CN111700611A

CN111700611A - Method for assessing insight capabilities and related device

Info

Publication number: CN111700611A
Application number: CN202010552504.XA
Authority: CN
Inventors: 于成龙; 黄艳; 叶文梓
Original assignee: Shenzhen Institute of Advanced Technology of CAS
Current assignee: Shenzhen Institute of Advanced Technology of CAS
Priority date: 2020-06-16
Filing date: 2020-06-16
Publication date: 2020-09-25
Anticipated expiration: 2040-06-16
Also published as: CN111700611B

Abstract

The embodiment of the invention discloses a method for evaluating the comprehension capacity and related equipment, which are used for respectively detecting the comprehension capacity of a tester according to English word data and a graphic database to obtain a first detection parameter and a second detection parameter, wherein the first detection parameter and the second detection parameter are associated with the comprehension capacity; and then, according to the first detection parameter and the second detection parameter, respectively obtaining first and second demonstration evaluation information of the tester, and comprehensively evaluating the demonstration ability of the tester according to the first and second demonstration evaluation information. Particularly, due to the fact that the image comprehension detection branch is less affected by the vocabulary amount, age, culture background and the like of the testee, the evaluation method provided by the embodiment of the invention can adapt to comprehension detection requirements of the testees in different age groups and education levels, and the comprehension is evaluated by combining English word comprehension detection and image comprehension detection, so that the evaluation accuracy is higher.

Description

Method for assessing insight capabilities and related device

Technical Field

The invention relates to the technical field of internet, in particular to a method for evaluating the comprehension ability and related equipment.

Background

The existing solutions for detecting the inroads are mainly CRAT (Compound remove Associates task) and word guessing method (anticancer procedure). The CRAT scheme first presents several unrelated words to the subject, asking the subject to provide a new word that can link the several unrelated words (Kizilirmak et al, 2019). The word guessing method is to show a string of English words in disordered alphabetical order to the testee and to require the testee to give correct English words. After solving the problem, the subject is asked to tell whether the solution to the problem is emergent ("insight") or derived via analytical reasoning ("analytical") (Oh et al, 2020). Both schemes require a tester to have a certain mastery basis on English or Chinese words, and the comprehension ability and the analysis ability achievement are evaluated by detecting the time and the accuracy of the tester for solving the problems.

The existing scheme is based on association and arrangement tasks of words, and a testee needs to have a certain Chinese and English word basis. The popularization of the method is not strong for the testees of different ages and culture backgrounds, the detection result is too dependent on the mastering capability of the testees on words, and if the testees do not master some English words, the accuracy of the detection result is very limited.

Disclosure of Invention

The embodiment of the invention provides a comprehension ability evaluation method and related equipment, which can effectively improve the comprehension ability evaluation accuracy.

In a first aspect, an embodiment of the present invention provides a method for assessing a comprehension ability, including:

detecting the comprehension ability of a tester according to an English word database to obtain a first detection parameter;

detecting the comprehension ability of the tester according to a graph database, and acquiring a second detection parameter, wherein the first detection parameter and the second detection parameter are associated with the comprehension ability;

and respectively obtaining first and second insight evaluation information of the tester according to the first and second detection parameters.

Optionally, the first detection parameter comprises a first electroencephalogram index, a first reaction time, a first subjective evaluation, a first accuracy, and a first proportion of an expression;

according to the English word database, the comprehension ability of the tester is detected, and a first detection parameter is obtained, wherein the method comprises the following steps:

determining a test word for capability detection according to the English word database;

outputting the test words with disordered alphabetical order to the tester;

acquiring a first electroencephalogram index and a first reaction time of a word recovery process of the tester, and a first subjective evaluation of the way of recovering the word by the tester, wherein the first electroencephalogram index is related to a comprehension capacity, and the first subjective evaluation comprises a comprehension formula and an analysis formula;

obtaining a first ratio of a first correctness of the tester recovering the test words and a first proportion of the expressions of the recovery words.

Optionally, performing a comprehension ability detection on the tester according to the english word database, and acquiring a first detection parameter, including:

and determining an English word database for detecting the comprehension ability of the tester from English word databases with different difficulty levels according to the age and education condition of the tester.

after the comprehension ability is detected, acquiring the recognition result of the tester on the test word;

if the identification result is identification failure, eliminating a first electroencephalogram index, a first response time and a first subjective evaluation of the test word;

and acquiring the first accuracy and the first ratio according to the removed data.

Optionally, the second detection parameter includes a second electroencephalogram index, a second reaction time, a second subjective evaluation, a second accuracy, and a second proportion of an insight;

detecting the comprehension ability of the tester according to the graph database, and acquiring a second detection parameter, wherein the method comprises the following steps:

determining a test image for capability detection according to the graph database, wherein the graph database comprises a plurality of images, and the images are composed of a plurality of basic graphs;

outputting the test image with the basic image position disturbed to the tester;

acquiring a second electroencephalogram index in the process of recovering the test image by the tester, a second reaction time for recovering the test image, and a second subjective evaluation of the mode of recovering the test image by the tester, wherein the second electroencephalogram index is related to the comprehension capacity, and the second subjective evaluation comprises a comprehension formula and an analysis formula;

and acquiring a second ratio of the second correctness rate of the tester for recovering the test image and the second proportion of the expressions in the word recovering mode.

Optionally, obtaining first and second demonstration evaluation information of the tester according to the first and second detection parameters, respectively, includes:

calculating a relative Euclidean distance between the first detection parameter and the parameters of the first upper limit and the first lower limit of the achievement force according to the first detection parameter, the parameters of the first upper limit of the achievement force and the parameters of the first lower limit of the achievement force, and determining first achievement force evaluation information of the tester according to the relative Euclidean distance;

and calculating a relative Euclidean distance between the second detection parameter and the parameters of the second upper limit and the second lower limit of the achievement force according to the second detection parameter, the parameters of the second upper limit of the achievement force and the parameters of the second lower limit of the achievement force, and determining second achievement force evaluation information of the tester according to the relative Euclidean distance.

clustering the first detection parameters and the second detection parameters of a plurality of testers respectively to obtain a plurality of first sets and a plurality of second sets;

calculating relative Euclidean distances between the cluster centers of the first set and the parameters of the first upper limit of the learning force and the parameters of the first lower limit of the learning force according to the cluster centers of the first set, the parameters of the first upper limit of the learning force and the parameters of the first lower limit of the learning force, and determining first learning force evaluation information of the testers in the first set according to the relative Euclidean distances;

and calculating relative Euclidean distances between the cluster centers of the second set and the parameters of the second upper limit and the second lower limit of the comprehension force according to the cluster centers of the second set, the parameters of the second upper limit and the parameters of the second lower limit of the comprehension force, and determining second comprehension force evaluation information of the testers in the second set according to the relative Euclidean distances.

In a second aspect, an embodiment of the present invention provides a device for evaluating comprehension ability, including:

the first detection module is used for detecting the comprehension ability of the tester according to the English word database to obtain a first detection parameter;

the second detection module is used for detecting the comprehension ability of the tester according to the graph database and acquiring a second detection parameter, wherein the first detection parameter and the second detection parameter are associated with the comprehension ability;

and the evaluation module is used for respectively obtaining first insight evaluation information and second insight evaluation information of the tester according to the first detection parameter and the second detection parameter.

In a third aspect, an embodiment of the present invention provides a device for assessing comprehension ability, including: a processor and a memory;

the processor is coupled to the memory, wherein the memory is configured to store program code, and the processor is configured to invoke the program code to perform the insight capability assessment method of the first aspect.

In a fourth aspect, embodiments of the present invention provide a computer storage medium storing a computer program comprising program instructions which, when executed by a processor, perform the insight capability assessment method according to the first aspect.

The embodiment of the invention respectively detects the comprehension ability of a tester according to English word data and a graphic database to obtain a first detection parameter and a second detection parameter, wherein the first detection parameter and the second detection parameter are associated with the comprehension ability; and then, according to the first detection parameter and the second detection parameter, respectively obtaining first and second demonstration evaluation information of the tester, and comprehensively evaluating the demonstration ability of the tester according to the first and second demonstration evaluation information. Particularly, due to the fact that the image comprehension detection branch is less affected by the vocabulary amount, age, culture background and the like of the testee, the evaluation method provided by the embodiment of the invention can adapt to comprehension detection requirements of the testees in different age groups and education levels, and the comprehension is evaluated by combining English word comprehension detection and image comprehension detection, so that the evaluation accuracy is higher.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

FIG. 1 is a schematic flow chart of a method for assessing insight provided by an embodiment of the present invention;

FIGS. 2a and 2b are schematic diagrams of an understanding ability evaluation method provided by an embodiment of the present invention;

FIG. 3 is a schematic structural diagram of an exemplary device for assessing throughput of a computer system according to the present invention;

fig. 4 is a schematic structural diagram of a device for evaluating throughput provided by an embodiment of the present invention.

Detailed Description

The technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

It should be understood that the terms "first," "second," and the like in the description and claims of this application and in the drawings are used for distinguishing between different objects and not for describing a particular order. Furthermore, the terms "include" and "have," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements listed, but may alternatively include other steps or elements not listed, or inherent to such process, method, article, or apparatus.

Reference in the specification to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the invention. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is explicitly and implicitly understood by the person skilled in the art that the described embodiments of the invention can be combined with other embodiments.

Creativity is a high-level mental activity, which does not depend on a single mental activity or activation of a brain region, but rather is a complex cognitive process involving multiple brain regions in a coordinated manner. The detection of creative thinking often needs to be done in a specific cognitive activity. For example, in the task of solving problems, creative thinking is embodied as insights, usually expressed as an idea, solution to a problem, etc., that appear suddenly in consciousness in an "all or nothing" manner, also referred to as "Ah ha time". Since a sudden emergence is generally a completely new idea or solution, the insight is considered to be a manifestation of the ability to innovate. Corresponding to this is the ability to analyze, emphasizing the progressive analysis of the problem and the solution proposed, with the participation of awareness. Analysis capabilities often suggest solving problems with existing solutions, or a combination of existing solutions. The insight tends to create a new solution in an "all or nothing" manner. Furthermore, unlike analytical capabilities, the insight is generally accompanied by a stronger rewarding feedback. In the creative thinking culture process, cognitive activities such as the insight and the like often form a benign positive feedback process, namely, the spiritual pleasure brought by the insight can stimulate students to further think to strengthen the process.

In the prior art, in order to realize the detection of the comprehension capacity, the CRAT method and the word guessing method can be adopted for detection, however, the accuracy of the detection results of the two methods is determined to a certain extent by the vocabulary mastering basis of a tester, so that the detection accuracy of the comprehension capacity of the two methods is not high. Therefore, the application provides an evaluation method for the comprehension ability, which can meet the detection requirements of testers with different ages and culture backgrounds and has high evaluation accuracy.

Fig. 1 is a schematic flow chart illustrating an exemplary embodiment of a method for evaluating comprehension ability of a computer; the comprehension capacity evaluation method comprises the following steps:

101. detecting the comprehension ability of a tester according to an English word database to obtain a first detection parameter;

specifically, the comprehension capacity detection is performed by using an English word database, namely a word guessing method, generally, during detection, the letter sequence of words in the database is disordered, a tester performs letter recovery, and a relevant parameter of the tester in the letter recovery process is obtained as a first detection parameter, and the first detection parameter is associated with the comprehension capacity.

102. Detecting the comprehension ability of the tester according to the graph database to obtain a second detection parameter;

specifically, the method for detecting the mental capacity of the tester according to the graph database is a nonverbal graph detection paradigm, the graph detection paradigm requires the tester to combine several basic geometric graphs into a common graph, the recognition and understanding abilities of the testers including children and the old are strong, and the limitation of the vocabulary of the testers on the accuracy of detection results can be effectively avoided. Similarly, the related parameters of the tester in the graph combination process are obtained as second detection parameters, and the second detection parameters are associated with the comprehension capability.

103. And respectively obtaining first and second comprehension evaluation information of the tester according to the first and second detection parameters.

Specifically, the first demonstration evaluation information of the tester can be obtained according to the first detection parameter, and the second demonstration evaluation information of the tester can be obtained according to the second detection parameter, wherein both the first demonstration evaluation information and the second demonstration evaluation information can be demonstration ability scores, or demonstration ability grades/gears and the like show the demonstration information of the demonstration ability.

The method of the figure 1 can be used for obtaining the first and second demonstration evaluation information of the tester, the graph detection method is less influenced by the vocabulary, the age, the culture background and the like, the development condition of the demonstration ability of people at different ages can be favorably comprehensively investigated by flexibly comparing the results of the graph detection and word guessing method, the influence of the vocabulary mastering degree of the tester on the accuracy of the demonstration ability evaluation result can be effectively overcome, the accuracy of the demonstration ability evaluation can be guaranteed, and the method is suitable for the demonstration ability detection requirements of the testers at different ages and education levels.

In particular, the execution order of step 101 and step 102 is not limited, and step 101 may be executed first, and then step 102 may be executed. Step 102 may be performed first, and then step 101 may be performed.

In one possible embodiment, the first detection parameter includes a first brain electrical indicator, a first reaction time, a first subjective evaluation, a first accuracy, and a first percentage of an insight; step 101 comprises:

1011. determining a test word for capability detection according to an English word database;

specifically, an english word database for detecting the comprehension ability is established in advance, wherein words in the english word database may include words of primary english, middle english, and advanced english, that is, english words with different difficulties are included in the english word database. When the comprehension ability of the tester is detected according to the English word database, a certain number of words can be randomly extracted from the English word database to serve as test words.

1012. Outputting the test words with disordered alphabetical order to a tester;

specifically, at the beginning of the test, according to the test words selected in step 1011, each test word is output to the tester in sequence after the alphabetical order is broken. The alphabetic sequence can be randomly disturbed to avoid the influence on the accuracy of the comprehension capability detection caused by the fact that a tester knows the rule of the disordered alphabetic sequence; and requires the tester to enter the correct english word. For example, at the computer side or the mobile phone side, "goody" is output to the tester, and the tester is required to input a correct english word, and the correct word corresponding to "goody" is "today".

1013. Acquiring a first electroencephalogram index and a first reaction time of a word recovery process of a tester, and a first subjective evaluation of the tester on a word recovery mode, wherein the first electroencephalogram index is related to a comprehension capacity, and the first subjective evaluation comprises a comprehension formula and an analysis formula;

specifically, because apparent EEG signal changes when the wu takes place, consequently, when the tester is done the wu kui ability and is detected, the EEG signal of tester can be monitored simultaneously to obtain first EEG index. In this embodiment, the first electroencephalogram index is related to the comprehension capability, and includes energy of a gamma frequency band of a prefrontal lobe and energy of an alpha frequency band of a temporal lobe. The monitoring signal is subjected to noise reduction and filtering, and interference signals such as eye movement and the like are removed, so that a relatively pure electroencephalogram signal can be obtained. Then, the dynamic change of the insight and creative thinking of the testee can be objectively analyzed by calculating the energy change of the relevant gamma and alpha frequency bands.

In addition, in the process of recovering the words by the tester, the time length of the answer input by the tester is monitored, namely, the test words with disordered alphabetical order are output to the tester in a statistical manner, the time length until the answer is input by the tester is taken as the first reaction time of the tester about the test words. Generally, the reaction time of a insight solution is generally earlier than the reaction time of an analytical solution.

Finally, after the tester inputs the answer, the tester is required to give a first subjective evaluation to the mode of recovering the word, and the first subjective evaluation comprises a contemplation formula and an analysis formula; the choice is made whether the answer appears suddenly in the brain in an "all or nothing" manner, i.e. a dread answer, or by stepwise inferential analysis, i.e. an analytic answer. For example, "insight" and "analytic" can be displayed on the computer or the mobile phone for the tester to select.

1014. A first ratio of a first correctness of a tester recovering a test word and a first proportion of a realization in a recovery word mode is obtained.

Specifically, for all the test words selected in step 1011, the frequency of appearance of the tentative answers, i.e., the proportion of the tentative answers, is calculated as the first proportion used to measure whether the tester is more inclined to solve the problem analytically or tentatively. In addition, the correctness of the tester in the test is counted as a first correctness.

In one possible embodiment, in order to improve the accuracy of the comprehension ability detection, it is necessary to determine test words of different difficulty levels according to different situations of the testers, and the step 101 further includes:

1015. and determining an English word database for detecting the comprehension ability of the tester from English word databases with different difficulty levels according to the age and education condition of the tester.

Specifically, the ages are 5-12 years old, 13-18 years old, 19-40 years old, 41-60 years old and more than 61 years old; the education condition is below primary school, junior middle school, high school, specialty, this department, and more than master. English word databases with different difficulties are established in advance, words in the databases are divided into 5 grades according to indexes such as word common degree, character string length and the like, namely, English word databases with 5 different difficulties are respectively established from low to high according to the difficulty degree, and are respectively a first-grade difficulty English word database, a second-grade difficulty English word database, a third-grade difficulty English word database, a fourth-grade difficulty English word database and a fifth-grade difficulty English word database. In actual selection, word data with different difficulties are called according to the conditions of testers, such as age, education degree and the like. For example, for a 38-year-old tester a, the education condition is elementary school, and in order to improve the accuracy of the detection result of the comprehension ability, an english word database originally suitable for 13 to 18 years old and having a high school calendar (for example, the english word database is a second-level difficulty english word database) may be selected to detect the comprehension ability of the tester a.

On the other hand, when the English word database corresponding to the difficulty is selected according to the age and the education condition of the tester, weighting (the respective weights of the age and the education condition can be adjusted as required) can be carried out according to the age and the education condition to obtain an identification score, different ages correspond to different scores (the specific score can be set as required), different education conditions also correspond to different scores (the specific score can be set as required), and the corresponding English word database can be found according to the identification score and the corresponding table of the English word database. For example, a pre-established english word database correspondence table is shown in table 1.

TABLE 1

Assuming that the weight of age is 0.4 and the weight of education is 0.6, the identification score S is age N × 0.4+ education J × 0.6, assuming that the age-score correspondence table is shown in table 2 and the education-score correspondence table is shown in table 3.

TABLE 2

Age group/year	Score of
		5-12	5
13-18	10
		19-40	20
41-60	10
		Above 61	5

TABLE 3

Educational situation	Score of
		Below primary school	1
Middle school	5
		High school	10
Special section	15
		This section	20
Greater than Master	25

In addition, after the english word database is determined, a certain number of english words can be randomly selected from the english word database as the test words of the tester, and different numbers of test words can be determined according to the age and education condition of the tester, for example, the number of test words is divided into 5 grades, i.e., 5, 10, 15, 20 and 25 in the order of difficulty level, and the different numbers of test words correspond to the english word database with different difficulty levels.

For example, when testing, the first difficulty english word database selects 5 test words.

The second difficulty English word database selects 10 test words.

The third difficulty English word database selects 15 test words.

The fourth difficulty English word database selects 20 test words.

The fifth difficulty English word database selects 25 test words.

Different from the step 1011, test words with different numbers of words and different difficulties need to be selected from the single english word database with large word quantity according to the condition of the tester, so that the speed of testing the words is determined to be slow, and the experience of the tester is poor. In combination with step 1011 and step 1015, since the english word databases with different difficulties are pre-established in step 1015, when selecting a test word, only the english word database with the corresponding difficulty level needs to be determined according to the age and education condition of the tester, and then a certain number of test words are selected from the determined english word database, in contrast, the speed of determining the test word is faster.

In one possible embodiment, to further improve the accuracy of the insight capability detection result, step 101 further comprises:

s1, obtaining the recognition result of the tester to the testing word after the detection of the ability of the realization;

specifically, after the tester completes the test word determined in step 1011, the tester is required to recognize the test word used in the above-mentioned detection process again, so as to obtain the recognition result of the tester for the test word. The test word list may be displayed again at the computer or mobile terminal, and options are provided for the tester to choose whether to recognize, for example, "know" or "not know" is displayed. Or a paper list can be provided and selected by a tester.

S2, if the identification result is identification failure, eliminating a first electroencephalogram index, a first response time and a first subjective evaluation of the test word;

specifically, if the recognition result is a recognition failure, for example, the tester checks that "no knowledge" is found, the first electroencephalogram index, the first reaction time and the first subjective evaluation of the test word are rejected, and the test word does not participate in the calculation of other first detection parameters.

And S3, acquiring a first accuracy and a first proportion according to the data after being eliminated.

Specifically, the calculation of the first accuracy and the first proportion is performed according to the data left after the data of the test word with failed recognition is eliminated. For example, if there are 10 total test words, and 2 word testers fail to recognize, after the data corresponding to the two words are eliminated, the first correctness rate and the first proportion corresponding to the remaining 8 test words are calculated as the final first correctness rate and the final first proportion of the tester.

Through after the test, carry out the word once more and discern to reject the test data that the test word that the tester does not know corresponds, promote the degree of accuracy that the ability detected of feeling immediately.

In particular, the word resolution test can be performed between the first period (for example, the first two weeks) before the formal test starts, so that the formal test result is not affected by the vocabulary of the tester and other factors, and the english word database called during the formal test can be specifically adjusted. And the interference effect of the preliminary experiment on the normal detection can be reduced to the maximum extent by carrying out formal tests later. For example, when the first accuracy is lower than 50% in the preliminary test, the word difficulty may be appropriately adjusted to be lower, and the word in the english word database with the lower level of difficulty is used as the test word in the formal test.

In one possible embodiment, the second detection parameter includes a second electroencephalogram index, a second reaction time, a second subjective evaluation, a second accuracy, and a second percentage of the insight; step 102 comprises:

1021. determining a test image for capability detection according to a graphic database, wherein the graphic database comprises a plurality of images, and the images are composed of a plurality of basic graphics;

specifically, each image in the graph database is formed by splicing a plurality of basic graphs, and the basic graphs comprise circles, triangles, squares, ladders and the like. Reference is made to fig. 2a and 2b, wherein the image of fig. 2b is stitched from the five basic patterns of fig. 2 a. Similar to the english word database, a graph database for detecting the difficulty level needs to be established in advance, where the graphs in the graph database may include primary difficulty graphs, intermediate difficulty graphs, and advanced difficulty graphs, that is, the graph database includes graphs with different spelling difficulties, the number of basic graphs of each primary difficulty graph is smaller than that of basic graphs of each intermediate difficulty graph, and the number of basic graphs of each intermediate difficulty graph is smaller than that of basic graphs of each advanced difficulty graph, for example, the number of basic graphs of one primary difficulty graph may be 3 to 5, the number of basic graphs of one primary difficulty graph may be 6 to 10, and the number of basic graphs of one advanced difficulty graph may be 11 to 15. When the comprehension ability of the tester is detected according to the graphic database, a certain number of images can be randomly extracted from the graphic database to serve as test images.

1022. Outputting the test image with the basic image position disturbed to a tester;

specifically, the test image with the disordered basic pattern position is output to the tester, for example, fig. 2a may be displayed on a computer terminal or a mobile phone terminal, and the tester is required to sequentially select basic patterns through a mouse to combine the basic patterns into a correct complex pattern, where the correct pattern is fig. 2 b. The position of the basic graph can be randomly disturbed, so that the principle of disturbing the position of the basic graph is known by testers, and the accuracy of the detection of the understanding ability is influenced.

1023. Acquiring a second electroencephalogram index in the process of recovering the test image by the tester, a second reaction time for recovering the test image, and a second subjective evaluation of the mode of recovering the test image by the tester, wherein the second electroencephalogram index is related to the comprehension capacity, and the second subjective evaluation comprises a comprehension formula and an analysis formula;

specifically, similar to the word guessing method, when the mental power is detected according to the image of the graphic database, the electroencephalogram signal of the tester is monitored in real time to obtain a second electroencephalogram index; the second EEG index can be the energy of the gamma frequency band of the prefrontal lobe and the energy of the alpha frequency band of the temporal lobe. Meanwhile, the time spent by the tester in recovering the test image is also required to be acquired, and the time is taken as second reaction time, namely, the time from outputting the test image at the disordered graphic position to completing the image splicing by the tester is counted, and the time from the tester clicking a 'confirmation' button displayed on a computer terminal or a mobile phone terminal to submitting the splicing answer is taken as the deadline time. Finally, the method adopted by the tester for recovering the image is also required to be obtained to make self evaluation, namely, a second subjective evaluation is given.

1024. A second ratio of a second accuracy rate of the tester's recovery of the test image and a second ratio of the expressions in the recovery of the words is obtained.

Specifically, after all the test images selected in step 1021 are completed, the accuracy of the tester in the recovery test is calculated as a second accuracy, and a second proportion of the expressions in the second subjective evaluation, that is, a second proportion, is calculated, so that the ability of the tester to recognize is evaluated according to the second detection parameters.

In one possible embodiment, step 102 further comprises:

1025. and determining a graphic database for detecting the comprehension ability of the tester from the graphic databases with different difficulty levels according to the age and education condition of the tester.

Specifically, the graph databases with different difficulties are established in advance according to the number, the intercrossing condition, the spatial resolution and the like of each basic graph, for example, the graph databases with five difficulties may be established, and the graph databases with five difficulties may be respectively established as a first-level difficulty graph database, a second-level difficulty graph database, a third-level difficulty graph database, a fourth-level difficulty graph database and a fifth-level difficulty graph database. For example, for a 38-year-old tester a, the education condition is primary school, and in order to improve the accuracy of the comprehension ability detection result, a graphic database originally suitable for ages 13 to 18 and having an early school calendar (for example, the graphic database is a second-level difficulty graphic database) may be selected to perform comprehension ability detection on the tester a.

Similarly, when the graphic database of the difficulty level is determined according to the age and education condition of the tester, another identification score can be calculated by weighting according to the age and education condition (the respective weights of the age and education condition can be adjusted as required), and the different identification scores correspond to the graphic databases of different difficulty levels. For example, the age is weighted to 0.4, the education is weighted to 0.6, the recognition score of the tester is determined by weighting calculation according to tables 2 and 3, and the graph database for the test of the comprehension of the tester can be determined according to table 4.

TABLE 4

During testing, a certain number of images can be randomly selected from the determined graphic database to serve as test images, or different numbers of test images can be determined according to the age and education condition of a tester, for example, the number of the test images is divided into 5 grades, namely 5, 10, 15, 20 and 25 according to the sequence of the difficulty level from low to high, and the different numbers of the test images correspond to the graphic databases with different difficulty levels.

For example, in the test, the first difficulty figure database selects 5 test images.

The second difficulty graphic database selects 10 test images.

The third difficulty graphic database selects 15 test images.

The fourth difficulty graphic database selects 20 test images.

The fifth difficulty graphic database selects 25 test images.

With the method of step 1021, it is necessary to select a test image that matches the age and education of the tester from a single graphic database with a large number of graphics, and the time taken for selection is long. And step 1021 and step 1025 are combined, and step 1025 is utilized to determine the graphic database which is consistent with the age and the education condition of the tester, and then a certain number of test images are selected from the graphic database, so that the selection time is short, and the experience and the feeling of the tester are improved.

Particularly, the pattern distinguishing test is carried out between the first period (for example, the first two weeks) before the formal test is started, so that the formal test result is ensured not to be influenced by factors such as the image recognition quantity of a testee, and the pattern database called in the formal test can be adjusted in a targeted manner. And the interference effect of the preliminary experiment on the normal detection can be reduced to the maximum extent by carrying out formal tests later. For example, when the first accuracy is lower than 50% in the preliminary test, the difficulty of the graph may be adjusted appropriately, and the image in the graph database with the lower difficulty may be used as the test image of the formal test.

In one possible embodiment, whether the tester makes a pause or not can be judged according to the first electroencephalogram index and the first electroencephalogram index threshold, and when the first electroencephalogram index is larger than the first electroencephalogram index threshold, the tester makes a pause or not. Otherwise, it indicates that the tester did not realize. For example, the first electroencephalogram index threshold is a gamma frequency band energy threshold and a temporal lobe alpha frequency band energy threshold, when the gamma frequency band energy corresponding to a certain test word is greater than the gamma frequency band energy threshold, and the temporal lobe alpha frequency band energy corresponding to the test word is greater than the temporal lobe alpha frequency band energy threshold, it is indicated that the tester makes a pause, otherwise, the tester does not make a pause. Similarly, whether the tester makes a pause or not can be judged according to the second electroencephalogram index and the second electroencephalogram index threshold, and the repeated description is omitted.

In one possible embodiment, step 103 comprises:

1031. calculating a relative Euclidean distance between the first detection parameter and the parameters of the first achievement upper limit and the first achievement lower limit according to the first detection parameter, the parameters of the first achievement upper limit and the parameters of the first achievement lower limit, and determining first achievement evaluation information of the tester according to the relative Euclidean distance;

specifically, a detection parameter corresponding to a first lower limit of the learning ability is preset, and a detection parameter corresponding to a first lower limit of the learning ability is preset, wherein the detection parameter of the first lower limit of the learning ability is a detection parameter in a state without a tester, a first electroencephalogram index of the first lower limit of the learning ability is idle white noise, and the first reaction time and the first subjective evaluation are null; and the first accuracy and the first percentage are 0. And the detection parameter of the first mental effort upper limit can be the parameter under the ideal mental state, namely to every test word, its first brain electric index all accords with the mental index (being that first brain electric index is greater than first brain electric index threshold value), and first reaction time is less than ideal mental time (can be according to the statistics of the limited number of experiments and draw), and first correct rate is 100%, and first subjective evaluation is the mental formula. The detection parameter corresponding to the first learning upper limit can be finely adjusted as required, for example, the first accuracy can be 99%.

The first detection parameter (including the first electroencephalogram index, the first reaction time, the first subjective evaluation, the first accuracy, and the first proportion of the first insight) corresponds to one point in the multidimensional space, and similarly, the parameter of the first insight upper limit corresponds to the point P1 in the multidimensional space, and the parameter of the first insight lower limit corresponds to the point P2 (corresponding to the origin in the multidimensional space). For example, the first detection parameter of tester A is point A in the multidimensional space, and the relative Euclidean distances between point A, point P1 and point P2 are equivalent to calculating the relative position of point A between point P1 and point P2. The first demonstration evaluation information of the tester can be obtained according to the relative euclidean distance, and assuming that the first demonstration evaluation information is the demonstration score, since the point P1 represents the highest score of the demonstration ability, and assuming that the score is 1000, the evaluation information can be converted into the corresponding demonstration score according to the relative euclidean distance, for example, the euclidean distance between the points P1 and P2 is 100, and the relative euclidean distance between the points a, P1 and P2 is 40, and the demonstration score of the tester a is 400.

When the first contemplation evaluation information is contemplation of the ability level/gear, the first contemplation evaluation information may be obtained according to a correspondence relationship between different relative euclidean distances and different contemplation ability levels/gears. Assuming that the relative Euclidean distance corresponding to the first upper limit of the instant recognizability is 100, the relative Euclidean distance is 100-80 corresponding to the first recognizability of the first recognizability, 80-60 corresponding to the second recognizability of the first recognizability, 60-40 corresponding to the first recognizability of the second recognizability, and 40-0 corresponding to the second recognizability of the first recognizability. If tester A has a relative Euclidean distance of 40, then tester A's comprehension capability is rated Ding et al comprehension capability.

1032. And calculating the relative Euclidean distance between the second detection parameter and the parameters of the second upper limit of the recognizability and the parameters of the second lower limit of the recognizability according to the second detection parameter, the parameters of the second upper limit of the recognizability and the parameters of the second lower limit of the recognizability, and determining the second recognizability evaluation information of the tester according to the relative Euclidean distance.

Specifically, similarly to the first detection parameter, it is necessary to preset a detection parameter corresponding to the second recognition upper limit and a detection parameter corresponding to the second recognition lower limit. The detection parameter of the second lower limit of the comprehension is a detection parameter in a state without a tester, the second electroencephalogram index of the second lower limit of the comprehension is idle white noise, and the second reaction time and the second subjective evaluation are null; and the second accuracy and the second percentage are 0. And the detection parameter of the second contemplation upper limit can be the parameter under the ideal contemplation state, namely for each test image, the second brain electrical index accords with the contemplation index, the second reaction time is less than the ideal contemplation time (can be obtained according to limited experimental statistics), the second correct rate is 100%, and the second subjective evaluation is the contemplation. The detection parameter corresponding to the second learning upper limit can be finely adjusted as required, for example, the second accuracy can be 99%.

According to the same method as the first detection parameter, the relative euclidean distance between the second detection parameter and the parameter of the second superior comprehension limit and the parameter of the second inferior comprehension limit can be calculated, and the second comprehension evaluation information can be determined according to the relative euclidean distance, and the second comprehension evaluation information can be a comprehension ability score or a comprehension ability grade/gear.

In particular, if the comprehension evaluation information is calculated using only the once-time graph and word comprehension detection result, the comprehension evaluation accuracy is low. Therefore, the same tester can perform detection for a plurality of times (more than 2 times) to obtain a plurality of first detection parameters and a plurality of second detection parameters, obtain a plurality of relative Euclidean distances according to the plurality of first detection parameters, average the plurality of relative Euclidean distances, use the average as the final relative Euclidean distance of the tester, and calculate first contemplation evaluation information according to the final relative Euclidean distance, thereby ensuring the accuracy of the first contemplation evaluation information and avoiding contemplation evaluation errors caused by single detection. Similarly, according to the second detection parameters, calculating to obtain a plurality of relative Euclidean distances, then averaging, and calculating second comprehension evaluation information according to the average.

The numbers in step 1031 and step 1032 do not limit the execution sequence therebetween, and step 1031 may be executed first, and then step 1032 may be executed; step 1032 may be executed first, and then step 1031 may be executed; step 1031 and step 1032 may also be performed simultaneously.

In another possible embodiment, step 103 comprises:

1033. clustering first detection parameters and second detection parameters of a plurality of testers respectively to obtain a plurality of first sets and a plurality of second sets;

specifically, after the first detection parameter and the second detection parameter of different testers are obtained through the

steps

101 and 102, the insight evaluation can be performed on all testers at one time. Clustering the first detection parameters to obtain a plurality of first sets; and clustering the second detection parameters to obtain a second set. Wherein the plurality of fingers is more than 2.

The first detection parameter is used as one of input parameters of the convolutional neural network, a first electroencephalogram index, a first accuracy, a first response time and a first proportion are integrated, classification results of a plurality of first detection parameters are obtained through an unsupervised learning clustering analysis method, and a plurality of first sets are obtained, wherein each first set is provided with a clustering center.

Similarly, clustering processing of unsupervised learning is performed according to the second detection parameters to obtain a plurality of second sets, and each second set has a cluster center.

1034. Calculating relative Euclidean distances between the cluster center of the first set and the parameters of the first achievement upper limit and the first achievement lower limit according to the cluster center of the first set, the parameters of the first achievement upper limit and the parameters of the first achievement lower limit, and determining first achievement evaluation information of the testers in the first set according to the relative Euclidean distances;

specifically, according to the method of step 1031, the cluster center of the first set is treated in the same way as the first detection parameter of step 1031, so as to obtain the relative euclidean distance between the cluster center of the first set and the parameters of the first learning upper limit and the parameters of the first learning lower limit, and then the first learning evaluation information of the cluster center can be determined according to the relative euclidean distance.

For example, the first set includes 10 first detection parameters, which respectively correspond to 10 testers, the first insight evaluation information of the cluster center of the first set is the equal insight capability, and then the first insight evaluation information of the 10 testers in the first set is the equal insight capability.

1035. And calculating the relative Euclidean distance between the cluster center of the second set and the parameters of the second upper limit and the second lower limit of the comprehension force according to the cluster center of the second set, the parameters of the second upper limit of the comprehension force and the parameters of the second lower limit of the comprehension force, and determining the second comprehension force evaluation information of the testers in the second set according to the relative Euclidean distance.

Specifically, similarly to the method in step 1034, referring to the method in step 1032, the cluster centers of the second set are equal to the second detection parameter in step 1032, and the second insight evaluation information of the cluster centers of the second set is calculated. The second comprehension evaluation information is used as second comprehension evaluation information of all testers corresponding to the second set.

By using the method of step 1033, step 1034 and step 1035, the tester can perform the detection of the comprehension ability only once, that is, relatively accurate comprehension evaluation information can be obtained, and the testing experience of the tester is improved without multiple detections.

The numbers in step 1034 and step 1035 do not limit the execution sequence therebetween, and step 1034 may be executed first, and then step 1035 is executed; alternatively, step 1035 may be performed first, and then step 1034 may be performed; step 1034 and step 1035 may also be performed simultaneously.

In step 103, first and second comprehension evaluation information are obtained according to first and second detection parameters of a tester respectively, and comprehension ability of the tester is comprehensively judged through subjective evaluation and objective reaction time, accuracy and electroencephalogram characteristic data of the tester, so that the problem of excessive dependence on subjective evaluation of the tester is avoided, and a basis is provided for deep analysis of electroencephalogram data.

Further, the comprehension ability evaluation method further comprises the following steps:

and obtaining a comprehensive analysis report of the tester about the insight according to the first insight evaluation information and the second insight evaluation information of the tester.

The non-language graph detection system is introduced in the process of the early-learning detection, so that the dependence of the traditional scheme on Chinese and English vocabulary is greatly compensated, and the method is suitable for being popularized to student groups with different age groups and development degrees. And two sets of normal forms of speech and nonspeech are adopted, and the comprehensive comparison of the comprehensive comprehension abilities of the students in processing the language tasks and the graphic tasks is carried out based on the detection results of the two normal forms, so that a depth report of the comprehensive comprehension abilities of the students is obtained, and the detection results of the speech tasks and the graphic tasks are separately explained. The comprehensive comparison of the performances of the students in the language task and the graphic task can systematically analyze the comprehension abilities of the students so as to obtain more comprehensive and reliable detection results. In addition, further analysis of the two types of detection conclusions helps to understand deeply the cognitive performance of testers (e.g., students) in different tasks and provides important references and basis for educational practices.

The integrated analysis report may include the first insight evaluation information, the second insight evaluation information, an analysis for the first insight evaluation information, an analysis for the second insight evaluation information, an insight culture suggestion, and the like. The comprehensive analysis report can be output on the computer side or the mobile phone side after the tester completes the detection of the comprehension capability. When the mobile phone end outputs the comprehensive analysis report, the cloud server can calculate the first demonstration evaluation information and the second demonstration evaluation information, generate the comprehensive analysis report and send the comprehensive analysis report to the mobile phone end of the tester, so that the tester can check the evaluation result conveniently.

Based on the description of the embodiment of the method for evaluating the comprehension ability, the embodiment of the present invention further discloses a device for evaluating the comprehension ability, and referring to fig. 3, fig. 3 is a schematic structural diagram of the device for evaluating the comprehension ability provided by the embodiment of the present invention, and the device for evaluating the comprehension ability includes:

the first detection module 301 is configured to perform comprehension capability detection on a tester according to an english word database, and acquire a first detection parameter;

the second detection module 302 is configured to perform comprehension capability detection on the tester according to the graph database, and obtain a second detection parameter, where the first detection parameter and the second detection parameter are associated with the comprehension capability;

the evaluation module 303 is configured to obtain first insight evaluation information and second insight evaluation information of the tester according to the first detection parameter and the second detection parameter, respectively.

In one possible embodiment, the first detection parameter includes a first brain electrical indicator, a first reaction time, a first subjective evaluation, a first accuracy, and a first percentage of an insight; the first detection module 301 includes:

a first sub-module 3011, configured to determine a test word for capability detection according to an english word database;

a second sub-module 3012 for outputting the test words after the alphabetical order is disturbed to the tester;

the third sub-module 3013, configured to obtain a first electroencephalogram index of the tester in the process of recovering the word, a first reaction time of recovering the word, and a first subjective evaluation of the way of recovering the word by the tester, where the first electroencephalogram index is related to a comprehension capability, and the first subjective evaluation includes a comprehension formula and an analysis formula;

the fourth sub-module 3014 is configured to obtain a first correctness rate of the test word recovered by the tester and a first ratio of the expressions in the word recovery mode.

In one possible embodiment, the first detection module 301 comprises:

a fifth sub-module 3015, configured to determine, according to the age and education condition of the tester, an english word database for detecting the recognition ability of the tester from among english word databases of different difficulty levels;

in one possible embodiment, the first detection module 301 comprises:

a sixth sub-module 3016, configured to obtain, after the detection of the comprehension ability, a recognition result of the test word by the tester;

the seventh sub-module 3017, configured to reject the first electroencephalogram index of the test word, the first response time, and the first subjective evaluation if the identification result is an identification failure;

the fourth sub-module 3014 is specifically configured to obtain a first accuracy and a first percentage according to the data after being culled.

In one possible embodiment, the second detection parameter includes a second electroencephalogram index, a second reaction time, a second subjective evaluation, a second accuracy, and a second percentage of the insight; the second detection module 302 includes:

an eighth sub-module 3021, configured to determine a test image for capability detection according to a graph database, where the graph database includes a plurality of images, and each image is composed of a plurality of basic graphs;

a ninth sub-module 3022 for outputting the test image with the position of the basic pattern being disturbed to the tester;

a tenth submodule 3023, configured to obtain a second electroencephalogram index of the tester in the process of recovering the test image, a second reaction time of recovering the test image, and a second subjective evaluation of the tester on a manner of recovering the test image, where the second electroencephalogram index is related to an ability of performing a mental function, and the second subjective evaluation includes a mental formula and an analysis formula;

an eleventh sub-module 3024 for obtaining a second ratio of the second correctness rate of the tester recovering the test image and the comprehension in the word recovery mode.

In one possible embodiment, the second detection module 302 includes:

a twelfth sub-module 3025 for determining a graphic database for detecting the ability of the tester to perceive from graphic databases of different difficulty levels according to the age and education of the tester.

In one possible embodiment, the evaluation module 303 comprises:

a thirteenth sub-module 3031, configured to calculate a relative euclidean distance between the first detection parameter and the first upper limit of the learning force and between the first detection parameter and the first lower limit of the learning force according to the first detection parameter, the first upper limit of the learning force and the first lower limit of the learning force, and determine first learning force evaluation information of the tester according to the relative euclidean distance;

a fourteenth sub-module 3032, configured to calculate, according to the second detection parameter, the second parameter of the upper limit of the recognizability, and the second parameter of the lower limit of the recognizability, a relative euclidean distance between the second detection parameter and the second parameter of the upper limit of the recognizability, and the second parameter of the lower limit of the recognizability, and determine, according to the relative euclidean distance, the second recognizability evaluation information of the tester.

In one possible embodiment, the evaluation module 303 comprises:

a fifteenth sub-module 3033, configured to cluster the first detection parameters and the second detection parameters of the multiple testers respectively to obtain multiple first sets and multiple second sets;

a sixteenth sub-module 3034, configured to calculate, according to the cluster center of the first set, the parameter of the first upper limit of the learning force, and the parameter of the first lower limit of the learning force, a relative euclidean distance between the cluster center of the first set and the parameter of the first upper limit of the learning force, and the parameter of the first lower limit of the learning force, and determine, according to the relative euclidean distance, first learning force evaluation information of the tester in the first set;

a seventeenth sub-module 3035, configured to calculate, according to the cluster center of the second set, the parameter of the second upper limit of the learning force, and the parameter of the second lower limit of the learning force, a relative euclidean distance between the cluster center of the second set and the parameter of the second upper limit of the learning force, and the parameter of the second lower limit of the learning force, and determine, according to the relative euclidean distance, second learning force evaluation information of the tester in the second set.

It should be noted that, for the specific functional implementation of the comprehension ability evaluation apparatus, reference may be made to the description of the comprehension ability evaluation method above, and details thereof are not repeated here. The units or modules in the comprehension ability evaluation device may be respectively or completely combined into one or several other units or modules to form the unit or module, or some unit(s) or module(s) thereof may be further split into multiple functionally smaller units or modules to form the unit or module(s), which may achieve the same operation without affecting the achievement of the technical effect of the embodiments of the present invention. The above units or modules are divided based on logic functions, and in practical applications, the functions of one unit (or module) may also be implemented by a plurality of units (or modules), or the functions of a plurality of units (or modules) may be implemented by one unit (or module).

Based on the above description of the method embodiment and the apparatus embodiment, the embodiment of the present invention further provides an insight capability evaluation device.

Fig. 4 is a schematic structural diagram of an understanding ability evaluation apparatus according to an embodiment of the present invention. As shown in fig. 4, the above-mentioned comprehension ability evaluation means may be applied to the comprehension ability evaluation apparatus 400, and the comprehension ability evaluation apparatus 400 may include: processor 401, network interface 404 and memory 405, the insight capability assessment apparatus 400 may further comprise: a user interface 403, and at least one communication bus 402. Wherein a communication bus 402 is used to enable connective communication between these components. The user interface 403 may include a Display (Display) and a Keyboard (Keyboard), and the selectable user interface 403 may also include a standard wired interface and a standard wireless interface. The network interface 404 may optionally include a standard wired interface, a wireless interface (e.g., WI-FI interface). The memory 405 may be a high-speed RAM memory or a non-volatile memory (e.g., at least one disk memory). The memory 405 may alternatively be at least one storage device located remotely from the aforementioned processor 401. As shown in fig. 4, the memory 405, which is a type of computer storage medium, may include therein an operating system, a network communication module, a user interface module, and a device control application program.

In the device 400 for assessing insight capabilities shown in FIG. 4, a network interface 404 may provide network communication functionality; and the user interface 403 is primarily an interface for providing input to a user; and processor 401 may be configured to invoke a device control application stored in memory 405 to implement the method steps of the insight capability assessment method described above.

It should be understood that the apparatus 400 for evaluating comprehension ability described in the embodiments of the present invention may perform the method for evaluating comprehension ability described above, and may also perform the device for evaluating comprehension ability described above, which is not described herein again. In addition, the beneficial effects of the same method are not described in detail.

Further, here, it is to be noted that: an embodiment of the present invention further provides a computer storage medium, and the computer storage medium stores a computer program executed by the aforementioned device for assessing the comprehension ability, and the computer program includes program instructions, which when executed by the processor, are capable of performing the foregoing description of the method for assessing the comprehension ability, and therefore, the detailed description thereof will not be repeated here. In addition, the beneficial effects of the same method are not described in detail. For technical details not disclosed in the embodiments of the computer storage medium to which the present invention relates, reference is made to the description of the method embodiments of the present invention.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.

The above disclosure is only for the purpose of illustrating the preferred embodiments of the present invention, and it is therefore to be understood that the invention is not limited by the scope of the appended claims.

Claims

1. A method of assessing insight capabilities, comprising:

2. The method of claim 1, wherein the first detection parameter comprises a first brain electrical indicator, a first reaction time, a first subjective evaluation, a first accuracy, and a first proportion of an insight;

outputting the test words with disordered alphabetical order to the tester;

3. The method of claim 2, wherein performing a comprehension ability test on the tester from an english word database to obtain a first test parameter comprises:

4. The method of claim 2 or 3, wherein the obtaining the first detection parameter from the comprehension ability detection of the tester from the English word database comprises:

5. The method of any of claims 1 to 3, wherein the second detection parameter comprises a second brain electrical indicator, a second reaction time, a second subjective assessment, a second accuracy, and a second aspect of an insight;

6. The method of any one of claims 1 to 3, wherein obtaining first and second insight evaluation information of the tester from the first and second detection parameters, respectively, comprises:

7. The method of any one of claims 1 to 3, wherein obtaining first and second insight evaluation information of the tester from the first and second detection parameters, respectively, comprises:

8. A device for assessing a mental ability, comprising:

9. A device for assessing a mental ability, comprising: a processor and a memory;

the processor is coupled to the memory, wherein the memory is configured to store program code and the processor is configured to invoke the program code to perform the insight capability assessment method of any of claims 1-7.

10. A computer storage medium, characterized in that the computer storage medium stores a computer program comprising program instructions that, when executed by a processor, perform the insight capability assessment method of any one of claims 1-7.