CN115910325A - Modeling method for cognitive task evaluation, cognitive task evaluation method and system - Google Patents
Modeling method for cognitive task evaluation, cognitive task evaluation method and system Download PDFInfo
- Publication number
- CN115910325A CN115910325A CN202211512702.9A CN202211512702A CN115910325A CN 115910325 A CN115910325 A CN 115910325A CN 202211512702 A CN202211512702 A CN 202211512702A CN 115910325 A CN115910325 A CN 115910325A
- Authority
- CN
- China
- Prior art keywords
- task
- cognitive training
- cognitive
- user
- training task
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a modeling method for cognitive task evaluation, a cognitive task evaluation method and a cognitive task evaluation system. The modeling method comprises the steps of obtaining user information of each user in a user set; acquiring behavior reflection data and evaluation results of each user on the same cognitive training task; acquiring a plurality of task indexes corresponding to the cognitive training task based on a diffusion project reflection theory according to the user information, the behavior reflection data and the evaluation result of each user; acquiring a plurality of inherent attributes of the cognitive training task; repeating the steps until each cognitive training task in the task set acquires a plurality of corresponding task indexes and a plurality of corresponding inherent attributes; and constructing a prediction model by utilizing a plurality of task indexes and a plurality of inherent attributes which respectively correspond to each cognitive training task. Therefore, the content quality of the cognitive training task can be automatically evaluated based on the constructed prediction model, the content quality of the cognitive training task can be evaluated without collecting real user data, and low-quality content is removed.
Description
Technical Field
The invention relates to a modeling method for cognitive task evaluation, and also relates to a cognitive task evaluation method and a cognitive task evaluation system for a model constructed by using the modeling method, belonging to the technical field of automatic cognitive task evaluation.
Background
At present, electronic equipment (such as computers and tablets) based cognitive digital therapy is receiving more and more attention, and the treatment effect of the cognitive digital therapy on various cognitive impairments is supported by more and more clinical test results. As a core content in cognitive digital therapy, the design of training tasks directly affects clinical efficacy. However, there is currently a lack of quantitative assessment criteria for the content of the training task, which adds uncertainty to subsequent clinical trials and efficacy validation. Therefore, if the research of the cognitive digital therapy is further deepened and the application of the cognitive digital therapy is promoted, a set of evaluation methods for the cognitive digital therapy must be formed.
Psychometrics have been widely used as a traditional neuropsychological verification method (e.g., validity analysis of paper screening tools). Among them, a new generation psychometric theory framework represented by a project reaction theory is increasingly applied. The project reaction theory framework is a set of topic evaluation technology based on a potential feature theory, and the quality of a topic is quantitatively evaluated in the topic development stage by evaluating different features of the topic, so that the topic has more pertinence in application. The project reaction theory framework comprises a plurality of models, can solve different problems, is widely applied to education evaluation, and has started to have application cases in recent years in cognitive evaluation tasks for developing or optimizing paper neuropsychological measuring tools.
However, quantitative assessment methods based on psychometrics require collection of large amounts of data, raising the development cost of new training tasks. Automatic evaluation based on machine learning algorithms has been widely used in many fields. For example, in the field of marketing, machine learning algorithms are used to predict the effectiveness of an advertisement and select the best ad creative, thereby maximizing the input-output ratio. In the entertainment field, machine learning algorithms are used to select good artwork to optimize the input-output ratio of the investor. However, in the field of evaluation of content quality of cognitive training tasks, there is no application of automatic evaluation based on artificial intelligence algorithms.
Disclosure of Invention
The invention aims to provide a modeling method for cognitive task evaluation.
The invention also aims to provide a cognitive task evaluation method.
The invention also aims to provide a cognitive task evaluation system.
In order to realize the technical purpose, the invention adopts the following technical scheme:
according to a first aspect of embodiments of the present invention, there is provided a modeling method for cognitive task evaluation, comprising the steps of:
acquiring user information of each user in a user set;
acquiring behavior reflection data and evaluation results of the users in the user set on the same cognitive training task;
acquiring a plurality of task indexes corresponding to the cognitive training task based on a diffusion project reflection theory according to the user information, behavior reflection data and evaluation results of the users in the user set;
acquiring a plurality of inherent attributes of the cognitive training task;
repeating the steps to obtain a plurality of task indexes and a plurality of inherent attributes corresponding to different cognitive training tasks until each cognitive training task in the task set obtains a plurality of corresponding task indexes and a plurality of inherent attributes;
and constructing a prediction model by utilizing a plurality of task indexes and a plurality of inherent attributes which respectively correspond to each cognitive training task in the task set.
Preferably, the plurality of task indicators of the cognitive training task at least include: a time pressure index, a task difficulty index and a satisfaction index;
the plurality of intrinsic attributes of the cognitive training task include at least: the method comprises the following steps of target object number, training time, pre-calibrated target response and cognitive domain aiming.
Preferably, the acquiring a plurality of task indexes corresponding to the cognitive training task based on a diffusion project reflection theory according to the user information, the behavior reflection data and the evaluation result of each user in the user set specifically includes:
according to the user information and behavior reflection data of each user in the user set, iterative calculation is carried out based on a formula 1 and a formula 2 to obtain a time pressure index a of the cognitive training task i And a task difficulty index v i ；
Wherein, a pi Representing a time pressure value of the cognitive training task i for the user p; yp represents the user response caution; a is a i A time pressure indicator representing a cognitive training task; p (X) pi (= 1 Theta) |) p Yp) represents a probability; theta p Representing a current cognitive ability of the user; x pi =1 represents training task i for user p;
and calculating a satisfaction index of the cognitive training task through a weight assignment method according to the evaluation result of each user in the user set.
Preferably, the constructing of the prediction model by using the plurality of task indexes and the plurality of inherent attributes respectively corresponding to each cognitive training task in the task set specifically includes:
time pressure index a based on each cognitive training task in the task set i Constructing a first predictor model according to a plurality of inherent attributes of the cognitive training task;
task difficulty index v based on cognitive training tasks in task set i And a plurality of inherits of the cognitive training taskAttribute, constructing a second predictor model;
constructing a third prediction submodel based on the satisfaction index of each cognitive training task in the task set and a plurality of inherent attributes of the cognitive training tasks;
wherein the first predictor model, the second predictor model and the third predictor model together form the prediction model.
Preferably, the first predictor model is constructed by:
constructing a first prediction submodel by an XGboost algorithm, wherein the model is expressed as follows:
wherein: m is the number of decision trees, T (x, theta M) is a certain decision tree, theta M is the number of corresponding decision trees, x is a vector formed by inherent attributes of the task, and fM (x) is a training task time pressure index a i The predicted value of (2).
Preferably, the second predictor model is constructed by:
and constructing a second predictor model through the XGboost algorithm, wherein the model is expressed as follows:
wherein: m is the number of decision trees, T (x, theta M) is a certain decision tree, theta M is the number of corresponding decision trees, x is a vector formed by inherent attributes of the task, fM (x) is a task difficulty index v of the training task i The predicted value of (2).
Preferably, the third predictor model is constructed by:
and constructing a third predictor model by an XGboost algorithm, wherein the model is expressed as follows:
wherein: m is the number of the decision trees, T (x, theta M) is a certain decision tree, theta M is the number of the corresponding decision trees, x is a vector formed by inherent attributes of the task, and fM (x) is a predicted value of a satisfaction index of the training task.
According to a second aspect of the embodiments of the present invention, there is provided a method for evaluating a cognitive training task, including the steps of:
acquiring a plurality of inherent attributes of the cognitive training task to be evaluated;
inputting the inherent attributes into a prediction model, and outputting a plurality of task indexes corresponding to the cognitive training task to be evaluated so as to comprehensively evaluate the cognitive training task to be evaluated through the task indexes;
the prediction model is constructed by the modeling method.
Preferably, the inputting the inherent attributes into a prediction model, and the outputting the task indexes corresponding to the cognitive training task to be evaluated specifically includes:
inputting the plurality of inherent attributes into a first predictor model to obtain a time pressure index a of the cognitive training task i ；
Inputting the plurality of inherent attributes into a second predictor model to obtain a task difficulty index v of the cognitive training task i ；
Inputting the inherent attributes into a third predictor model to obtain a satisfaction index of the cognitive training task;
wherein the first predictor model, the second predictor model and the third predictor model together form the prediction model.
According to a third aspect of the embodiments of the present invention, there is provided an evaluation system for cognitive training tasks, including a processor and a memory, the processor reading a computer program in the memory for performing the following operations:
acquiring a plurality of inherent attributes of the cognitive training task to be evaluated;
inputting the inherent attributes into a prediction model, and outputting a plurality of task indexes corresponding to the cognitive training task to be evaluated so as to comprehensively evaluate the cognitive training task to be evaluated through the task indexes;
the prediction model is constructed by the modeling method.
The invention has the following technical effects:
1. the method is based on a psychometric theory and a machine learning algorithm, and a set of evaluation model for cognitive task evaluation is constructed by comprehensively evaluating time pressure, problem difficulty and satisfaction degree of a task and combining the inherent attributes of learning a cognitive training task by the machine learning algorithm. And forming a set of evaluation method for rapidly evaluating the content quality of the new cognitive training task based on the evaluation model so as to achieve the aim of evaluating the content quality of the cognitive training task without collecting real user data.
2. Before a game-based cognitive training task designed for a cognitive impairment user is actually applied, the evaluation method based on the invention can quickly evaluate the content quality of the cognitive training task to obtain quantitative evaluation indexes, avoid excessive resource input to low-quality content, provide data support for computerized self-adaptive pushing and provide early-stage data guarantee for subsequent clinical tests.
Drawings
FIG. 1 provides an overall flow chart of a modeling method for cognitive task assessment according to a first embodiment of the present invention;
FIG. 2 is a detailed flow chart of a modeling method for cognitive task assessment according to a first embodiment of the present invention;
FIG. 3 is a flowchart of a cognitive task assessment method according to a second embodiment of the present invention;
FIG. 4 is a flowchart illustrating a cognitive task assessment method according to a second embodiment of the present invention;
fig. 5 is a structural diagram of a cognitive task evaluation system according to a third embodiment of the present invention.
Detailed Description
The technical contents of the invention are described in detail below with reference to the accompanying drawings and specific embodiments.
< first embodiment >
As shown in fig. 1, a first embodiment of the present invention provides a modeling method for cognitive task evaluation, which specifically includes steps S1 to S6:
s1: and acquiring user information of each user in the user set.
Specifically, the user set in this embodiment is mostly historical users under big data, and new users may also be added, so as to form a user set composed of one hundred thousand users or million users, so as to improve the comprehensiveness and accuracy of the evaluation model. As shown in fig. 2, the user information in this embodiment at least includes: statistical information and condition information. Wherein, the statistical information comprises information such as user name, gender, age, education degree and the like; the disease information includes information of the type of disease, the severity of disease, the time of illness, etc. of the user.
It can be understood that, for a historical user, since the historical user is subjected to cognition enhancement training, the historical user is subjected to cognition evaluation before training to obtain the current cognitive ability of the user; for a new user, certain cognitive assessment can be performed on the user through the cognitive scale so as to obtain the current cognitive ability of the user, and therefore data support is provided for subsequent model construction.
S2: acquiring behavior reflection data and evaluation results of the users in the user set on the same cognitive training task.
Specifically, in this embodiment, for a certain cognitive training task, all users in the user set are required to complete one training of the task, so that in the training process, behavior reflection data of each user for the cognitive training task can be obtained, and after the training is completed, an evaluation result of each user for the cognitive training task is obtained.
It can be understood that, for the historical user, the historical user is required to complete the cognitive training task when cognitive improvement training is performed, so that behavior reflection data and evaluation results of the historical user for the cognitive training task are obtained based on historical data. For a new user, the new user is required to complete the cognitive training task once, so that the behavior reflection data of the new user is obtained in the training process, and the evaluation result of the new user on the cognitive training task is obtained after the training is completed.
S3: and acquiring a plurality of task indexes corresponding to the cognitive training task based on a diffusion project reflection theory according to the user information, behavior reflection data and evaluation results of the users in the user set.
In this embodiment, the plurality of task indicators of the cognitive training task at least include: a time pressure index, a task difficulty index, and a satisfaction index. Specifically, the method comprises the following steps of S31-S32:
s31: obtaining a time pressure index a i And a task difficulty index v i ；
Specifically, according to the user information and behavior reflection data of each user in the user set, iterative calculation is performed based on formula 1 and formula 2 to obtain the time pressure index a of the cognitive training task i And a task difficulty index v i ；
Wherein, a pi Representing a time pressure value of the cognitive training task i for the user p; yp represents the user's degree of caution in responding; a is i A time pressure indicator representing a cognitive training task; p (X) pi (= 1 | theta) p Yp) represents a probability; theta.theta. p Representing a current cognitive ability of the user; x pi =1 represents training task i for user p.
S32: obtaining a satisfaction index;
after the user completes the cognitive training task, the patient is required to fill in a questionnaire. The questionnaire in this embodiment at least includes the following:
(1) In general, how much is you satisfied with the set of cognitive training?
(2) How much do you think the set of cognitive training work to improve your cognitive function?
| Has no effect | Is very useful | |||||||||
| 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 |
(3) In the set of cognitive training, is there a part that is hard for you to understand?
| Is that | Whether or not |
| 1 | 2 |
(4) What aspects of the cognitive training are difficult for you to understand? Please select all parts that are hard to understand for you. [ show only for the user who selects 1 out of 3 ]
Instruction words
Picture design
Answering mode
Logic arrangement
<xnotran> , _________ </xnotran>
(5) How much you expect the next cognitive training after completing the set of cognitive training?
(6) How much do you feel engaged in completing cognitive training?
(7) How much do the set of cognitive training feel your pleased?
And when a certain user finishes the questionnaire, evaluating the cognitive training task based on the specific options of the user in the questionnaire to obtain the evaluation result of the user on the cognitive training task. When all the users in the user set evaluate the cognitive training task based on the questionnaire, a plurality of evaluation results can be obtained.
And calculating a satisfaction index of the cognitive training task by a weight assignment method according to the evaluation result of each user in the user set.
It should be understood that the questionnaire in this embodiment only shows a part of the content, and the rest of the content may be adaptively increased or decreased according to the survey requirement, and is not limited in particular.
S4: a plurality of inherent attributes of the cognitive training task are obtained.
In this embodiment, the plurality of inherent attributes of the cognitive training task at least include: the number of target objects, the training time, the pre-calibrated target response and the cognitive domain.
It is understood that the plurality of intrinsic attributes are possessed by each cognitive training task, and only the specific attribute values are different. For example: the number of the target objects of the rat hitting task is 3, the training time is 2 minutes, the pre-calibrated target response is reflected by hitting all the target rats, and the targeted cognitive domain is attention. The number of the targets of the similar object recognition task is 5, the training time is 2 minutes, the pre-calibrated target response is reflected by recognizing the object which just appears, and the specific cognitive domain is the memory.
S5: and repeating the steps S2-S4 until each cognitive training task in the task set acquires a plurality of corresponding task indexes and a plurality of corresponding inherent attributes.
In this embodiment, the plurality of task indexes and the plurality of inherent attributes of a certain cognitive training task in the task set may be obtained through steps S2 to S4, and the plurality of task indexes and the plurality of inherent attributes corresponding to different cognitive training tasks may be obtained by repeating the steps S2 to S4 until each cognitive training task in the task set obtains the corresponding plurality of task indexes and the plurality of inherent attributes.
S6: and constructing a prediction model by utilizing a plurality of task indexes and a plurality of inherent attributes which respectively correspond to each cognitive training task in the task set.
The prediction model in this embodiment includes three different sub-models respectively corresponding to the time pressure index a i Task difficulty index v i And a satisfaction index. The method specifically comprises the following steps of S61-S63:
s61: time pressure index a based on each cognitive training task in task set i Establishing a first prediction submodel by a plurality of inherent attributes of the cognitive training task;
specifically, in this embodiment, a first predictor model is constructed through the XGBoost algorithm, and the model expression is as follows:
wherein: m is the number of decision trees, T (x, theta M) is a certain decision tree, theta M is the number of corresponding decision trees, x is a vector formed by inherent attributes of the task, and fM (x) is a time pressure index a of the training task i The predicted value of (2).
S62: task difficulty index v based on task centralization cognitive training tasks i Establishing a second predictor model by a plurality of inherent attributes of the cognitive training task;
specifically, in this embodiment, a second predictor model is constructed through the XGBoost algorithm, and the model expression is as follows:
wherein: m is the number of decision trees, T (x, theta M) is a certain decision tree, theta M is the number of corresponding decision trees, x is a vector formed by inherent attributes of the task, fM (x) is a task difficulty index v of the training task i The predicted value of (2).
S63: constructing a third prediction submodel based on the satisfaction index of each cognitive training task in the task set and a plurality of inherent attributes of the cognitive training tasks;
specifically, in this embodiment, a third predictor model is constructed through the XGBoost algorithm, and the model expression is as follows:
wherein: m is the number of the decision trees, T (x, theta M) is a certain decision tree, theta M is the number of the corresponding decision trees, x is a vector formed by inherent attributes of the task, and fM (x) is a predicted value of a satisfaction index of the training task.
It can be understood that, based on the first predictor model constructed in step S61, by inputting a plurality of attribute values of the cognitive training task to be evaluated, the time pressure index a of the cognitive training task to be evaluated can be calculated through the model i . Similarly, based on the second predictor model constructed in the step S62, by inputting a plurality of attribute values of the cognitive training task to be evaluated, the task difficulty index v of the cognitive training task to be evaluated can be calculated by the model i . Based on the third predictor model constructed in the step S63, by inputting a plurality of attribute values of the cognitive training task to be evaluated, the satisfaction index of the cognitive training task to be evaluated can be calculated by the model.
Thus, the pressure index a can be output by using the prediction model i Task difficulty index v i And satisfaction index, thereby enabling recognition of the evaluationAnd (5) knowing the training task for evaluation.
< second embodiment >
As shown in fig. 3 and 4, on the basis of the first embodiment, a second embodiment of the present invention provides a method for evaluating a cognitive training task, which specifically includes steps S10 to S20:
s10: and acquiring a plurality of inherent attributes of the cognitive training task to be evaluated.
The plurality of intrinsic attributes are the same as the intrinsic attributes in step S4, but have different specific attribute values.
S20: and comprehensively evaluating the cognitive training task to be evaluated.
Inputting a plurality of inherent attributes into a prediction model, and outputting a plurality of task indexes corresponding to the cognitive training task to be evaluated so as to comprehensively evaluate the cognitive training task to be evaluated through the plurality of task indexes; the prediction model is constructed by the modeling method.
Specifically, the method comprises the steps S201-S203:
s201: inputting a plurality of inherent attributes into a first predictor model to obtain a time pressure index a of the cognitive training task i ；
S202: inputting a plurality of inherent attributes into a second prediction submodel to obtain a task difficulty index v of the cognitive training task i ；
S203: inputting a plurality of inherent attributes into a third predictor model to obtain a satisfaction index of the cognitive training task;
thereby, based on the acquired time pressure index a i Task difficulty index v i And a satisfaction index for evaluating the cognitive training task.
< third embodiment >
On the basis of the second embodiment, the invention further provides an evaluation system for cognitive training tasks. As shown in fig. 5, the evaluation system includes one or more processors 21 and a memory 22. Wherein the memory 22 is coupled to the processor 21 for storing one or more programs, which when executed by the one or more processors 21, cause the one or more processors 21 to implement the evaluation method of the cognitive training task as in the above embodiments.
The processor 21 is configured to control the overall operation of the evaluation system, so as to complete all or part of the steps of the evaluation method for the cognitive training task. The processor 21 may be a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), a Field Programmable Gate Array (FPGA), an Application Specific Integrated Circuit (ASIC), a Digital Signal Processing (DSP) chip, or the like. The memory 22 is used to store various types of data to support operation of the evaluation system, which may include, for example, instructions for any application or method operating on the evaluation system, as well as application-related data. The memory 22 may be implemented by any type or combination of volatile or non-volatile memory devices, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, and the like.
In an exemplary embodiment, the evaluation system may be implemented by a computer chip or an entity, or a product with certain functions, and is used for executing the evaluation method of the cognitive training task, and achieving the technical effect consistent with the method. One typical embodiment is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a vehicle-mounted human-computer interaction device, a cellular telephone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.
In another exemplary embodiment, the present invention further provides a computer readable storage medium including program instructions, which when executed by a processor, implement the steps of the assessment method of the cognitive training task in any one of the above embodiments. For example, the computer readable storage medium may be the memory described above including program instructions executable by the processor of the evaluation system to perform the method of evaluating a cognitive training task described above and achieve a technical effect consistent with the method described above.
The modeling method, the cognitive task evaluation method and the cognitive task evaluation system for cognitive task evaluation provided by the embodiment of the invention are explained in detail above. It will be apparent to those skilled in the art that any obvious modifications thereof can be made without departing from the spirit of the invention, which infringes the patent right of the invention and bears the corresponding legal responsibility.
1. A modeling method for cognitive task assessment, comprising the steps of:
acquiring user information of each user in a user set;
acquiring behavior reflection data and evaluation results of the users in the user set on the same cognitive training task;
acquiring a plurality of task indexes corresponding to the cognitive training task based on a diffusion project reflection theory according to user information, behavior reflection data and evaluation results of all the users in the user set;
acquiring a plurality of inherent attributes of the cognitive training task;
repeating the steps to obtain a plurality of task indexes and a plurality of inherent attributes corresponding to different cognitive training tasks until each cognitive training task in the task set obtains a plurality of corresponding task indexes and a plurality of corresponding inherent attributes;
and constructing a prediction model by utilizing a plurality of task indexes and a plurality of inherent attributes which respectively correspond to each cognitive training task in the task set.
2. A modeling method in accordance with claim 1, wherein:
the plurality of task indicators of the cognitive training task at least comprise: a time pressure index, a task difficulty index and a satisfaction index;
the plurality of intrinsic attributes of the cognitive training task include at least: the number of target objects, the training time, the pre-calibrated target response and the cognitive domain.
3. The modeling method according to claim 2, wherein the obtaining of the plurality of task indicators corresponding to the cognitive training task based on the diffusion item reflection theory according to the user information, the behavior reflection data, and the evaluation result of each user in the user set specifically includes:
according to the user information and behavior reflection data of each user in the user set, iterative calculation is carried out based on a formula 1 and a formula 2 to obtain a time pressure index a of the cognitive training task i And a task difficulty index v i ；
Wherein, a pi Representing a time pressure value of the cognitive training task i for the user p; yp represents the user response caution; a is a i A time pressure indicator representing a cognitive training task; p (X) pi (= 1 | theta) p Yp) represents a probability; theta p Representing a current cognitive ability of the user; x pi =1 represents training task i for user p;
and calculating a satisfaction index of the cognitive training task through a weight assignment method according to the evaluation result of each user in the user set.
4. The modeling method according to claim 2, wherein the constructing of the prediction model using the plurality of task indexes and the plurality of intrinsic attributes respectively corresponding to the cognitive training tasks in the task set specifically includes:
time pressure index a based on each cognitive training task in the task set i Constructing a first predictor model according to a plurality of inherent attributes of the cognitive training task;
task difficulty index v based on cognitive training tasks in task set i And a plurality of inherent attributes of the cognitive training task, and constructing a second predictor model;
constructing a third prediction submodel based on the satisfaction index of each cognitive training task in the task set and a plurality of inherent attributes of the cognitive training tasks;
wherein the first predictor model, the second predictor model and the third predictor model together form the prediction model.
5. A modeling method as claimed in claim 4, characterized in that the first predictor model is constructed by:
constructing a first prediction submodel by an XGboost algorithm, wherein the model is expressed as follows:
wherein: m is the number of decision trees, T (x, theta M) is a certain decision tree, theta M is the number of corresponding decision trees, x is a vector formed by inherent attributes of the task, and fM (x) is a training task time pressure index a i The predicted value of (2).
6. A modeling method as claimed in claim 4, characterized in that the second predictor model is constructed by:
and constructing a second predictor model through the XGboost algorithm, wherein the model is expressed as follows:
wherein: m is the number of decision trees, T (x, theta M) is a certain decision tree, theta M is the number of corresponding decision trees, x is a vector formed by inherent attributes of the task, fM (x) is a task difficulty index v of the training task i The predicted value of (2).
7. The modeling method of claim 5, wherein the third predictor model is constructed by:
and constructing a third predictor model through the XGboost algorithm, wherein the model is expressed as follows:
wherein: m is the number of the decision trees, T (x, theta M) is a certain decision tree, theta M is the number of the corresponding decision trees, x is a vector formed by inherent attributes of the task, and fM (x) is a predicted value of a satisfaction index of the training task.
8. The evaluation method of the cognitive training task is characterized by comprising the following steps of:
acquiring a plurality of inherent attributes of the cognitive training task to be evaluated;
inputting the inherent attributes into a prediction model, and outputting a plurality of task indexes corresponding to the cognitive training task to be evaluated so as to comprehensively evaluate the cognitive training task to be evaluated through the task indexes;
wherein the predictive model is constructed by a modeling method according to any one of claims 1 to 7.
9. The evaluation method according to claim 8, wherein the inputting of the plurality of intrinsic attributes into a prediction model and the outputting of a plurality of task indexes corresponding to the cognitive training task to be evaluated specifically comprises:
inputting the plurality of inherent attributes into a first predictor model to obtain a time pressure index a of the cognitive training task i ；
Inputting the inherent attributes into a second predictor model to obtain a task difficulty index v of the cognitive training task i ；
Inputting the inherent attributes into a third predictor model to obtain a satisfaction index of the cognitive training task;
wherein the first predictor model, the second predictor model and the third predictor model together form the prediction model.
10. An assessment system for cognitive training tasks, comprising a processor and a memory, the processor reading a computer program in the memory for performing the following operations:
acquiring a plurality of inherent attributes of the cognitive training task to be evaluated;
inputting the inherent attributes into a prediction model, and outputting a plurality of task indexes corresponding to the cognitive training task to be evaluated so as to comprehensively evaluate the cognitive training task to be evaluated through the task indexes;
wherein the predictive model is constructed by a modeling method according to any one of claims 1 to 7.
Claims (10)
1. A modeling method for cognitive task assessment, comprising the steps of:
acquiring user information of each user in a user set;
acquiring behavior reflection data and evaluation results of the users in the user set on the same cognitive training task;
acquiring a plurality of task indexes corresponding to the cognitive training task based on a diffusion project reflection theory according to the user information, behavior reflection data and evaluation results of the users in the user set;
acquiring a plurality of inherent attributes of the cognitive training task;
repeating the steps to obtain a plurality of task indexes and a plurality of inherent attributes corresponding to different cognitive training tasks until each cognitive training task in the task set obtains a plurality of corresponding task indexes and a plurality of inherent attributes;
and constructing a prediction model by utilizing a plurality of task indexes and a plurality of inherent attributes which respectively correspond to each cognitive training task in the task set.
2. A modeling method in accordance with claim 1, wherein:
the plurality of task indicators of the cognitive training task at least comprise: a time pressure index, a task difficulty index and a satisfaction index;
the plurality of intrinsic attributes of the cognitive training task include at least: the method comprises the following steps of target object number, training time, pre-calibrated target response and cognitive domain aiming.
3. The modeling method of claim 2, wherein obtaining a plurality of task indicators corresponding to the cognitive training task based on a diffusion project reflection theory according to the user information, the behavior reflection data, and the evaluation results of each user in the user set specifically comprises:
according to the user information and behavior reflection data of each user in the user set, iterative calculation is carried out based on a formula 1 and a formula 2 to obtain the time pressure index a of the cognitive training task i And a task difficulty index v i ；
Wherein, a pi Representing a time pressure value of the cognitive training task i for the user p; yp represents the user's degree of caution in responding; a is i A time pressure indicator representing a cognitive training task; p (X) pi (= 1 | theta) p Yp) represents a probability; theta p Representing a current cognitive ability of the user; x pi =1 represents training task i for user p;
and calculating a satisfaction index of the cognitive training task through a weight assignment method according to the evaluation result of each user in the user set.
4. The modeling method according to claim 2, wherein the constructing of the prediction model using the plurality of task indexes and the plurality of intrinsic attributes respectively corresponding to the cognitive training tasks in the task set specifically includes:
based on time pressure index a of each cognitive training task in the task set i Constructing a first predictor model according to a plurality of inherent attributes of the cognitive training task;
task difficulty index v based on cognitive training tasks in task set i And a plurality of the cognitive training tasksBuilding a second predictor model by the inherent attribute;
constructing a third prediction submodel based on the satisfaction index of each cognitive training task in the task set and a plurality of inherent attributes of the cognitive training tasks;
wherein the first predictor model, the second predictor model and the third predictor model together form the prediction model.
5. A modeling method as claimed in claim 4, characterized in that the first predictor model is constructed by:
constructing a first prediction submodel by an XGboost algorithm, wherein the model is expressed as follows:
wherein: m is the number of decision trees, T (x, theta M) is a certain decision tree, theta M is the number of corresponding decision trees, x is a vector formed by inherent attributes of the task, and fM (x) is a training task time pressure index a i The predicted value of (2).
6. A modeling method as claimed in claim 4, characterized in that the second predictor model is constructed by:
and constructing a second predictor model by an XGboost algorithm, wherein the model is expressed as follows:
wherein: m is the number of decision trees, T (x, theta M) is a certain decision tree, theta M is the number of corresponding decision trees, x is a vector formed by inherent attributes of the task, fM (x) is a task difficulty index v of the training task i The predicted value of (2).
7. The modeling method of claim 5, wherein the third predictor model is constructed by:
and constructing a third predictor model by an XGboost algorithm, wherein the model is expressed as follows:
wherein: m is the number of the decision trees, T (x, theta M) is a certain decision tree, theta M is the number of the corresponding decision trees, x is a vector formed by inherent attributes of the task, and fM (x) is a predicted value of a satisfaction index of the training task.
8. The evaluation method of the cognitive training task is characterized by comprising the following steps of:
acquiring a plurality of inherent attributes of the cognitive training task to be evaluated;
inputting the inherent attributes into a prediction model, and outputting a plurality of task indexes corresponding to the cognitive training task to be evaluated so as to comprehensively evaluate the cognitive training task to be evaluated through the task indexes;
wherein the predictive model is constructed by a modeling method according to any one of claims 1 to 7.
9. The evaluation method according to claim 8, wherein the inputting of the plurality of intrinsic attributes into a prediction model and the outputting of a plurality of task indexes corresponding to the cognitive training task to be evaluated specifically comprises:
inputting the plurality of inherent attributes into a first prediction submodel to obtain a time pressure index a of the cognitive training task i ；
Inputting the inherent attributes into a second predictor model to obtain a task difficulty index v of the cognitive training task i ；
Inputting the inherent attributes into a third predictor model to obtain a satisfaction index of the cognitive training task;
wherein the first predictor model, the second predictor model and the third predictor model together form the prediction model.
10. An assessment system for cognitive training tasks, comprising a processor and a memory, the processor reading a computer program in the memory for performing the following operations:
acquiring a plurality of inherent attributes of the cognitive training task to be evaluated;
inputting the inherent attributes into a prediction model, and outputting a plurality of task indexes corresponding to the cognitive training task to be evaluated so as to comprehensively evaluate the cognitive training task to be evaluated through the task indexes;
wherein the predictive model is constructed by a modeling method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211512702.9A CN115910325A (en) | 2022-11-25 | 2022-11-25 | Modeling method for cognitive task evaluation, cognitive task evaluation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211512702.9A CN115910325A (en) | 2022-11-25 | 2022-11-25 | Modeling method for cognitive task evaluation, cognitive task evaluation method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115910325A true CN115910325A (en) | 2023-04-04 |
Family
ID=86474310
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211512702.9A Pending CN115910325A (en) | 2022-11-25 | 2022-11-25 | Modeling method for cognitive task evaluation, cognitive task evaluation method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115910325A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117257304A (en) * | 2023-11-22 | 2023-12-22 | 暗物智能科技(广州)有限公司 | Cognitive ability evaluation method and device, electronic equipment and storage medium |
| CN117612712A (en) * | 2024-01-23 | 2024-02-27 | 首都医科大学宣武医院 | Method and system for detecting and improving cognition evaluation diagnosis precision |
-
2022
- 2022-11-25 CN CN202211512702.9A patent/CN115910325A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117257304A (en) * | 2023-11-22 | 2023-12-22 | 暗物智能科技(广州)有限公司 | Cognitive ability evaluation method and device, electronic equipment and storage medium |
| CN117257304B (en) * | 2023-11-22 | 2024-03-01 | 暗物智能科技(广州)有限公司 | Cognitive ability evaluation method and device, electronic equipment and storage medium |
| CN117612712A (en) * | 2024-01-23 | 2024-02-27 | 首都医科大学宣武医院 | Method and system for detecting and improving cognition evaluation diagnosis precision |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115910325A (en) | Modeling method for cognitive task evaluation, cognitive task evaluation method and system | |
| JP2022512065A (en) | Image classification model training method, image processing method and equipment | |
| CN111611390B (en) | Data processing method and device | |
| US11481648B2 (en) | Software categorization based on knowledge graph and machine learning techniques | |
| Pargent et al. | Predictive modeling with psychological panel data | |
| CN113987182A (en) | Fraud entity identification method, device and related equipment based on security intelligence | |
| CN112420125A (en) | Molecular attribute prediction method and device, intelligent equipment and terminal | |
| CN116991711A (en) | Test case generation method and device, terminal equipment and storage medium | |
| CN111475628A (en) | Session data processing method, device, computer equipment and storage medium | |
| CN110929169A (en) | Position recommendation method based on improved Canopy clustering collaborative filtering algorithm | |
| Jönsson et al. | Benchmarking k-nearest neighbour imputation with homogeneous Likert data | |
| CN113780365A (en) | Sample generation method and device | |
| CN110866000B (en) | Data quality evaluation method and device, electronic equipment and storage medium | |
| CN111062449A (en) | Prediction model training method, interestingness prediction device and storage medium | |
| CN110717787A (en) | User classification method and device | |
| CN116168403A (en) | Medical data classification model training method, classification method, device and related medium | |
| CN113627513A (en) | Training data generation method and system, electronic device and storage medium | |
| Schlegel et al. | Towards Human-Centred Explainability Benchmarks For Text Classification | |
| CN111858863B (en) | Reply recommendation method, reply recommendation device and electronic equipment | |
| CN113628080B (en) | Score prediction method and device, storage medium and electronic equipment | |
| CN110688371B (en) | Data adjustment method, device, electronic equipment and storage medium | |
| CN116167829B (en) | Multidimensional and multi-granularity user behavior analysis method | |
| CN111144973B (en) | Question ranking method and computer-readable storage medium | |
| CN113468477B (en) | Sensitive data investigation analysis method, storage medium and equipment | |
| CN115049289A (en) | Task allocation method and device based on artificial intelligence, computer equipment and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20230511 Address after: 100085 Information Center a308, building 1, Zhongguancun Software Park, Haidian District, Beijing Applicant after: Beijing Smart Spirit Technology Co.,Ltd. Address before: No. 45 Changchun Street, Xicheng District, Beijing 100000 Applicant before: XUANWU HOSPITAL OF CAPITAL MEDICAL University |
|
| TA01 | Transfer of patent application right |