CN113421175A - Capability test grading method and device - Google Patents
Capability test grading method and device Download PDFInfo
- Publication number
- CN113421175A CN113421175A CN202110758961.9A CN202110758961A CN113421175A CN 113421175 A CN113421175 A CN 113421175A CN 202110758961 A CN202110758961 A CN 202110758961A CN 113421175 A CN113421175 A CN 113421175A
- Authority
- CN
- China
- Prior art keywords
- knowledge
- module
- user
- test
- capability
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012360 testing method Methods 0.000 title claims abstract description 149
- 238000000034 method Methods 0.000 title claims abstract description 34
- 239000011159 matrix material Substances 0.000 claims abstract description 67
- 238000004458 analytical method Methods 0.000 claims abstract description 55
- 238000013528 artificial neural network Methods 0.000 claims abstract description 33
- 238000013101 initial test Methods 0.000 claims abstract description 29
- 238000011176 pooling Methods 0.000 claims description 11
- 230000004044 response Effects 0.000 claims description 7
- 230000007175 bidirectional communication Effects 0.000 claims description 3
- 230000006854 communication Effects 0.000 claims description 3
- 238000004891 communication Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 12
- 238000011156 evaluation Methods 0.000 description 6
- 238000010205 computational analysis Methods 0.000 description 5
- 230000007613 environmental effect Effects 0.000 description 4
- 238000011056 performance test Methods 0.000 description 4
- 238000010998 test method Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000011835 investigation Methods 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000003344 environmental pollutant Substances 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 231100000719 pollutant Toxicity 0.000 description 1
- 238000001303 quality assessment method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/20—Education
- G06Q50/205—Education administration or guidance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Educational Administration (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Tourism & Hospitality (AREA)
- Strategic Management (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Educational Technology (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- General Business, Economics & Management (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application provides a capability test grading method, which comprises the following steps: establishing a capability test question bank; generating an index of the capability test question bank; generating initial test information, and generating a first group of questions according to the initial test information; collecting first test data and transmitting the first test data to an analysis module; generating a user data matrix graph by using the first test data, inputting the user data matrix graph into the ability level hierarchical deep neural network, and carrying out classification identification on the user data matrix graph, wherein the result of the classification identification contains the ability level hierarchical information of the user; a second set of topics is generated based on the capability level rating information. In addition, this application still provides a capability test grading plant, includes: the system comprises an item bank storage module, an index module, a control module, an analysis module and a user terminal. The capacity test grading device is used for executing the capacity test grading method.
Description
Technical Field
The present application relates generally to the field of talent training and testing, and more particularly to a method and apparatus for competency test grading for people with different competency levels.
Background
The current testing method mainly provides a set of questions for all testers, and evaluates the tester level according to the correct answer rate of the testers. Such tests are typically one-dimensional, reflecting ultimately only a total score or rank, regardless of what the test covers. In addition, before testing, the current ability level of the tester cannot be known, and the same set of test questions cannot achieve the best evaluation effect for each tester. However, talent training in any skill or knowledge field includes many factors for investigation, talent ability is multidimensional, and the current testing method and testing technology based on computer and network systems cannot meet the evaluation of abilities of different testers in multiple dimensions.
In view of the above, how to provide personalized tests to different testers according to the current capability level of the testers and perform multi-dimensional and multi-level comprehensive investigation and evaluation on the capability level of the testers in a certain field is a technical problem to be solved in the field.
Disclosure of Invention
One object of the present application is to provide a capability test ranking method, comprising:
step S1, a capability test question library is created, which contains questions, knowledge structure labels of the questions, and question numbers, and is stored in the question library storage module 41. The knowledge structure labels indicate the categories of specific knowledge contents tested by the subjects, one subject has at least one knowledge structure label, and the subjects can be classified according to the knowledge structure labels. Preferably, the question bank storage module 41 is a server storing the capability test question bank.
Step S2, an index of the capability test question library is generated and stored in the index module 42, the index connects all knowledge structure labels included in the question library according to the correlation between the knowledge structure labels to form a structured system, and the question number of the question including the knowledge structure label is recorded under each knowledge structure label. The correlation among the knowledge structure labels is the correlation among the represented knowledge contents, and the formed structured system can embody the knowledge structure in the tested knowledge field, and is formed by connecting a plurality of nodes, wherein each node is a knowledge structure label. Preferably, the indexing module 42 is a server that stores the index.
Step S3, generating initial test information, the analyzing module 44 generating a first set of knowledge structure labels according to the initial test information and transmitting the first set of knowledge structure labels to the indexing module 42, the indexing module 42 generating a first set of topic numbers with the first set of knowledge structure labels, calling the first set of topics with the first set of topic numbers from the topic repository storage module 41, and transmitting the first set of topics to the user terminal 45. The initial test information is the initial information required by the user to participate in the test, and comprises user identification and personal information, selected test content, initial setting of the user knowledge ability level and the like.
In step S4, the user terminal 45 collects first test data generated by the user in response to the first set of questions and transmits the first test data to the analysis module 44, wherein the first test data includes answers to the first set of questions input by the user from the user terminal 45.
In step S5, the analysis module 44 generates a user data matrix map by using the first test data, and inputs the user data matrix map into the ability level hierarchical deep neural network 441 in the analysis module 44, so as to perform classification and identification on the user data matrix map, where the result of the classification and identification includes the ability level hierarchical information of the user. The analysis module 44 analyzes the first test data, and arranges and organizes the analysis result in a data matrix manner to generate a user data matrix map, where the user data matrix map includes a plurality of data units, and each data unit corresponds to an image pixel that can be identified and analyzed by the deep neural network. The competence level hierarchical deep neural network 441 is a well-trained deep neural network model capable of classifying and identifying the knowledge competence level of a user according to a user data matrix diagram.
In step S6, the analysis module 44 generates a second group of knowledge structure labels according to the capability level classification information, transmits the second group of knowledge structure labels to the indexing module 42, and the indexing module 42 generates a second group of topic numbers with the second group of knowledge structure labels, and calls the second group of topics with the second group of topic numbers from the topic library storage module 41, and transmits the second group of topics to the user terminal 45. The second set of knowledge structure tags based on the capability level ranking information more accurately represents the user's current capability level within the tested knowledge domain than the first set of knowledge structure tags, and thus the second set of topics invoked according to the second set of knowledge structure tags may provide better, user-personalized test results.
Preferably, the knowledge structure labels comprise knowledge level labels and knowledge point number labels, the knowledge level labels divide the capability test question bank into at least two knowledge levels, each knowledge level has at least one knowledge point, each knowledge point has a number, and the number is the knowledge point number label of the knowledge point.
Preferably, the knowledge structure label further comprises a knowledge module label, the knowledge module label divides the capability test question library into at least two knowledge modules, each knowledge module is provided with at least one knowledge point, each knowledge point corresponds to one knowledge module and one knowledge grade and is provided with a number, and the number is the knowledge point number label of the knowledge point.
Preferably, the index comprises a structured system formed by connecting and forming according to at least two correlation relations among the knowledge module labels, the knowledge level labels and the knowledge point number labels. The structured system can be constructed mainly according to knowledge modules, each knowledge module subdivides a knowledge level; the structured system may also be constructed primarily in terms of knowledge levels, each knowledge level subdividing a knowledge module.
Preferably, the initial test information comprises user-defined knowledge modules and knowledge levels input from the user terminal 45, and the first set of knowledge structure tags are knowledge module tags and knowledge level tags corresponding to the user-defined knowledge modules and knowledge levels.
Preferably, the initial test information is default initial test information, and the first set of knowledge structure tags are knowledge module tags and knowledge level tags corresponding to default knowledge modules and default knowledge levels. Default means that the default is preset and does not need user specification. The default initial test information, default knowledge module, and default knowledge level are stored in a control module 43 communicatively coupled to the user terminal 45, the analysis module 44, and the indexing module 42. Control module 43, upon receiving the default initial test information, will invoke a first set of topics with default knowledge module labels and default knowledge level labels.
Preferably, the user data matrix map is a two-dimensional data matrix arranged according to the knowledge structure labels, and the ability level hierarchical deep neural network 441 performs a two-dimensional convolution operation on the user data matrix map. Preferably, the user data matrix map is a two-dimensional data matrix arranged according to the knowledge level labels, and each data unit in the two-dimensional data matrix is user data of one knowledge point at one knowledge level. Preferably, the user data in the data unit is obtained by statistical analysis of the user response results of the plurality of topics under the knowledge point represented by the data unit.
Preferably, the user data matrix map is a three-dimensional data matrix arranged according to the knowledge structure labels, and the ability level hierarchical deep neural network 441 performs a three-dimensional convolution operation on the user matrix map. Preferably, the user data matrix map is a three-dimensional data matrix arranged in terms of knowledge module labels and knowledge level labels. Each data unit in the three-dimensional data matrix is user data of a knowledge point under a knowledge level in a knowledge module. Preferably, the user data in the data unit is obtained by statistical analysis of the user response results of the plurality of topics under the knowledge point represented by the data unit.
Preferably, the capability level hierarchical deep neural network 441 deep neural network comprises convolutional layers, pooling layers, and fully-connected layers. The convolution layer uses convolution core to carry out convolution operation on the user data matrix diagram, the pooling layer carries out down sampling on the output of the convolution layer, and finally classification is realized through the full connection layer and the grading information of the user capacity level is output. Preferably, the capability level hierarchical deep neural network 441 has at least one convolutional layer and at least one pooling layer therein.
Preferably, between step S2 and step S3, further comprising: step S21, a user profile is created and stored in the index module 42, where the user profile includes all user data generated during the performance test classification process using the performance test classification method of the present application, including a first group of questions, first test data, performance level classification information, a second group of questions, answer schedule, test time, and the like. Any data or information in the same user profile stored in the indexing module 42 may be invoked at a different user terminal 45.
Preferably, after step S6, the method further includes: in step S7, the user terminal 45 collects second test data generated by the user in response to the second set of questions and transmits the second test data to the analysis module 44, where the second test data includes answers of the second set of questions input by the user from the user terminal 45, and the analysis module 44 analyzes the second test data, returns the analysis result to the user terminal 45, and stores the analysis result in the user profile of the indexing module 42.
Preferably, after step S7, the method further includes: in step S8, returning to step S3, the analysis result of the second test data stored in the user profile is called from the index module 42 by the control module 43 as the initial test information, and steps S3 to S7 are performed.
Preferably, the user terminal 45 further comprises a biometric acquisition device, and the test data further comprises biometric data of the user generated by the biometric acquisition device. Preferably, the biological characteristic acquisition device comprises an infrared emission device and an infrared receiving device, and the acquired biological characteristic data is the body surface temperature change and/or the heart rate change of the user in the test process. Preferably, the biometric acquisition device further comprises a skin electrical sensor, and the acquired biometric data is skin resistance change of the user during the test.
Another object of the present application is to provide a capability test ranking device, comprising: the question bank comprises a question bank storage module 41, an index module 42, a control module 43, an analysis module 44 and a user terminal 45; wherein, the question storage module stores the capability test question bank and transmits the questions in the capability test question bank to the user terminal 45; the index module 42 stores the index of the capability test question bank and calls the question in the question storage module according to the index; the control module 43 comprises a processor and a memory which is in communication connection with the processor, instructions for executing the capability test classification are stored in the memory, the control module 43 is in bidirectional communication with the user terminal 45, the index module 42 and the analysis module 44 respectively, namely the control module 43 outputs the instructions and data to the user terminal 45, the index module 42 and the analysis module 44 and receives the data and information from the user terminal 45, the index module 42 and the analysis module 44; the analysis module 44 analyzes the user data and comprises a capability level hierarchical deep neural network 441 for generating capability level hierarchical information of the user, wherein the capability level hierarchical deep neural network 441 comprises a convolutional layer, a pooling layer and a full-link layer; the user terminal 45 collects user data and transmits the user data to the analysis module 44. The capacity test grading device is used for executing the capacity test grading method.
Preferably, the question bank storage module 41 is a server storing the capability test question bank. Preferably, the indexing module 42 is a server that stores the index. Preferably, the user terminal 45 further comprises biometric acquisition means. Preferably, the biological characteristic acquisition device comprises an infrared transmitting device and an infrared receiving device, and is used for measuring the body surface temperature change and/or the heart rate change of the user in the test process. Preferably, the biometric acquisition device further comprises a galvanic sensor for measuring the skin resistance change of the user during the test.
The method and the device provided by the application have the beneficial effects that: adding a knowledge structure tag to questions in a capability test question bank, generating indexes of the capability test question bank by using the knowledge structure tag, and dividing the capability test question bank into a plurality of knowledge modules and knowledge levels to realize calling different question combinations according to user requirements and provide personalized test contents for users; imaging the test data and performing computational analysis by using a deep neural network, so that multi-dimensional test result evaluation is realized, and the capability level of the user is more refined and accurately graded according to different elements; thirdly, by establishing a user profile in the index module 42 and storing the test data of the user, the user can call the completed test content and test result at different user terminals 45 or different time and place; and fourthly, the user can continuously iterate the hierarchical information of the ability level through the hierarchical test of the ability level provided by the application, and can selectively refresh different aspects in the ability level at different iteration rates according to the requirements on different knowledge ability elements.
In summary, the present application provides a capability test classification method, including: s1, establishing a capability test question bank, and storing the capability test question bank in the question bank storage module 41; s2, generating an index of the capability test item library and storing the index in the index module 42; s3, generating initial test information, generating a first group of questions according to the initial test information and transmitting the first group of questions to the user terminal 45; s4, the user terminal 45 collects the first test data and transmits the first test data to the analysis module 44; s5, the analysis module 44 generates a user data matrix map by using the first test data, and inputs the user data matrix map into the ability level hierarchical deep neural network 441 in the analysis module 44, and performs classification and identification on the user data matrix map, where the result of the classification and identification includes the ability level hierarchical information of the user; and S6, generating a second group of topics according to the capability level classification, and transmitting the second group of topics to the user terminal 45. In addition, this application still provides a capability test grading plant, includes: the question bank storage module 41, the index module 42, the control module 43, the analysis module 44 and the user terminal 45. The capacity test grading device is used for executing the capacity test grading method.
Drawings
In order to more clearly illustrate the technical solution of the present application, the drawings used in the description of the embodiments will be briefly described below. It should be apparent that the drawings depict only some embodiments of the application and that others can be derived from them by those skilled in the art without the exercise of inventive faculty.
FIG. 1 shows a flow diagram of one embodiment of a capability ranking test method provided herein;
FIG. 2 shows a schematic diagram of one embodiment of the ability level hierarchical deep neural network 441 convolving a two-dimensional user data matrix map 21;
FIG. 3 shows a schematic diagram of one embodiment of the ability level hierarchical deep neural network 441 convolving the three-dimensional user data matrix map 31;
FIG. 4 shows a schematic diagram of one embodiment of a capability test grading apparatus provided herein.
In the figure, 21-a two-dimensional user data matrix diagram; 22-two-dimensional convolution kernel; 23-two-dimensional convolution output; 31-three dimensional user data matrix map; 32-three-dimensional convolution kernel; 33-three-dimensional convolution output; 41-question bank storage module; 42-an index module; 43-a control module; 44-an analysis module; 441-competency level hierarchical deep neural network; 45-user terminal.
Detailed Description
In order to make the objects and advantages of the present invention more apparent, the following detailed description of the present invention will be made with reference to the accompanying drawings. It should be apparent that the described embodiments are only a few embodiments of the present application, and not all embodiments. All other embodiments that can be conceived by a person skilled in the art based on the embodiments in the present application without making any creative effort belong to the protection scope of the present application.
Referring to fig. 1, fig. 1 is a flow chart of one embodiment of a capability ranking test method provided herein. The capacity test grading method comprises the following steps:
step S1, a capability test question library is created, which contains questions, knowledge structure labels of the questions, and question numbers, and is stored in the question library storage module 41. The knowledge structure labels indicate the categories of specific knowledge contents tested by the subjects, one subject has at least one knowledge structure label, and the subjects can be classified according to the knowledge structure labels. In some embodiments, the question bank storage module 41 is a server that stores the capability test question bank.
In some embodiments, the knowledge structure labels include knowledge level labels and knowledge point number labels, the knowledge level labels divide the capability test question base into at least two knowledge levels, each knowledge level has at least one knowledge point, each knowledge point has a number, and the number is the knowledge point number label of the knowledge point.
In some embodiments, the knowledge structure labels further comprise knowledge module labels, the knowledge module labels divide the capability test question base into at least two knowledge modules, each knowledge module has at least one knowledge point therein, each knowledge point corresponds to a knowledge module and a knowledge level, and has a number, and the number is a knowledge point number label of the knowledge point.
Step S2, an index of the capability test question library is generated and stored in the index module 42, the index connects all knowledge structure labels included in the question library according to the correlation between the knowledge structure labels to form a structured system, and the question number of the question including the knowledge structure label is recorded under each knowledge structure label. The correlation among the knowledge structure labels is the correlation among the represented knowledge contents, and the formed structured system can embody the knowledge structure in the tested knowledge field, and is formed by connecting a plurality of nodes, wherein each node is a knowledge structure label. In some embodiments, the indexing module 42 is a server that stores the index.
Taking the capability grading test method in the field of "environmental quality assessment" as an example, the capability test question bank includes a knowledge module label, a knowledge grade label, a knowledge point number label and a question number (represented by letters M, L, N and T, respectively), and a part of the index of the capability test question bank is shown in table 1 below:
TABLE 1
In some embodiments, the index comprises a structured system that is connected and formed according to at least two correlations between knowledge module tags, knowledge level tags, and knowledge point number tags. The structured system can be constructed mainly according to knowledge modules, each knowledge module subdivides a knowledge level; the structured system may also be constructed primarily in terms of knowledge levels, each knowledge level subdividing a knowledge module.
Step S3, generating initial test information, the analyzing module 44 generating a first set of knowledge structure labels according to the initial test information and transmitting the first set of knowledge structure labels to the indexing module 42, the indexing module 42 generating a first set of topic numbers with the first set of knowledge structure labels, calling the first set of topics with the first set of topic numbers from the topic repository storage module 41, and transmitting the first set of topics to the user terminal 45. The initial test information is the initial information required by the user to participate in the test, and comprises user identification and personal information, selected test content, initial setting of the user knowledge ability level and the like.
In some embodiments, the initial test information comprises user-defined knowledge modules and knowledge levels input from the user terminal 45, and the first set of knowledge structure tags are knowledge module tags and knowledge level tags corresponding to the user-defined knowledge modules and knowledge levels.
In some embodiments, the initial test information is default initial test information and the first set of knowledge structure tags are knowledge module tags and knowledge level tags corresponding to default knowledge modules and default knowledge levels. Default means that the default is preset and does not need user specification. The default initial test information, default knowledge module, and default knowledge level are stored in a control module 43 communicatively coupled to the user terminal 45, the analysis module 44, and the indexing module 42. Control module 43, upon receiving the default initial test information, will invoke a first set of topics with default knowledge module labels and default knowledge level labels.
Taking table 1 as an example, the user may simply select the "environmental impact evaluation techniques" module and select a default knowledge level for the tested knowledge levels since the user does not know his own level. The control module 43, after receiving the initial test information from the user terminal 45, issues an instruction to the index module 42 to call a title having a M1 tag, more specifically, to call titles having M1L1 tags, M1L2 tags, and M1L3 tags. After receiving the instruction, the indexing module 42 selects the topic number with the corresponding tag according to the index, and calls the corresponding topic from the topic library storage module 41 to send to the user terminal 45.
The user can also select to test the three modules of the environmental impact evaluation technology, the environmental quality standard and the pollutant emission standard, and select one level as the knowledge level of the test. The control module 43, after receiving the initial test information from the user terminal 45, issues an instruction to call a title having an M1L1 tag, an M2L1 tag, and an M3L1 tag to the index module 42. After receiving the instruction, the indexing module 42 selects the topic number with the corresponding tag according to the index, and calls the corresponding topic from the topic library storage module 41 to send to the user terminal 45.
In some embodiments, indexing module 42 selects partial knowledge points from the knowledge points having the first label combination according to an algorithm and picks an optimized topic combination from among the selected partial knowledge points such that the best test result is obtained with as few test topics as possible.
In step S4, the user terminal 45 collects first test data generated by the user in response to the first set of questions and transmits the first test data to the analysis module 44, wherein the first test data includes answers to the first set of questions input by the user from the user terminal 45.
In step S5, the analysis module 44 generates a user data matrix map by using the first test data, and inputs the user data matrix map into the ability level hierarchical deep neural network 441 in the analysis module 44, so as to perform classification and identification on the user data matrix map, where the result of the classification and identification includes the ability level hierarchical information of the user. The analysis module 44 analyzes the first test data, and arranges and organizes the analysis result in a data matrix manner to generate a user data matrix map, where the user data matrix map includes a plurality of data units, and each data unit corresponds to an image pixel that can be identified and analyzed by the deep neural network. The competence level hierarchical deep neural network 441 is a well-trained deep neural network model capable of classifying and identifying the knowledge competence level of a user according to a user data matrix diagram.
In some embodiments, the user data matrix map is a two-dimensional data matrix arranged in accordance with knowledge structure labels, and the competency level hierarchical deep neural network 441 performs a two-dimensional convolution operation on the user data matrix map. In some embodiments, the user data matrix map is a two-dimensional data matrix arranged according to knowledge level labels, each data element in the two-dimensional data matrix being user data for a knowledge point at a knowledge level. In some embodiments, the user data in a data unit is statistically analyzed from user answers to multiple topics at the knowledge point represented by the data unit.
Referring to fig. 2, fig. 2 is a schematic diagram of one embodiment of the ability level hierarchical deep neural network 441 convolving the two-dimensional user data matrix map 21. The two-dimensional user data matrix map 21 is generated by selecting nine knowledge points of three different knowledge levels under the same knowledge module for testing, and knowledge structure labels of the nine knowledge points are respectively L1N001, L2N001, L3N001, L1N002, L2N002, L3N002, L1N003, L2N003 and L3N 003. Each data unit in the matrix corresponds to data of a knowledge point, and the data can be the correct rate of all test questions related to the knowledge point or data obtained by processing all test questions related to the knowledge point through an algorithm on the basis of the data of all test questions related to the knowledge point. In some embodiments, the data is data representing the user's ability value as a result of the correct rate of all questions related to the knowledge point being transformed by a particular model under the project testing theory (IRT). The two-dimensional user data matrix map 21 is subjected to a convolution operation by a two-dimensional convolution kernel 22 to produce a two-dimensional convolution output 23 which can be used for further computational analysis. The further computational analysis may be another layer of convolution, pooling, or full concatenation.
In some embodiments, the user data matrix map is a three-dimensional data matrix arranged in accordance with knowledge structure labels, and the competency level hierarchical deep neural network 441 performs a three-dimensional convolution operation on the user matrix map. In some embodiments, the user data matrix map is a three-dimensional data matrix arranged in terms of knowledge module labels and knowledge level labels. Each data unit in the three-dimensional data matrix is user data of a knowledge point under a knowledge level in a knowledge module. In some embodiments, the user data in a data unit is statistically analyzed from user answers to multiple topics at the knowledge point represented by the data unit.
Referring to fig. 3, fig. 3 is a schematic diagram of one embodiment of the ability level hierarchical deep neural network 441 convolving the three-dimensional user data matrix map 31. The three-dimensional user data matrix map 31 is generated by selecting twenty-seven knowledge points of three knowledge modules and three different knowledge levels for testing, wherein the knowledge modules and the knowledge levels respectively generate a dimension. As shown in fig. 3, the knowledge point structure labels for the knowledge points associated with the M1 module are M1L1N001, M1L2N001, M1L3N001, M1L1N002, M1L2N002, M1L3N002, M1L1N003, M1L2N003, and M1L3N 003. Each data unit in the matrix corresponds to data of a knowledge point, and the data can be the correct rate of all test questions related to the knowledge point or data obtained by processing all test questions related to the knowledge point through an algorithm on the basis of the data of all test questions related to the knowledge point. In some embodiments, the data is data representing the user's ability value as a result of the correct rate of all questions related to the knowledge point being transformed by a particular model under the project testing theory (IRT). The three-dimensional user data matrix map 31 is subjected to convolution operations by a three-dimensional convolution kernel 32 to produce a three-dimensional convolution output 33 which can be used for further computational analysis. The further computational analysis may be another layer of convolution, pooling, or full concatenation.
In some embodiments, the capability level hierarchical deep neural network 441 deep neural network includes a convolutional layer, a pooling layer, and a fully-connected layer. The convolution layer uses convolution core to carry out convolution operation on the user data matrix diagram, the pooling layer carries out down sampling on the output of the convolution layer, and finally classification is realized through the full connection layer and the grading information of the user capacity level is output. In some embodiments, the capability level hierarchical deep neural network 441 has at least one convolutional layer and at least one pooling layer therein.
In step S6, the analysis module 44 generates a second group of knowledge structure labels according to the capability level classification information, transmits the second group of knowledge structure labels to the indexing module 42, and the indexing module 42 generates a second group of topic numbers with the second group of knowledge structure labels, and calls the second group of topics with the second group of topic numbers from the topic library storage module 41, and transmits the second group of topics to the user terminal 45. The second set of knowledge structure tags based on the capability level ranking information more accurately represents the user's current capability level within the tested knowledge domain than the first set of knowledge structure tags, and thus the second set of topics invoked according to the second set of knowledge structure tags may provide better, user-personalized test results.
In some embodiments, between step S2 and step S3 further comprising: step S21, a user profile is created and stored in the index module 42, where the user profile includes all user data generated during the performance test classification process using the performance test classification method of the present application, including a first group of questions, first test data, performance level classification information, a second group of questions, answer schedule, test time, and the like. Any data or information in the same user profile stored in the indexing module 42 may be invoked at a different user terminal 45.
In some embodiments, after step S6, further comprising: in step S7, the user terminal 45 collects second test data generated by the user in response to the second set of questions and transmits the second test data to the analysis module 44, where the second test data includes answers of the second set of questions input by the user from the user terminal 45, and the analysis module 44 analyzes the second test data, returns the analysis result to the user terminal 45, and stores the analysis result in the user profile of the indexing module 42.
In some embodiments, after step S7, further comprising: in step S8, returning to step S3, the analysis result of the second test data stored in the user profile is called from the index module 42 by the control module 43 as the initial test information, and steps S3 to S7 are performed.
In some embodiments, the user terminal 45 further comprises a biometric acquisition device, and the test data further comprises biometric data of the user generated by the biometric acquisition device. In some embodiments, the biometric acquisition device comprises an infrared emitting device and an infrared receiving device, and the acquired biometric data is the body surface temperature change and/or the heart rate change of the user during the test. In some embodiments, the biometric acquisition device further comprises a skin electrical sensor, and the acquired biometric data is a change in skin resistance of the user during the test.
Referring to fig. 4, fig. 4 is a schematic diagram of one embodiment of a competency testing and grading apparatus provided herein. The capability test grading device comprises: the question bank comprises a question bank storage module 41, an index module 42, a control module 43, an analysis module 44 and a user terminal 45; wherein, the question storage module stores the capability test question bank and transmits the questions in the capability test question bank to the user terminal 45; the index module 42 stores the index of the capability test question bank and calls the question in the question storage module according to the index; the control module 43 comprises a processor and a memory which is in communication connection with the processor, instructions for executing the capability test classification are stored in the memory, the control module 43 is in bidirectional communication with the user terminal 45, the index module 42 and the analysis module 44 respectively, namely the control module 43 outputs the instructions and data to the user terminal 45, the index module 42 and the analysis module 44 and receives the data and information from the user terminal 45, the index module 42 and the analysis module 44; the analysis module 44 analyzes the user data and comprises a capability level hierarchical deep neural network 441 for generating capability level hierarchical information of the user, wherein the capability level hierarchical deep neural network 441 comprises a convolutional layer, a pooling layer and a full-link layer; the user terminal 45 collects user data and transmits the user data to the analysis module 44. The capacity test grading device is used for executing the capacity test grading method.
In some embodiments, the question bank storage module 41 is a server that stores the capability test question bank. In some embodiments, the indexing module 42 is a server that stores the index. In some embodiments, the user terminal 45 further comprises a biometric acquisition device. In some embodiments, the biometric acquisition device includes an infrared emitting device and an infrared receiving device for measuring changes in body surface temperature and/or changes in heart rate of the user during the test. In some embodiments, the biometric acquisition device further comprises a skin electrical sensor for measuring changes in skin resistance of the user during the test.
It should be understood that the various steps recited in the method embodiments of the present application may be performed in a different order, and/or performed in parallel. Moreover, method embodiments may include additional steps and/or omit performing the recited steps. The scope of the present application is not limited in this respect.
In the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the application may be practiced without these specific details. In some embodiments, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.
While preferred embodiments of the present application have been shown and described herein, it will be readily understood by those skilled in the art that these embodiments are provided by way of example only. Numerous variations, changes, and substitutions will now occur to those skilled in the art without departing from the application. It should be understood that various alternatives to the embodiments of the application described herein may be employed in practicing the application. It is intended that the following claims define the scope of the application and that methods and structures within the scope of these claims and their equivalents be covered thereby.
Claims (10)
1. A capability test ranking method, comprising:
s1, establishing a capability test question bank, wherein the capability test question bank comprises questions, knowledge structure labels of the questions and question numbers, and storing the capability test question bank in a question bank storage module (41);
s2, generating an index of the ability test question bank and storing the index in an index module (42), wherein the index connects all the knowledge structure labels contained in the question bank according to the correlation among the knowledge structure labels to form a structured system, and the question number of the question containing the knowledge structure label is recorded under each knowledge structure label;
s3, generating initial test information, generating a first group of knowledge structure labels by an analysis module (44) according to the initial test information and transmitting the first group of knowledge structure labels to an index module (42), generating a first group of topic numbers with the first group of knowledge structure labels by the index module (42), calling the first group of topics with the first group of topic numbers from the topic bank storage module (41), and transmitting the first group of topics to the user terminal (45);
s4, the user terminal (45) collects first test data generated by the user in response to the first group of questions and transmits the first test data to an analysis module (44), and the first test data comprises answers of the first group of questions input by the user from the user terminal (45);
s5, the analysis module (44) generates a user data matrix map by using the first test data, inputs the user data matrix map into a competence level hierarchical deep neural network (441) in the analysis module (44), and carries out classification identification on the user data matrix map, wherein the result of the classification identification contains competence level hierarchical information of the user;
s6, the analysis module (44) generates a second group of knowledge structure labels according to the ability level grading information, transmits the second group of knowledge structure labels to the index module (42), the index module (42) generates a second group of topic numbers with the second group of knowledge structure labels, and calls the second group of topics with the second group of topic numbers from the topic library storage module (41) to transmit to the user terminal (45).
2. The method of claim 1, wherein the knowledge structure tags comprise a knowledge level tag, a knowledge module tag, and a knowledge point number tag, wherein the knowledge level tag divides the capability test problem base into at least two knowledge levels, each knowledge level having at least one knowledge point therein; the knowledge module label divides the capability test question bank into at least two knowledge modules, and each knowledge module is provided with at least one knowledge point; the index comprises a structured system formed by connecting at least two correlation relations among the knowledge module labels, the knowledge level labels and the knowledge point number labels.
3. The method of claim 2, wherein the initial test information comprises a user-defined knowledge module and a knowledge level input from the user terminal (45), and wherein the first set of knowledge structure tags are the knowledge module tags and the knowledge level tags corresponding to the user-defined knowledge module and knowledge level.
4. The method of claim 2, wherein the initial test information is default initial test information, and wherein the first set of knowledge structure tags are the knowledge module tags and the knowledge level tags corresponding to default knowledge modules and default knowledge levels.
5. The method according to claim 1 or 2, wherein the user data matrix map is a two-dimensional data matrix arranged in accordance with the knowledge structure labels, the competency level hierarchical deep neural network (441) performing a two-dimensional convolution operation on the user data matrix map.
6. The method according to claim 2, wherein the user data matrix map is a three-dimensional data matrix arranged according to the knowledge structure labels, and the competency level hierarchical deep neural network (441) performs a three-dimensional convolution operation on the user matrix map.
7. The method of claim 1, further comprising S21 between S2 and S3, and further comprising S7 after S6:
s21, creating and storing in an indexing module (42) a user profile containing all user data generated during said capability test ranking, any data or information in the same user profile stored in said indexing module (42) being retrievable at a different said user terminal (45);
s7, the user terminal (45) collects second test data generated by the user answering the second group of questions and transmits the second test data to the analysis module (44), the second test data comprises answers of the second group of questions input by the user from the user terminal (45), the analysis module (44) analyzes the second test data, and analysis results are returned to the user terminal (45) and stored in the user profile of the index module (42).
8. The method according to claim 1, characterized in that the user terminal (45) further comprises a biometric acquisition device, the test data further comprising biometric data of the user generated by the biometric acquisition device.
9. A capability test rating device comprising:
an item bank storage module (41);
an indexing module (42);
a control module (43);
an analysis module (44); and
a user terminal (45);
wherein, the question storage module stores a capability test question bank and transmits the questions in the capability test question bank to the user terminal (45); the index module (42) stores the index of the capability test question bank and calls the question in the question storage module according to the index; the control module (43) comprises a processor and a memory which is in communication connection with the processor and in which instructions for performing a capability test hierarchy are stored, the control module (43) being in respective bidirectional communication with the user terminal (45), the indexing module (42) and the analysis module (44); the analysis module (44) analyzes user data and comprises a capability level hierarchical deep neural network (441) used for generating capability level hierarchical information of a user, wherein the capability level hierarchical deep neural network (441) comprises a convolutional layer, a pooling layer and a full-link layer; the user terminal (45) collects user data and transmits the user data to an analysis module (44).
10. The device according to claim 9, characterized in that the user terminal (45) further comprises biometric acquisition means.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110758961.9A CN113421175A (en) | 2021-07-05 | 2021-07-05 | Capability test grading method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110758961.9A CN113421175A (en) | 2021-07-05 | 2021-07-05 | Capability test grading method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113421175A true CN113421175A (en) | 2021-09-21 |
Family
ID=77721529
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110758961.9A Pending CN113421175A (en) | 2021-07-05 | 2021-07-05 | Capability test grading method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113421175A (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140025612A1 (en) * | 2011-03-28 | 2014-01-23 | Ivaylo Popov | Adaptive cognitive method |
| CN103942993A (en) * | 2014-03-17 | 2014-07-23 | 深圳市承儒科技有限公司 | Self-adaptive online assessment system and method based on IRT |
| CN105373703A (en) * | 2015-12-02 | 2016-03-02 | 武汉慧人信息科技有限公司 | Self-adaptive capacity testing system based on forgetting curve |
| CN106779166A (en) * | 2016-11-23 | 2017-05-31 | 北京师范大学 | Grasp the forecasting system and method for state in a kind of knowledge point based on data-driven |
| CN107157499A (en) * | 2017-05-27 | 2017-09-15 | 西南交通大学 | A kind of method of testing and system of high ferro dispatcher prospective memory ability |
| CN109086837A (en) * | 2018-10-24 | 2018-12-25 | 高嵩 | User property classification method, storage medium, device and electronic equipment based on convolutional neural networks |
| CN110491218A (en) * | 2019-08-21 | 2019-11-22 | 北京大米科技有限公司 | A kind of online teaching exchange method, device, storage medium and electronic equipment |
| CN110533563A (en) * | 2019-09-05 | 2019-12-03 | 河南师范大学 | A kind of the layer level label and intelligent Application method of exam pool |
| CN111813921A (en) * | 2020-08-20 | 2020-10-23 | 浙江学海教育科技有限公司 | Topic recommendation method, electronic device and computer-readable storage medium |
-
2021
- 2021-07-05 CN CN202110758961.9A patent/CN113421175A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140025612A1 (en) * | 2011-03-28 | 2014-01-23 | Ivaylo Popov | Adaptive cognitive method |
| CN103942993A (en) * | 2014-03-17 | 2014-07-23 | 深圳市承儒科技有限公司 | Self-adaptive online assessment system and method based on IRT |
| CN105373703A (en) * | 2015-12-02 | 2016-03-02 | 武汉慧人信息科技有限公司 | Self-adaptive capacity testing system based on forgetting curve |
| CN106779166A (en) * | 2016-11-23 | 2017-05-31 | 北京师范大学 | Grasp the forecasting system and method for state in a kind of knowledge point based on data-driven |
| CN107157499A (en) * | 2017-05-27 | 2017-09-15 | 西南交通大学 | A kind of method of testing and system of high ferro dispatcher prospective memory ability |
| CN109086837A (en) * | 2018-10-24 | 2018-12-25 | 高嵩 | User property classification method, storage medium, device and electronic equipment based on convolutional neural networks |
| CN110491218A (en) * | 2019-08-21 | 2019-11-22 | 北京大米科技有限公司 | A kind of online teaching exchange method, device, storage medium and electronic equipment |
| CN110533563A (en) * | 2019-09-05 | 2019-12-03 | 河南师范大学 | A kind of the layer level label and intelligent Application method of exam pool |
| CN111813921A (en) * | 2020-08-20 | 2020-10-23 | 浙江学海教育科技有限公司 | Topic recommendation method, electronic device and computer-readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110516085A (en) | The mutual search method of image text based on two-way attention | |
| CN110674292B (en) | Man-machine interaction method, device, equipment and medium | |
| CN110059716B (en) | Construction of CNN-LSTM-SVM network model and MOOC conjugate prediction method | |
| CN110472090A (en) | Image search method and relevant apparatus, storage medium based on semantic label | |
| CN108228674B (en) | DKT-based information processing method and device | |
| CN111652291A (en) | Method for establishing student growth portrait based on group sparse fusion hospital big data | |
| Akagunduz et al. | Defining image memorability using the visual memory schema | |
| CN106354701A (en) | Chinese character processing method and device | |
| CN109189922B (en) | Comment evaluation model training method and device | |
| CN109614480A (en) | A kind of generation method and device of the autoabstract based on production confrontation network | |
| CN110222838A (en) | Deep neural network and its training method, device, electronic equipment and storage medium | |
| CN110197213A (en) | Image matching method, device and equipment neural network based | |
| Kuzma et al. | Predicting user’s preferences using neural networks and psychology models | |
| Qayyum et al. | Ios mobile application for food and location image prediction using convolutional neural networks | |
| CN113421175A (en) | Capability test grading method and device | |
| CN116228361A (en) | Course recommendation method, device, equipment and storage medium based on feature matching | |
| CN113821610A (en) | Information matching method, device, equipment and storage medium | |
| CN115033700A (en) | Cross-domain emotion analysis method, device and equipment based on mutual learning network | |
| CN113919983A (en) | Test question portrait method, device, electronic equipment and storage medium | |
| CN114625986A (en) | Method, device and equipment for sorting search results and storage medium | |
| CN114443956A (en) | Content recommendation method and related equipment | |
| CN114329181A (en) | Question recommendation method and device and electronic equipment | |
| CN110334353A (en) | Analysis method, device, equipment and the storage medium of word order recognition performance | |
| CN114328797B (en) | Content search method, device, electronic apparatus, storage medium, and program product | |
| CN117040942B (en) | Network security test evaluation method and system based on deep learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210921 |
|
| RJ01 | Rejection of invention patent application after publication |