnosis characteristic data, the historical psychological state data, the historical mental state data, the historical psychological adjustment information, the current psychological state data and the current mental state data into a preset psychological adjustment information prediction model to perform adjustment information prediction, so as to obtain psychological adjustment information of the re-diagnosis object aiming at the current heart consultation process.
Based on the embodiment of the method shown in fig. 1 and the device shown in fig. 3, the embodiment of the invention further provides a physical structure diagram of a computer device, as shown in fig. 5, where the computer device includes: a processor 41, a memory 42, and a computer program stored on the memory 42 and executable on the processor, wherein the memory 42 and the processor 41 are both arranged on a bus 43, the processor 41 performing the following steps when said program is executed: responding to a psychological consultation signal of a psychological consultation object, and judging whether the psychological consultation object is a re-diagnosis object or not; if the psychological consultation object is a re-diagnosis object, acquiring re-diagnosis characteristic data of the re-diagnosis object, historical psychological state data and historical mental state data of the re-diagnosis object in each historical consultation process, historical psychological adjustment information, and current psychological state data and current mental state data of the re-diagnosis object in the current mental consultation process; and inputting the re-diagnosis characteristic data, the historical psychological state data, the historical mental state data, the historical psychological adjustment information, the current psychological state data and the current mental state data into a preset psychological adjustment information prediction model to perform adjustment information prediction, so as to obtain psychological adjustment information of the re-diagnosis object aiming at the current heart consultation process.
According to the technical scheme, whether the psychological consultation object is a re-diagnosis object is judged by responding to the psychological consultation signal of the psychological consultation object; then if the psychological consultation object is a re-diagnosis object, acquiring re-diagnosis characteristic data of the re-diagnosis object, historical psychological state data and historical mental state data of the re-diagnosis object in each historical consultation process, historical psychological adjustment information, and current psychological state data and current mental state data of the re-diagnosis object in the current mental consultation process; and finally, inputting the re-diagnosis characteristic data, the historical psychological state data, the historical mental state data, the historical psychological adjustment information, the current psychological state data and the current mental state data into a preset psychological adjustment information prediction model to perform adjustment information prediction, and obtaining psychological adjustment information of the re-diagnosis object aiming at the current heart consultation process. Therefore, under the condition that the psychological consultation object is a re-diagnosis object, the psychological adjustment information of the re-diagnosis object is determined by comprehensively analyzing the re-diagnosis characteristic data, the historical psychological state data, the historical mental state data, the historical psychological adjustment information, the current psychological state data and the current mental state data of the re-diagnosis object by utilizing the preset psychological adjustment information prediction model, so that the problems of low determination efficiency and low determination accuracy caused by manually determining the psychological adjustment information can be avoided, and the determination efficiency and the determination accuracy of the psychological adjustment information can be improved.
It will be appreciated by those skilled in the art that the modules or steps of the invention described above may be implemented in a general purpose computing device, they may be concentrated on a single computing device, or distributed across a network of computing devices, they may alternatively be implemented in program code executable by computing devices, so that they may be stored in a memory device for execution by computing devices, and in some cases, the steps shown or described may be performed in a different order than that shown or described, or they may be separately fabricated into individual integrated circuit modules, or multiple modules or steps within them may be fabricated into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (10)

1. A method for determining psychological adjustment information, comprising:
Responding to a psychological consultation signal of a psychological consultation object, and judging whether the psychological consultation object is a re-diagnosis object or not;
if the psychological consultation object is a re-diagnosis object, acquiring re-diagnosis characteristic data of the re-diagnosis object, historical psychological state data and historical mental state data of the re-diagnosis object in each historical consultation process, historical psychological adjustment information, and current psychological state data and current mental state data of the re-diagnosis object in the current mental consultation process;
and inputting the re-diagnosis characteristic data, the historical psychological state data, the historical mental state data, the historical psychological adjustment information, the current psychological state data and the current mental state data into a preset psychological adjustment information prediction model to perform adjustment information prediction, so as to obtain psychological adjustment information of the re-diagnosis object aiming at the current heart consultation process.
2. The method according to claim 1, wherein the inputting the review feature data, the historical mental state adjustment information, the current mental state data, and the current mental state data into a preset mental state adjustment information prediction model to perform adjustment information prediction, to obtain the mental adjustment information of the review object for the current mental consultation process, includes:
Performing cross processing on the re-diagnosis feature data, the historical psychological state data, the historical mental state data, the historical psychological adjustment information, the current psychological state data and the current mental state data to obtain psychological cross feature vectors;
and inputting the psychological cross characteristic vector into a preset psychological adjustment information prediction model to predict adjustment information, so as to obtain the psychological adjustment information of the review object aiming at the current psychological consultation process.
3. The method according to claim 2, wherein the cross-processing the review feature data, the historical mental state adjustment information, the current mental state data, and the current mental state data to obtain a mental cross feature vector includes:
determining a feature vector corresponding to the re-diagnosis feature data, a historical psychological state vector corresponding to the historical psychological state data, a mental state vector corresponding to the historical mental state data, an adjustment vector corresponding to the historical psychological adjustment information and a current psychological state vector corresponding to the current psychological state data;
performing feature level cross processing on the feature vector, the historical psychological state vector, the mental state vector, the adjustment vector and the current psychological state vector to obtain a feature level cross vector;
Performing element level cross processing on the feature vector, the historical psychological state vector, the mental state vector, the adjustment vector and the current psychological state vector to obtain an element level cross vector;
performing low-order cross processing on the feature vector, the historical psychological state vector, the mental state vector, the adjustment vector and the current psychological state vector to obtain a low-order cross vector;
and carrying out transformation processing on the feature level cross vector, the element level cross vector and the low-order cross vector by using a preset transformation function to obtain a psychological cross feature vector.
4. The method of claim 2, wherein the pre-set psychometric information prediction model comprises a name layer, a type layer, a sequence layer, a plurality of fully connected layers, a sigmoid layer; inputting the psychological cross feature vector into a preset psychological adjustment information prediction model for adjustment information prediction to obtain psychological adjustment information of the review object aiming at the current psychological consultation process, wherein the method comprises the following steps:
inputting the psychological cross feature vector into the name layer for feature extraction to obtain a name feature vector;
inputting the psychological cross feature vector into the type layer for feature extraction to obtain a type feature vector;
Inputting the psychological cross feature vector into the sequence layer for feature extraction to obtain a sequence feature vector;
adding the name feature vector, the type feature vector and the order feature vector to obtain a fusion feature vector;
sequentially inputting the fusion feature vectors to the plurality of full-connection layers for feature extraction, and extracting a psychological adjustment feature vector output by the last full-connection layer;
and inputting the psychological adjustment feature vector to the sigmoid layer to obtain psychological adjustment information of the re-diagnosis object aiming at the current heart consultation process.
5. The method according to claim 1, wherein before the inputting the review feature data, the historical mental state adjustment information, the current mental state data, and the current mental state data into a preset mental state adjustment information prediction model for adjustment information prediction, the method further comprises:
constructing a preset initial psychological adjustment information prediction model, and acquiring training characteristic data of a sample training object which is cured after completing a target psychological consultation course, psychological state training data, mental state training data and corresponding actual psychological adjustment information of the sample training object in each sample consultation process, wherein the target psychological consultation course comprises a plurality of sample consultation processes;
Inputting training characteristic data of the sample training object, psychological state training data, mental state training data of the sample training object in the nth sample consultation process, historical psychological state training data and historical mental state training data of the sample training object in the previous n-1 sample consultation processes and corresponding actual mental regulation information of the historical psychological state training data into the preset initial psychological regulation information prediction model for regulation information prediction, and obtaining predicted psychological regulation information of the sample training object in the nth sample consultation process, wherein n is more than or equal to 2;
constructing a loss function corresponding to the preset initial psychological adjustment information prediction model based on the actual psychological adjustment information and the predicted psychological adjustment information in the nth sample consultation process;
and constructing a preset psychological adjustment information prediction model based on the loss function.
6. The method of claim 5, wherein after said constructing a predictive model of preset psychometric information based on said loss function, said method further comprises:
acquiring verification feature data of a sample verification object, historical psychological state verification data, historical mental state verification data and historical psychological adjustment verification information of the sample verification object in the previous m-1 sample historical consultation processes, and current psychological state verification data of the sample verification object in the mth psychological consultation process, wherein if the psychological consultation process of the sample verification object comprises y consultation processes, m is more than 1 and less than y-1, and m is an integer;
Inputting the verification feature data, the historical psychological state verification data, the historical psychological adjustment verification information and the current psychological state verification data into the preset psychological adjustment information prediction model for mediation information prediction to obtain psychological adjustment verification information of the sample verification object in the m-th psychological consultation process;
obtaining psychological state verification data of the sample verification object after psychological adjustment is carried out on the psychological adjustment verification information;
scoring the psychological state mediated by the sample verification object based on the psychological state verification data to obtain a first verification scoring value;
predicting the psychological adjustment information of the sample verification object in each psychological consultation process by using the preset psychological adjustment information prediction model to obtain the last psychological adjustment information of the sample verification object in the last psychological consultation process in the psychological consultation course;
acquiring last psychological state data of the sample verification object after psychological adjustment is carried out on the last psychological adjustment information;
scoring the psychological state of the sample verification object after the psychological consultation treatment course is finished based on the last psychological state data to obtain a second verification scoring value;
Determining weight coefficients of the first verification score value and the second verification score value respectively;
based on the weight coefficient, adding the first verification score value and the second verification score value to obtain a prediction effect score value corresponding to the preset psychological adjustment information prediction model;
if the prediction effect evaluation value is larger than a preset threshold value, predicting the psychological adjustment information of the psychological consultation object by using the preset psychological adjustment information prediction model;
and if the prediction effect evaluation value is smaller than or equal to a preset threshold value, optimizing and adjusting model parameters of the preset psychological adjustment information prediction model by using a preset near-end strategy optimization algorithm to obtain an adjusted preset psychological adjustment information prediction model, and predicting psychological adjustment information of the psychological consultation object by using the adjusted preset psychological adjustment information prediction model.
7. The method of claim 1, wherein after said determining whether the psychological consulting object is a review object, the method further comprises:
if the psychological consultation object is a primary consultation object, primary consultation characteristic data of the primary consultation object, primary consultation psychological state data and primary consultation mental state data of the primary consultation object in the current heart consultation process are obtained;
Inputting the initial diagnosis characteristic data, the initial diagnosis mental state data and the initial diagnosis mental state data into a preset mental regulation information prediction model to predict regulation information, and obtaining mental regulation information of the initial diagnosis object aiming at the current mental consultation process.
8. A device for determining psychological adjustment information, comprising:
the judging unit is used for responding to the psychological consultation signal of the psychological consultation object and judging whether the psychological consultation object is a review object or not;
the system comprises an acquisition unit, a review unit and a review unit, wherein the acquisition unit is used for acquiring review characteristic data of a review object, historical psychological state data and historical mental state data of the review object in each historical consultation process, historical psychological adjustment information and current psychological state data and current mental state data of the review object in the current mental consultation process if the psychological consultation object is the review object;
the prediction unit is used for inputting the review feature data, the historical psychological state data, the historical mental state data, the historical psychological adjustment information, the current psychological state data and the current mental state data into a preset psychological adjustment information prediction model to perform adjustment information prediction, so that psychological adjustment information of the review object in the current heart consultation process is obtained.
9. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 7.
10. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the computer program when executed by the processor implements the steps of the method according to any one of claims 1 to 7.
CN202311720130.8A 2023-12-14 2023-12-14 Psychological adjustment information determining method and device, storage medium and computer equipment Pending CN117711572A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311720130.8A CN117711572A (en) 2023-12-14 2023-12-14 Psychological adjustment information determining method and device, storage medium and computer equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311720130.8A CN117711572A (en) 2023-12-14 2023-12-14 Psychological adjustment information determining method and device, storage medium and computer equipment

Publications (1)

Publication Number Publication Date
CN117711572A true CN117711572A (en) 2024-03-15

Family

ID=90154729

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311720130.8A Pending CN117711572A (en) 2023-12-14 2023-12-14 Psychological adjustment information determining method and device, storage medium and computer equipment

Country Status (1)

Country Link
CN (1) CN117711572A (en)

Similar Documents

Publication Publication Date Title
US10854335B2 (en) Computer aided medical method and medical system for medical prediction
KR102190325B1 (en) A method for building a computer-implemented computing device for assessment of qualitative features from face images
KR20210066697A (en) Apparatus and method for predicting human depression level using multi-layer bi-lstm with spatial and dynamic information of video frames
US20220319234A1 (en) Object Behavior Analysis Method, Information Display Method, and Electronic Device
CN115620384B (en) Model training method, fundus image prediction method and fundus image prediction device
CN115547437A (en) Medical federal party-based training system and method
CN116113356A (en) Method and device for determining user dementia degree
CN113712573A (en) Electroencephalogram signal classification method, device, equipment and storage medium
CN110704668B (en) Grid-based collaborative attention VQA method and device
CN113282840B (en) Comprehensive training acquisition management platform
CN112262399A (en) Action learning device, action learning method, action learning system, program, and recording medium
CN117747087A (en) Training method of large inquiry model, inquiry method and device based on large inquiry model
CN116680391A (en) Custom dialogue method, training method, device and equipment for custom dialogue model
CN117711572A (en) Psychological adjustment information determining method and device, storage medium and computer equipment
CN116628161A (en) Answer generation method, device, equipment and storage medium
JP7014761B2 (en) Cognitive function estimation method, computer program and cognitive function estimation device
CN113704496A (en) Medical knowledge map repairing method and device, computer equipment and storage medium
Serrano et al. Plausibility validation of a decision making model using subjects’ explanations of decisions
CN111968632A (en) Call voice acquisition method and device, computer equipment and storage medium
KR102610267B1 (en) Method for analyzing status of specific user corresponding to specific avatar by referring to interactions between the specific avatar and other avatars in the metaverse world and providing service to the specific user and device using the same
KR102485600B1 (en) a depression management system using an AI(artificial intelligence) robot
WO2021025094A1 (en) Action learning device, action learning method, action determination device, action determination method, action learning system, program, and recording medium
CN117197613B (en) Image quality prediction model training method and device and image quality prediction method and device
CN113256368B (en) Product pushing method and device, computer equipment and storage medium
Bhattacharya et al. Detection of Psychological Trauma in Refugee Children by Analysis of Drawings using Machine Learning

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