Method and device for evaluating cognitive competence of ship-borne pilot, medium and product

Abstract

The invention discloses a method, equipment, medium and product for evaluating the cognitive ability of a ship-borne pilot, and relates to the field of data processing, the method comprises the following steps: obtaining multi-modal cognitive data of the ship-borne pilot in different cognitive evaluation task test processes; and obtaining a cognitive ability evaluation result of the ship-borne pilot based on the multi-modal cognitive data by adopting a random forest algorithm. According to the method, the comprehensive cognitive competence of the ship-based pilot can be comprehensively evaluated, and the ship-based flight related cognitive competence level of the pilot can be accurately reflected.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular to a method, device, medium and product for evaluating the cognitive ability of carrier-based pilots. Background Art

[0002] Carrier-based flight requires extremely high cognitive abilities from pilots. Accurately assessing pilots' ability to fly in the unique takeoff and landing scenarios is crucial for the selection and training of carrier-based pilots. Existing cognitive ability assessment methods lack a comprehensive and integrated assessment of core cognitive abilities required for carrier-based flight. Simple behavioral tests or brain imaging analysis cannot accurately reflect a pilot's level of cognitive abilities related to carrier-based flight. Therefore, there is an urgent need to develop an assessment method or system that can comprehensively reflect the comprehensive cognitive abilities of carrier-based pilots. Summary of the Invention

[0003] The purpose of this application is to provide a method, equipment, medium and product for evaluating the cognitive ability of carrier-based pilots, which can comprehensively evaluate the comprehensive cognitive ability of carrier-based pilots and accurately reflect the pilot's level of cognitive ability related to carrier-based flight.

[0004] To achieve the above objectives, this application provides the following solutions:

[0005] In a first aspect, the present application provides a method for evaluating the cognitive ability of carrier-based pilots, comprising:

[0006] Obtain multimodal cognitive data of carrier-based pilots during different cognitive assessment task tests;

[0007] The random forest algorithm is used to obtain the cognitive ability assessment results of carrier-based pilots based on multimodal cognitive data.

[0008] Optionally, a random forest algorithm is used based on multimodal cognitive data to obtain the cognitive ability assessment results of carrier-based pilots, including:

[0009] Forming feature datasets based on multimodal cognitive data;

[0010] Using the random forest algorithm, multiple sample data are extracted from the feature data set with replacement, and repeated K times to generate K initial decision trees;

[0011] Randomly extract multiple attributes from the attributes of the initial decision tree, and select the optimal attribute from the multiple attributes as the split attribute, split the initial decision tree until the stopping condition is reached, and obtain a decision tree;

[0012] The harmonic mean of the F measure is used to determine the voting weight of each decision tree, and each decision tree is regarded as a classifier;

[0013] The classification results of each classifier are weighted based on the voting weight to obtain the carrier-based pilot cognitive ability evaluation result.

[0014] Optionally, the process of selecting the best attribute from the multiple attributes includes:

[0015] Determine the Gini index of each attribute, and take the attribute with the smallest Gini index as the optimal attribute.

[0016] Optionally, the voting weight of each decision tree is expressed as:

[0017]

[0018] Where W j is the voting weight of the jth decision tree, F j is the harmonic mean of the F measure of the j-th decision tree, TP j is the number of positive samples predicted as positive by the j-th decision tree, TN j is the number of negative samples predicted as negative by the j-th decision tree, FP j is the number of negative samples predicted as positive by the j-th decision tree, FN j is the number of positive samples predicted as negative by the j-th decision tree.

[0019] Optionally, weighting the classification results of each classifier based on the voting weight to obtain the carrier-based pilot cognitive ability assessment result includes:

[0020] Weighting the classification results of each classifier based on the voting weight to obtain a classification decision result;

[0021] A classification category is obtained based on the classification decision result, and the classification category is used as a cognitive ability assessment result of the carrier-based pilot.

[0022] Optionally, the classification decision result is expressed as:

[0023]

[0024] Where H(x) is the classification decision result, Y is the classification category, K is the number of decision trees, and W j is the voting weight of the jth decision tree, h j (x) is the classification result of the jth classifier, I() is the indicator function, and x is the input data.

[0025] Optionally, cognitive assessment tasks are designed to include a multiple-target tracking task, a spatial memory task, a motion perception task, a decision-making task under uncertain information, and an impulse inhibition task;

[0026] Multimodal cognitive data includes behavioral feature data, EEG feature data, and functional near-infrared feature data.

[0027] In a second aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-mentioned method for evaluating the cognitive ability of carrier-based pilots.

[0028] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-mentioned method for evaluating the cognitive ability of carrier-based pilots.

[0029] In a fourth aspect, the present application provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of the above-mentioned method for evaluating the cognitive ability of carrier-based pilots.

[0030] According to the specific embodiments provided in this application, this application has the following technical effects:

[0031] The present application provides a method, device, medium and product for evaluating the cognitive ability of carrier-based pilots. By adopting a random forest algorithm based on multimodal cognitive data, the cognitive ability evaluation results of carrier-based pilots are obtained. The method can comprehensively evaluate the comprehensive cognitive ability of carrier-based pilots and accurately reflect the pilot's level of cognitive ability related to carrier-based flight, thereby more accurately realizing the cognitive ability of carrier-based pilots. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0033] Figure 1 A flowchart of a method for evaluating the cognitive ability of carrier-based pilots provided in one embodiment of the present application is provided;

[0034] Figure 2 A schematic diagram of a multi-target tracking paradigm provided in one embodiment of the present application;

[0035] Figure 3 A schematic diagram of a spatial memory task paradigm provided in one embodiment of the present application;

[0036] Figure 4 A schematic diagram of a motion perception ability test paradigm provided in one embodiment of the present application;

[0037] Figure 5 A schematic diagram of an uncertain information decision-making paradigm provided in an embodiment of the present application;

[0038] Figure 6 A schematic diagram of an impulse suppression task provided in one embodiment of the present application;

[0039] Figure 7 A schematic diagram of the weighted random forest algorithm flow provided in one embodiment of the present application;

[0040] Figure 8 This is a diagram illustrating an implementation framework of a method for evaluating the cognitive abilities of carrier-based pilots provided in one embodiment of the present application;

[0041] Figure 9 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0043] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0044] In an exemplary embodiment, the present application provides a method for evaluating the cognitive ability of carrier-based pilots. The method is executed by a computer device, specifically a computer device such as a terminal or a server, or a terminal and a server. In the embodiment of the present application, the method is described by taking the application of the method to the server as an example. Figure 1 As shown, the method includes:

[0045] Step 100: Acquire multimodal cognitive data of carrier-based pilots during different cognitive assessment task tests.

[0046] Step 101: Use a random forest algorithm based on multimodal cognitive data to obtain a cognitive ability assessment result of the carrier-based pilot.

[0047] In another exemplary embodiment of the present application, in order to extract the core cognitive decision-making capabilities involved in shipboard flight missions, a cognitive ability assessment method is developed through a machine learning algorithm by combining multimodal data such as behavioral, EEG, and functional near-infrared. Based on this, in this embodiment, the cognitive assessment tasks used in step 100 of the present application may include: a multi-target tracking task, a spatial memory task, a motion perception task, an uncertain information decision-making task, and an impulse inhibition task. Among them:

[0048] (1) Multi-target tracking task.

[0049] The multi-target tracking task is a behavioral task that can be implemented using the Multiple Object Tracking paradigm. Figure 2 As shown in , many random dots will be displayed on the screen at the same time. Figure 2 As shown in part (a) of the test, the objects that the subject (i.e., the carrier-based pilot to be evaluated) needs to track simultaneously will be marked on the screen. Figure 2 As shown in part (b) of , the mark disappears and all the dots start to move randomly on the screen. After a while, the movement stops, as shown in Figure 2 As shown in part (c) of the task, participants are asked to click on a white dot to confirm their tracking request. The participant's tracking accuracy serves as the performance indicator for the behavioral test. This task is divided into different difficulty levels, with adjustable difficulty indicators including the number of objects tracked, the speed of object movement, and the duration of movement. The number of objects tracked primarily assesses the participant's ability to process multiple visual target information, the speed of object movement primarily assesses the participant's ability to process visual information under varying noise conditions, and the duration of movement primarily assesses the participant's ability to maintain visual attention.

[0050] (2) Spatial memory task.

[0051] Spatial memory tasks can be divided into behavioral and ERP experimental versions, which are implemented using the delayed spatial memory paradigm to examine the subject's ability to retain spatial information. In this task, Figure 3 As shown, the experimental stimulus is 2 or 6 visual icons presented in an 8*8 checkerboard. The visual stimulus is divided into two colors, blue and orange, with a presentation time of 500ms. After the presentation, there is a time delay of 2000ms (i.e., delay 1). After that, a clue stimulus will be presented to the subject, telling the subject whether the upcoming test is related to the blue object or the orange object. The presentation time is also 500ms. After the presentation, there is still a time delay of 2000ms (i.e., delay 2). After that, a blank checkerboard will continue to be presented to the subject, with a certain position marked with a circle in it. The subject is required to judge whether the marked position is in the target stimulus of the specific color just presented. The presentation time is also 500ms.

[0052] This task examines a subject's ability to retain spatial information in working memory and to suppress irrelevant information. Research has shown that alpha waves in the EEG play a key role in the selection and storage of information. Furthermore, the intensity of alpha wave activity is associated with the encoding of spatial representations. Therefore, analyzing the intensity of alpha wave activity during the test can reflect the subject's spatial working memory representational capacity.

[0053] (3) Motion perception task.

[0054] Motion perception is the process by which individuals process information about the spatial displacement of objects. Research has shown that the strength of motion perception directly influences individual adaptation to the environment and performance in numerous sensorimotor tasks, such as driver accident frequency, athlete skill performance, and operator control performance. Time to contact (TTC) is a typical paradigm for assessing motion perception, reflecting a subject's ability to comprehensively process spatial and temporal information.

[0055] The motion perception task is also a behavioral task. Figure 4 In the TTC paradigm shown, the subjects need to predict the time it takes for a moving object to arrive at a specified location when there is partial occlusion in the moving object's path.

[0056] During the experiment, the white dot represents the moving object, moving either upward or downward. When passing through the occluded area, the subject can only estimate the dot's position based on its movement during the unoccluded period. The subject's task is to determine when the dot reaches the white horizontal line and respond by clicking the left mouse button at that moment. The experiment allows for setting three different dot movement speeds and three different occluded area lengths. The performance metric for this experiment is the error between the click time and the actual arrival time of the ball. A smaller error indicates a stronger motion estimation ability.

[0057] (4) Uncertain information decision-making tasks.

[0058] The uncertain information decision-making task is an ERP task, and the experimental materials are Gabor stripes in different directions. The Gabor stripes are randomly distributed on the screen. Figure 5 As shown in the figure, some Gabor stripes form a left-pointing arrow or a right-pointing arrow on the screen by maintaining the consistency of the direction. The direction of the Gabor stripes that make up the arrow can change within a certain range (jitter). The larger the range of change in the direction, the less obvious the shape of the arrow. After the stimulus material is presented, the subject needs to use the left and right mouse buttons to respond to the direction of the arrow. If the arrow is facing left, press the left mouse button, otherwise press the right mouse button. Figure 5As shown in the figure, the experiment included stimuli of high difficulty level (jitter = 60 degrees) and low difficulty level (jitter = 45 degrees), which were presented in a random order. The presentation time of the stimuli was 500ms, and the interstimulus interval (ISI) was 2000ms. Figure 5 Part (a) shows a low-difficulty picture. Figure 5 Part (b) shows a difficult picture.

[0059] In the uncertain information decision-making task, the material is presented briefly, focusing on examining the subject's ability to integrate perceptual information. This ability plays a crucial role in making decisions based on ambiguous information in noisy environments. Research has shown that the P1 / N1 components of the ERP are closely related to early visual perceptual information processing, with the P1 primarily related to information processing and the N1 primarily to feature discrimination. Therefore, in this test, it is appropriate to focus on examining differences in the P1 / N1 components across different populations.

[0060] (5) Impulse inhibition task.

[0061] The impulse inhibition task is a near-infrared optical imaging experimental task. The experimental presentation of pictures is divided into a cue phase and a response phase, such as Figure 6 As shown in the figure, during the cue phase, a white cross-shaped fixation point will appear in the center of a black background screen. At the same time, there will be white arrow clue stimuli above and below the fixation point, pointing to the left or right, indicating that the left or right mouse button needs to be pressed in the response phase. After the cue disappears, a brief blank screen will appear, followed by the response phase stimulus. The response phase stimuli are divided into two types: red dots and green dots. When the red dot appears (NOGO stimulus), the subject does not need to make any response. When the green dot appears (GO stimulus), the subject needs to click the corresponding mouse button according to the prompt of the cue phase.

[0062] In all of the cognitive assessment tasks described above, a computer records the subject's reaction time and accuracy using a mouse or keyboard. For the impulse inhibition task, a functional near-infrared data acquisition device can be used. For the motion perception and spatial memory tasks, EEG signals can be collected synchronously using an EEG acquisition device.

[0063] Furthermore, for each of the above tasks, we can obtain a subject's temporal error in the motion perception task, accuracy in the multi-target tracking task, reaction time and accuracy in the impulse inhibition task, the degree of oxygenated hemoglobin signal activation in the prefrontal region during the impulse inhibition task, accuracy and reaction time in the motor decision-making task, the latency and amplitude of the N1 component in the occipital region during the motor decision-making task, accuracy and reaction time in the spatial memory task, and alpha band (8-12 Hz) power intensity in the prefrontal region during the spatial memory task. Among them, the temporal error in the motion perception task, accuracy in the multi-target tracking task, reaction time and accuracy in the impulse inhibition task, accuracy and reaction time in the motor decision-making task, and accuracy and reaction time in the spatial memory task are behavioral data, while the latency and amplitude of the N1 component in the occipital region during the motor decision-making task and alpha band (8-12 Hz) power intensity in the prefrontal region during the spatial memory task are EEG data, and the degree of oxygenated hemoglobin signal activation in the prefrontal region during the impulse inhibition task is functional near-infrared data.

[0064] In another exemplary embodiment of the present application, carrier-based pilots are divided into three levels: A, B, and C, based on their actual assessment scores. The top third are ranked A, the middle third are ranked B, and the bottom third are ranked C. 30 individuals from each of the A, B, and C levels are selected to complete the cognitive assessment task described above, obtaining multimodal cognitive data for a total of 90 individuals. The acquired multimodal cognitive data may include behavioral characteristic data (reaction time, accuracy, etc.), EEG characteristic data (ERP amplitude, latency, etc.), and functional near-infrared characteristic data (oxygenated hemoglobin activation value, etc.). Based on this, in step 101 above, a random forest algorithm can be used to construct a classification and discriminant model for the multimodal cognitive data collected from the 90 individuals. The purpose of the constructed classification and discriminant model is to, after inputting all relevant assessment task features for a specific carrier-based pilot, classify the pilot's cognitive ability into one of the following levels: A, B, or C.

[0065] When implementing the above step 101 provided in this application, the following steps may be included:

[0066] Step 1: Form a feature dataset based on multimodal cognitive data.

[0067] For example, according to the above description, the cognitive ability features used by the classification and discrimination model include behavioral features, EEG features, and functional near-infrared features, thus obtaining a feature dataset Q containing 90 samples. The dimension of the feature dataset Q is 90×N, where N is the total number of features.

[0068] Step 2: Use the random forest algorithm to extract multiple sample data with replacement from the feature dataset, repeating this process K times to generate K initial decision trees. For example, from all samples in the feature dataset Q, extract n samples with replacement, repeat this process K times, and generate K initial decision trees.

[0069] Step 3: Randomly extract multiple attributes from the attributes of the initial decision tree, and select the optimal attribute from the multiple attributes as the splitting attribute. Split the initial decision tree until the stopping condition is reached to obtain a decision tree.

[0070] For example, when splitting each node of the initial decision tree, m attributes are randomly selected from all M attributes of the initial decision tree, and then the optimal attribute is selected from the m attributes as the splitting attribute. Here, m is the largest positive integer less than or equal to log2(M)+1.

[0071] For the selection of split attributes, the Gini index is used as the division method. When the feature data in the feature data set Q is used as a random variable, it is assumed that the proportion of the i-th class sample in the random variable D of the current feature data set Q is p i , then the Gini value of the random variable D is:

[0072]

[0073] Where Gini(D) is the Gini value of the random variable D, which ranges from 0 to 1. |y| is the number of class samples, where the classes are A, B, and C.

[0074] Use the test attribute a to divide the random variable D into two variables, obtaining variables D1 and D2. The Gini value of the test attribute a is:

[0075]

[0076] Where Gini(D,a) is the Gini value of the test attribute a, Gini(D1) is the Gini value of the variable D1, and Gini(D2) is the Gini value of the variable D2.

[0077] Based on the above processing, the candidate attribute set A = a1, a2, ..., am is obtained. Then, the attribute that minimizes the Gini index after partitioning in the candidate attribute set A is selected as the optimal attribute, that is,

[0078] a * =argmin(Gini_index(D,a)).

[0079] Where a * represents the optimal attribute, and Gini_index(D, a) represents the Gini index.

[0080] Each initial decision tree continues to split until the stopping condition is reached, for example, the maximum tree depth is reached or the number of samples at each leaf node is less than the set threshold, and a decision tree is obtained to avoid the problem of insufficient information gain after splitting.

[0081] Step 4: Use the harmonic mean of the F-measure to determine the voting weight of each decision tree, and treat each decision tree as a classifier. The calculation formula of the harmonic mean of the F-measure is:

[0082]

[0083] Where F-measure is the harmonic mean of the F-measure, and recall is the recall rate. Precision is the difference between the measured value and the true value. TP is the number of positive examples predicted as positive by the decision tree, and TN is the number of negative examples predicted as negative by the decision tree. FP is the number of negative examples predicted as positive by the decision tree, i.e., the number of false positives. FN is the number of positive examples predicted as negative by the decision tree, i.e., the number of false negatives.

[0084] Furthermore, after obtaining the TP, FP, FN, and TN of each decision tree, the voting weight of each decision tree can be calculated as follows:

[0085]

[0086] Where W j is the voting weight of the jth decision tree, F j is the harmonic mean of the F measure of the j-th decision tree, TP j is the number of positive samples predicted as positive by the j-th decision tree, TN j is the number of negative samples predicted as negative by the j-th decision tree, FP j is the number of negative samples predicted as positive by the j-th decision tree, FN j is the number of positive samples predicted as negative by the j-th decision tree.

[0087] Step 5: Weight the classification results of each classifier based on the voting weight to obtain the evaluation results of the cognitive ability of carrier-based pilots.

[0088] Furthermore, after each decision tree is used as a classifier, the final classification result depends on the weighted classification results of each classifier in the classification discriminant model. Based on this, the implementation process of step 5 may include:

[0089] Step 51: Weight the classification results of each classifier based on the voting weight to obtain the classification decision result. For example, the classification result of the j-th classifier (corresponding to the j-th decision tree) is h j(x), the classification weight of the j-th classifier is W j , the final classification decision result is:

[0090] The classification decision result is expressed as:

[0091]

[0092] Where H(x) is the classification decision, Y is the classification category (A, B, or C), I() is the indicator function, which is 1 if the independent variable is true and 0 if the independent variable is false. x is the input data.

[0093] Based on the above description, the entire process of determining the classification decision result by the weighted random forest algorithm provided in this embodiment is as follows: Figure 7 shown.

[0094] Step 52: Obtain a classification category based on the classification decision result, and use the classification category as the evaluation result of the cognitive ability of the carrier-based pilot.

[0095] Based on the above description, the overall implementation process of the carrier-based pilot cognitive ability assessment method provided by this application is as follows: Figure 8 As shown, it simultaneously incorporates behavioral data, EEG data, and functional near-infrared data into an analysis model (i.e., a classification and discrimination model), capable of categorizing carrier-based pilots into three quantitative levels (the specific levels can be adjusted based on actual needs). Furthermore, the cognitive abilities assessed are closely related to carrier-based flight missions, truly reflecting the flight performance of carrier-based pilots, enabling more accurate assessments and comprehensively reflecting their comprehensive cognitive abilities.

[0096] In another exemplary embodiment of the present application, the above-mentioned carrier-based pilot cognitive ability assessment method can be implemented in a smart terminal and combined with a mouse, keyboard, functional near-infrared data acquisition device, EEG data acquisition device, etc. to form a carrier-based pilot cognitive ability system. The smart terminal can be, but is not limited to, various desktop computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices.

[0097] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 9As shown. The computer device includes a processor, a memory, an input / output interface (I / O) and a communication interface. The processor, memory and input / output interface are connected via a system bus, and the communication interface is connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store carrier-based pilot cognitive ability data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a carrier-based pilot cognitive ability assessment method is implemented.

[0098] Those skilled in the art will understand that Figure 9 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present application and does not constitute a limitation on the computer device to which the solution of the present application is applied. A specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement. In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps of the above-mentioned method embodiments when executing the computer program.

[0099] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0100] In an exemplary embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0101] It should be noted that the subject information (including but not limited to subject device information, subject personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the subjects or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0102] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0103] The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may include, but are not limited to, general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic units, data processing logic units based on quantum computing, and the like.

[0104] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0105] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the method and core concept of this application. At the same time, for those skilled in the art, based on the concept of this application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for evaluating the cognitive ability of carrier-based pilots, characterized in that: include: Obtain multimodal cognitive data of carrier-based pilots during different cognitive assessment task tests; The random forest algorithm is used to obtain the cognitive ability assessment results of carrier-based pilots based on multimodal cognitive data.

2. The method for evaluating the cognitive ability of carrier-based pilots according to claim 1, characterized in that: The random forest algorithm is used based on multimodal cognitive data to obtain the cognitive ability assessment results of carrier-based pilots, including: Forming feature datasets based on multimodal cognitive data; Using the random forest algorithm, multiple sample data are extracted from the feature data set with replacement, and repeated K times to generate K initial decision trees; Randomly extract multiple attributes from the attributes of the initial decision tree, and select the optimal attribute from the multiple attributes as the split attribute, split the initial decision tree until the stopping condition is reached, and obtain a decision tree; The harmonic mean of the F measure is used to determine the voting weight of each decision tree, and each decision tree is regarded as a classifier; The classification results of each classifier are weighted based on the voting weight to obtain the carrier-based pilot cognitive ability evaluation result.

3. The method for evaluating the cognitive ability of carrier-based pilots according to claim 2, characterized in that: The process of selecting the best attribute from multiple attributes includes: Determine the Gini index of each attribute, and take the attribute with the smallest Gini index as the optimal attribute.

4. The method for evaluating the cognitive ability of carrier-based pilots according to claim 2, characterized in that: The voting weight of each decision tree is expressed as: Where W j is the voting weight of the jth decision tree, F j is the harmonic mean of the F measure of the j-th decision tree, TP j is the number of positive samples predicted as positive by the j-th decision tree, TN j is the number of negative samples predicted as negative by the j-th decision tree, FP j is the number of negative samples predicted as positive by the j-th decision tree, FN j is the number of positive samples predicted as negative by the j-th decision tree.

5. The method for evaluating the cognitive ability of carrier-based pilots according to claim 2, characterized in that: The classification results of each classifier are weighted based on the voting weight to obtain the carrier-based pilot cognitive ability assessment result, including: Weighting the classification results of each classifier based on the voting weight to obtain a classification decision result; A classification category is obtained based on the classification decision result, and the classification category is used as a cognitive ability assessment result of the carrier-based pilot.

6. The method for evaluating the cognitive ability of carrier-based pilots according to claim 5, characterized in that: The classification decision result is expressed as: Where H(x) is the classification decision result, Y is the classification category, K is the number of decision trees, and W j is the voting weight of the jth decision tree, h j (x) is the classification result of the jth classifier, I() is the indicator function, and x is the input data.

7. The method for evaluating the cognitive ability of carrier-based pilots according to claim 1, characterized in that: The cognitive assessment tasks were designed to include a multiple-target tracking task, a spatial memory task, a motion perception task, a decision-making task under uncertain information, and an impulse inhibition task; Multimodal cognitive data includes behavioral feature data, EEG feature data, and functional near-infrared feature data.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for evaluating the cognitive ability of carrier-based pilots according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for evaluating the cognitive ability of carrier-based pilots according to any one of claims 1 to 7 is implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method for evaluating the cognitive ability of carrier-based pilots according to any one of claims 1 to 7 is implemented.

Images