CN114238546A - Self-adaptive cognitive diagnosis test equipment and method based on cavity convolution - Google Patents
Self-adaptive cognitive diagnosis test equipment and method based on cavity convolution Download PDFInfo
- Publication number
- CN114238546A CN114238546A CN202210170060.2A CN202210170060A CN114238546A CN 114238546 A CN114238546 A CN 114238546A CN 202210170060 A CN202210170060 A CN 202210170060A CN 114238546 A CN114238546 A CN 114238546A
- Authority
- CN
- China
- Prior art keywords
- test
- module
- attention
- convolution
- test question
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- Biophysics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Animal Behavior & Ethology (AREA)
- Biomedical Technology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a self-adaptive cognitive diagnosis test device and a self-adaptive cognitive diagnosis test method based on cavity convolution, wherein the test device comprises a test board, and a micro-control module, a login system and a Redis storage system are arranged through the test board; the micro-control module comprises a hole convolution module based on an attention mechanism, a multi-channel hole convolution module based on the attention mechanism, an attention residual block and a multi-channel feature extraction module, and a hole convolution neural network based on the attention mechanism is designed. The invention uses the hole convolution to extract the characteristic information of a single test question from a plurality of different user labels, extracts the information of the question bank by using the attention residual block, and fully utilizes the correlation among the characteristic information of the single test question, so that the test is more accurate.
Description
Technical Field
The invention belongs to the technical field of computer intelligence, and particularly relates to self-adaptive cognitive diagnosis testing equipment and method based on cavity convolution.
Background
With the improvement of the technological level, information technology, multimedia technology and the like are introduced into the test field, and a new implementation mode for testing also appears, namely, a computer is used as a tool for implementing test evaluation.
Computer-Based Testing (CBT) is a test carrier using a Computer instead of a conventional paper pen, and according to a classical test theory, all testees answer completely identical test questions with the same quantity and the same questions, but without considering the capability difference of different testees.
Different from CBT, the idea of Computer Adaptive Testing (CAT) is to select the most suitable test Item for each subject by using Item Response Theory (IRT), record the reaction of the subject on the Item, and achieve more accurate estimation of the capability of the subject.
The Cognitive Diagnosis Computerized Adaptive Testing (CD-CAT) is a combination of Cognitive Diagnosis evaluation and Computerized Adaptive Testing, and has the characteristics of Cognitive Diagnosis and Adaptive Testing. The existing Cognitive Diagnosis Computerized Adaptive Test (CD-CAT) mainly focuses on language ability tests, and there is no Cognitive Diagnosis Computerized Adaptive Test for testing scientific potential of users.
The soft and hard integrated test mainly comprises dermatoglyph test, saliva test paper and the like, but the hand photography, the craniography, the handwriting, the neuro-linguistic programming, the constellation, the psychology control and the like are pseudo psychology. Pseudopsychology refers to those systems that look psychologically but do not have any evidence of this.
The essence of the saliva test paper is that DNA genes are used as test items, but in reality, no "innate genes" exist, no gene can completely determine the level of intelligence quotient, the functions of different genes are different and mutually influenced, and the complexity is aggravated by the growth environment and acquired education guidance of children. Such as a story of impaired sex. The series of soft and hard integrated equipment is controversial in scientific basis, test safety and the like.
The traditional test content becomes a fixed template basically after years of development, and a part of test questions do not accord with the current test population, so that the test result is not accurate.
In recent years, algorithms based on deep learning have been widely proposed and have achieved good results. Dong et al first applies deep learning to Super-Resolution reconstruction, and proposes a Super-Resolution reconstruction algorithm (SRCNN) based on a Convolutional Neural Network, and the SRCNN algorithm uses a 3-layer Convolutional Network to realize end-to-end learning from a low-Resolution image to a high-Resolution image, so that the image reconstruction effect is greatly improved compared with the conventional algorithm. Kim et al propose a Deep convolutional Network-based image Super-Resolution reconstruction algorithm (VDSR), which applies a residual structure to Super-Resolution reconstruction, so that the number of layers of a convolutional neural Network is deepened to 20 layers, more feature information of an image can be extracted, and the image reconstruction effect is greatly improved. Lai et al propose a Laplacian Pyramid structure-based Super-Resolution reconstruction algorithm (LapSRN) that performs reconstruction by way of gradual upsampling. Ledig et al propose a Super-Resolution reconstruction algorithm (SRGAN) based on a generated countermeasure Network, which applies a generated countermeasure Network structure to Super-Resolution reconstruction and makes the reconstruction effect more realistic by using sensing loss and countermeasure loss as loss functions. Lim et al propose an Enhanced depth Residual Network Image Super-Resolution reconstruction algorithm (EDSR), which deletes redundant blocks in an original Residual block, uses more convolution layers to extract richer feature information, and thus obtains better Image reconstruction performance.
Although the above methods have good effects, the correlation between feature information cannot be fully utilized to extract more feature information.
Spatial convolution is a way of information processing, i.e. the enhancement (or reduction) of the spatial frequency of an image is achieved by processing of neighboring image elements around each image element. There are two steps to spatially convolve an image. First, a moving window is established that contains a series of correlation coefficients or weighting factors. Then, the window is moved over the whole image, and the brightness value of the central pixel of the window is replaced by the sum (or average value) obtained by multiplying the brightness value of each pixel covered by the window by the corresponding correlation coefficient or weight, so as to obtain a new image. The testing method is combined with the spatial convolution to construct, so that more characteristic information of a single test question is extracted, and the characteristic information of different channels is fused, so that the test is more complete and accurate.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides self-adaptive cognitive diagnosis test equipment and method based on hole convolution.
In order to realize the technical scheme, the invention discloses a self-adaptive cognitive diagnosis testing method based on hole convolution, which comprises the following steps of:
step 1: selecting a test question bank, building a user label, and training the test question bank;
step 2: selecting a training data set and a test data set, enhancing a training data set test question library, and expanding the training data set test question library;
and step 3: carrying out 1/N ratio down-sampling treatment on the training data set test question library obtained in the step 2 to obtain corresponding individual test questions, wherein N is a scaling factor;
and 4, step 4: cutting the training data set test question library obtained in the step 2 into H multiplied by W test question groups, and cutting the single test questions obtained in the step 3 into H/N multiplied by W/N test question groups;
and 5: respectively taking the two test question groups obtained in the step 4 as a complete test question library and an independent test question sample to generate a training data set file of HDF 5;
step 6: designing a void convolutional neural network based on an attention mechanism, which comprises the following specific steps:
6.1: design of void convolution module based on attention mechanism
The attention mechanism-based cavity convolution module is formed by connecting a cavity convolution layer and an attention block end to end, wherein the output end of the cavity convolution layer is connected with the input end of the attention block, and meanwhile, the output end of the cavity convolution layer is multiplied by the output end of the attention block to form the output of the attention mechanism-based cavity convolution module;
6.2: design multichannel is based on hole convolution module of attention mechanism
The multichannel attention-based cavity convolution module consists of an attention-based cavity convolution module and a feature fusion module which have m different cavity convolution coefficients d1, d2 and … dm, wherein the m cavity convolution modules are connected in parallel, and the output ends of the m cavity convolution modules are connected to the feature fusion module;
6.3: designing an attention residual block
The attention residual block is formed by connecting a residual block and an attention block end to end, the output end of the residual block is connected with the input end of the attention block, and meanwhile, the output end of the residual block is multiplied by the output end of the attention block to form the output of the attention residual block;
6.4: constructing a multi-channel feature extraction Module
The multi-channel feature extraction module is formed by connecting a multi-channel attention-based hole convolution module and an attention residual block end to end;
6.5: designing a void convolutional neural network based on an attention mechanism
The cavity convolution neural network based on the attention mechanism is composed of four parts, namely an input module, a deep feature extraction module, an up-sampling module and an output module, wherein:
the input module and the output module are both composed of convolution layers with convolution kernel size of 3 multiplied by 3, and the deep layer feature extraction module is composed of n step 6.4 multi-channel feature extraction modules and a convolution layer which are connected in series and connected in a residual error mode;
the up-sampling module consists of a test result convolution layer;
and 7: training a void convolutional neural network based on an attention mechanism, which is as follows:
7.1: setting a loss function, and estimating network parameters by minimizing the loss value of the reconstructed test question library and the corresponding individual test questions;
7.2: selecting an optimization algorithm, and performing iterative training on the hollow convolution neural network;
7.3: selecting a coincidence reconstruction evaluation index of test question test contents to objectively evaluate the reconstruction performance of the cavity convolution neural network model based on the attention mechanism, wherein the coincidence represents a test purpose or a test result;
7.4: setting the values of m and d1, d2 and … dm of the multi-channel attention-based hole convolution module in the step 6.2;
7.5: setting training parameters including learning rate, iteration times and batch training sample values of training;
7.6: training a hole convolution neural network based on an attention mechanism by using the HDF5 training data set file generated in the step 5 according to the parameters set in the step 6.5 to generate a network model;
7.7: testing the network model obtained in the step 7.6 by using a test data set, and recording a reconstruction performance index value of the test library; then returning to step 7.4, setting different values of m and d1, d2 and … dm, continuing training and testing, finally, storing a group of values of m and d1, d2 and … dm corresponding to the reconstructed performance index value of the highest test question library, and obtaining a final cavity convolution neural network model based on the attention mechanism;
and 8: inputting the single test question into the cavity convolution neural network model based on the attention mechanism, and outputting to obtain a reconstructed complete question library;
and step 9: and matching the user label with the reconstructed test question library, and finishing answering by the user in the test question library to obtain a test result.
Further: in the step 1, a test question bank is further constructed, specifically: 1. extracting a first knowledge point from the teaching outline by the test question bank; 2. extracting a second knowledge point from the book by the question bank, and integrating the first knowledge point and the second knowledge point into a knowledge map; 3. and extracting a third knowledge point from the network resources by the test question bank, and integrating the third knowledge point and the knowledge map into a subject knowledge map.
Further: in step 1, the user tag is built as follows: 1. analyzing user data, including analyzing user basic information data, test times data, test time data and interest field data, wherein the basic information data includes but is not limited to user name, gender and user image; 2. collecting the user basic information data, the test frequency data and the test time data into original data; 3. performing data statistical analysis on the original data, and collecting the original data into a fact label; 4. modeling and analyzing the fact labels and then collecting the fact labels into model labels; 5. and performing model prediction on the model tags and then collecting the model tags to the prediction tags.
Further: the step 9 specifically includes the following steps:
s91: the user replies the reconstructed complete test question bank;
s92: evaluating the user response result, if the evaluation result meets the standard, entering the step S93, and if the evaluation result does not meet the standard, returning to the step S91;
s93: and finishing the test and outputting a test result.
The invention also provides self-adaptive cognitive diagnosis test equipment based on the cavity convolution, which comprises the following steps: including the testboard, the testboard is equipped with micro-control module, login system, two display screens, gesture response module and Redis memory system, promptly:
the micro-control module comprises a micro-control chip and is used for analyzing and processing data;
the login system is used for logging in a user and starting the test board;
the double display screens are used for displaying test questions and user operation;
the gesture sensing module is used for identifying a gesture signal;
the Redis memory system is used for data storage and data tracking; the data comprises user data, test result data and a test question bank after the void convolution; the output end of the micro control module is respectively connected with the input ends of the Redis storage system and the double display screens, and the output ends of the gesture sensing module and the login system are respectively connected with the input end of the micro control module.
Compared with the prior art, the invention has the beneficial effects that:
(1) compared with the traditional convolutional layer, the cavity convolution module based on the attention mechanism has larger receptive field under the condition of not increasing excessive parameters and calculated amount, can extract more characteristic information of a single test question, and fully extracts the information of a test question library by using the attention block;
(2) the multi-channel attention-based cavity convolution module is provided with m different cavity convolution coefficients and is used for extracting the characteristics of a single test question, and the correlation among the characteristic information of the single test question is fully utilized to extract more characteristic information. Meanwhile, the multi-channel attention-based hole convolution module can fuse the characteristic information of different channels;
(3) the attention residual block can perform deep extraction on the features obtained by the multi-channel attention-based cavity convolution module to obtain more test question bank information, so that the reconstructed complete test question bank has richer test question types and test directions, and the test is more complete.
Drawings
FIG. 1 is a schematic diagram of a three-dimensional 3D visualization model of the void convolution of the present invention;
FIG. 2 is a schematic diagram of the hole residual convolution block of the present invention;
FIG. 3 is a diagram of knowledge graph construction according to the present invention;
FIG. 4 is a user tag modeling diagram of the present invention;
FIG. 5 is a schematic view of a storage structure according to the present invention.
The labels in the figure are: 101-a hole convolution module; 102-attention block; 103-mobile terminal; 104-Java Internet System; 105 — NoSQL database; 106-database.
Detailed Description
The technical solutions of the present invention will be described clearly and completely with reference to the following embodiments, and it should be understood that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without inventive effort based on the embodiments of the present invention, are within the scope of the present invention.
Referring to fig. 1-5, fig. 1 is a schematic diagram of a three-dimensional 3D visualization model of the void convolution of the present invention, fig. 2 is a schematic diagram of a void residual convolution block of the present invention, fig. 3 is a diagram of a knowledge graph construction of the present invention, fig. 4 is a user tag modeling diagram of the present invention, and fig. 5 is a schematic diagram of a storage structure of the present invention.
The invention provides a self-adaptive cognitive diagnosis test device based on cavity convolution, which comprises a test board, wherein the test board is provided with a micro-control module, a login system, a double display screen, a gesture sensing module and a Redis storage system, namely:
the micro-control module comprises a micro-control chip and is used for analyzing and processing data;
the login system is used for logging in a user and starting the test board;
the double display screens are used for displaying test questions and user operation;
the gesture sensing module is used for identifying a gesture signal;
the Redis memory system is used for data storage and data tracking; the data comprises user data, test result data and a test question bank after the void convolution; the output end of the micro control module is respectively connected with the input ends of the Redis storage system and the double display screens, and the output ends of the gesture sensing module and the login system are respectively connected with the input end of the micro control module.
The test method for the operation of the test bench specifically comprises the following steps:
step 1: selecting a test question bank, building a user label, and training the test question bank;
step 2: selecting a training data set and a test data set, enhancing a training data set test question library, and expanding the training data set test question library;
and step 3: carrying out 1/N ratio down-sampling treatment on the training data set test question library obtained in the step 2 to obtain corresponding individual test questions, wherein N is a scaling factor;
and 4, step 4: cutting the training data set test question library obtained in the step 2 into H multiplied by W test question groups, and cutting the single test questions obtained in the step 3 into H/N multiplied by W/N test question groups;
and 5: respectively taking the two test question groups obtained in the step 4 as a complete test question library and an independent test question sample to generate a training data set file of HDF 5;
step 6: designing a void convolutional neural network based on an attention mechanism, which comprises the following specific steps:
6.1: design of void convolution module based on attention mechanism
The attention mechanism-based cavity convolution module is formed by connecting a cavity convolution layer and an attention block end to end, wherein the output end of the cavity convolution layer is connected with the input end of the attention block, and meanwhile, the output end of the cavity convolution layer is multiplied by the output end of the attention block to form the output of the attention mechanism-based cavity convolution module;
6.2: design multichannel is based on hole convolution module of attention mechanism
The multichannel attention-based cavity convolution module consists of an attention-based cavity convolution module and a feature fusion module which have m different cavity convolution coefficients d1, d2 and … dm, wherein the m cavity convolution modules are connected in parallel, and the output ends of the m cavity convolution modules are connected to the feature fusion module;
6.3: designing an attention residual block
The attention residual block is formed by connecting a residual block and an attention block end to end, the output end of the residual block is connected with the input end of the attention block, and meanwhile, the output end of the residual block is multiplied by the output end of the attention block to form the output of the attention residual block;
6.4: constructing a multi-channel feature extraction Module
The multi-channel feature extraction module is formed by connecting a multi-channel attention-based hole convolution module and an attention residual block end to end;
6.5: designing a void convolutional neural network based on an attention mechanism
The cavity convolution neural network based on the attention mechanism is composed of four parts, namely an input module, a deep feature extraction module, an up-sampling module and an output module, wherein:
the input module and the output module are both composed of convolution layers with convolution kernel size of 3 multiplied by 3, and the deep layer feature extraction module is composed of n step 6.4 multi-channel feature extraction modules and a convolution layer which are connected in series and connected in a residual error mode;
the up-sampling module consists of a test result convolution layer;
and 7: training a void convolutional neural network based on an attention mechanism, which is as follows:
7.1: setting a loss function, and estimating network parameters by minimizing the loss value of the reconstructed test question library and the corresponding individual test questions;
7.2: selecting an optimization algorithm, and performing iterative training on the hollow convolution neural network;
7.3: selecting a coincidence reconstruction evaluation index of test question test contents to objectively evaluate the reconstruction performance of the cavity convolution neural network model based on the attention mechanism, wherein the coincidence represents a test purpose or a test result;
7.4: setting the values of m and d1, d2 and … dm of the multi-channel attention-based hole convolution module in the step 6.2;
7.5: setting training parameters including learning rate, iteration times and batch training sample values of training;
7.6: training a hole convolution neural network based on an attention mechanism by using the HDF5 training data set file generated in the step 5 according to the parameters set in the step 6.5 to generate a network model;
7.7: testing the network model obtained in the step 7.6 by using a test data set, and recording a reconstruction performance index value of the test library; then returning to step 7.4, setting different values of m and d1, d2 and … dm, continuing training and testing, finally, storing a group of values of m and d1, d2 and … dm corresponding to the reconstructed performance index value of the highest test question library, and obtaining a final cavity convolution neural network model based on the attention mechanism;
and 8: inputting the single test question into the cavity convolution neural network model based on the attention mechanism, and outputting to obtain a reconstructed complete question library;
and step 9: and matching the user label with the reconstructed test question library, and finishing answering by the user in the test question library to obtain a test result.
In addition, in the step 1, a question bank is further constructed, specifically: 1. extracting a first knowledge point from the teaching outline by the test question bank; 2. extracting a second knowledge point from the book by the question bank, and integrating the first knowledge point and the second knowledge point into a knowledge map; 3. and extracting a third knowledge point from the network resources by the test question bank, and integrating the third knowledge point and the knowledge map into a subject knowledge map. In step 1, the user tag is built as follows: 1. analyzing user data, including analyzing user basic information data, test times data, test time data and interest field data, wherein the basic information data includes but is not limited to user name, gender and user image; 2. collecting the user basic information data, the test frequency data and the test time data into original data; 3. performing data statistical analysis on the original data, and collecting the original data into a fact label; 4. modeling and analyzing the fact labels and then collecting the fact labels into model labels; 5. and performing model prediction on the model tags and then collecting the model tags to the prediction tags.
The step 9 specifically includes the following steps:
s91: the user replies the reconstructed complete test question bank;
s92: evaluating the user response result, if the evaluation result meets the standard, entering the step S93, and if the evaluation result does not meet the standard, returning to the step S91;
s93: and finishing the test and outputting a test result.
The foregoing is merely a preferred embodiment of the invention, it is to be understood that the invention is not limited to the forms disclosed herein, but is not intended to be exhaustive or to limit the invention to other embodiments, and to various other combinations, modifications, and environments and may be modified within the scope of the inventive concept as expressed herein, by the teachings or the skill or knowledge of the relevant art. And that modifications and variations may be effected by those skilled in the art without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (5)
1. A self-adaptive cognitive diagnosis test method based on hole convolution is characterized by comprising the following steps: the method comprises the following steps:
step 1: selecting a test question bank, building a user label, and training the test question bank;
step 2: selecting a training data set and a test data set, enhancing a training data set test question library, and expanding the training data set test question library;
and step 3: carrying out 1/N ratio down-sampling treatment on the training data set test question library obtained in the step 2 to obtain corresponding individual test questions, wherein N is a scaling factor;
and 4, step 4: cutting the training data set test question library obtained in the step 2 into H multiplied by W test question groups, and cutting the single test questions obtained in the step 3 into H/N multiplied by W/N test question groups;
and 5: respectively taking the two test question groups obtained in the step 4 as a complete test question library and an independent test question sample to generate a training data set file of HDF 5;
step 6: designing a void convolutional neural network based on an attention mechanism, which comprises the following specific steps:
6.1: design of void convolution module based on attention mechanism
The attention mechanism-based cavity convolution module is formed by connecting a cavity convolution layer and an attention block end to end, wherein the output end of the cavity convolution layer is connected with the input end of the attention block, and meanwhile, the output end of the cavity convolution layer is multiplied by the output end of the attention block to form the output of the attention mechanism-based cavity convolution module;
6.2: design multichannel is based on hole convolution module of attention mechanism
The multichannel attention-based cavity convolution module consists of an attention-based cavity convolution module and a feature fusion module which have m different cavity convolution coefficients d1, d2 and … dm, wherein the m cavity convolution modules are connected in parallel, and the output ends of the m cavity convolution modules are connected to the feature fusion module;
6.3: designing an attention residual block
The attention residual block is formed by connecting a residual block and an attention block end to end, the output end of the residual block is connected with the input end of the attention block, and meanwhile, the output end of the residual block is multiplied by the output end of the attention block to form the output of the attention residual block;
6.4: constructing a multi-channel feature extraction Module
The multi-channel feature extraction module is formed by connecting a multi-channel attention-based hole convolution module and an attention residual block end to end;
6.5: designing a void convolutional neural network based on an attention mechanism
The cavity convolution neural network based on the attention mechanism is composed of four parts, namely an input module, a deep feature extraction module, an up-sampling module and an output module, wherein:
the input module and the output module are both composed of convolution layers with convolution kernel size of 3 multiplied by 3, and the deep layer feature extraction module is composed of n step 6.4 multi-channel feature extraction modules and a convolution layer which are connected in series and connected in a residual error mode;
the up-sampling module consists of a test result convolution layer;
and 7: training a void convolutional neural network based on an attention mechanism, which is as follows:
7.1: setting a loss function, and estimating network parameters by minimizing the loss value of the reconstructed test question library and the corresponding individual test questions;
7.2: selecting an optimization algorithm, and performing iterative training on the hollow convolution neural network;
7.3: selecting a coincidence reconstruction evaluation index of test question test contents to objectively evaluate the reconstruction performance of the cavity convolution neural network model based on the attention mechanism, wherein the coincidence represents a test purpose or a test result;
7.4: setting the values of m and d1, d2 and … dm of the multi-channel attention-based hole convolution module in the step 6.2;
7.5: setting training parameters including learning rate, iteration times and batch training sample values of training;
7.6: training a hole convolution neural network based on an attention mechanism by using the HDF5 training data set file generated in the step 5 according to the parameters set in the step 6.5 to generate a network model;
7.7: testing the network model obtained in the step 7.6 by using a test data set, and recording a reconstruction performance index value of the test library; then returning to step 7.4, setting different values of m and d1, d2 and … dm, continuing training and testing, finally, storing a group of values of m and d1, d2 and … dm corresponding to the reconstructed performance index value of the highest test question library, and obtaining a final cavity convolution neural network model based on the attention mechanism;
and 8: inputting the single test question into the cavity convolution neural network model based on the attention mechanism, and outputting to obtain a reconstructed complete question library;
and step 9: and matching the user label with the reconstructed test question library, and finishing answering by the user in the test question library to obtain a test result.
2. The adaptive cognitive diagnostic test method based on hole convolution according to claim 1, characterized in that: in the step 1, a test question bank is further constructed, specifically: 1. extracting a first knowledge point from the teaching outline by the test question bank; 2. extracting a second knowledge point from the book by the question bank, and integrating the first knowledge point and the second knowledge point into a knowledge map; 3. and extracting a third knowledge point from the network resources by the test question bank, and integrating the third knowledge point and the knowledge map into a subject knowledge map.
3. The adaptive cognitive diagnostic test method based on hole convolution according to claim 1, characterized in that: in step 1, the user tag is built as follows: 1. analyzing user data, including analyzing user basic information data, test times data, test time data and interest field data, wherein the basic information data includes but is not limited to user name, gender and user image; 2. collecting the user basic information data, the test frequency data and the test time data into original data; 3. performing data statistical analysis on the original data, and collecting the original data into a fact label; 4. modeling and analyzing the fact labels and then collecting the fact labels into model labels; 5. and performing model prediction on the model tags and then collecting the model tags to the prediction tags.
4. The adaptive cognitive diagnostic test method based on hole convolution according to claim 1, characterized in that: the step 9 specifically includes the following steps:
s91: the user replies the reconstructed complete test question bank;
s92: evaluating the user response result, if the evaluation result meets the standard, entering the step S93, and if the evaluation result does not meet the standard, returning to the step S91;
s93: and finishing the test and outputting a test result.
5. An adaptive cognitive diagnosis test equipment based on cavity convolution is characterized in that: including the testboard, the testboard is equipped with micro-control module, login system, two display screens, gesture response module and Redis memory system, promptly:
the micro-control module comprises a micro-control chip and is used for analyzing and processing data;
the login system is used for logging in a user and starting the test board;
the double display screens are used for displaying test questions and user operation;
the gesture sensing module is used for identifying a gesture signal;
the Redis memory system is used for data storage and data tracking; the data comprises user data, test result data and a test question bank after the void convolution; the output end of the micro control module is respectively connected with the input ends of the Redis storage system and the double display screens, and the output ends of the gesture sensing module and the login system are respectively connected with the input end of the micro control module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210170060.2A CN114238546A (en) | 2022-02-24 | 2022-02-24 | Self-adaptive cognitive diagnosis test equipment and method based on cavity convolution |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210170060.2A CN114238546A (en) | 2022-02-24 | 2022-02-24 | Self-adaptive cognitive diagnosis test equipment and method based on cavity convolution |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114238546A true CN114238546A (en) | 2022-03-25 |
Family
ID=80747952
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210170060.2A Pending CN114238546A (en) | 2022-02-24 | 2022-02-24 | Self-adaptive cognitive diagnosis test equipment and method based on cavity convolution |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114238546A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116056074A (en) * | 2023-04-03 | 2023-05-02 | 微网优联科技(成都)有限公司 | Wireless communication control method based on multiple verification and wireless router applying same |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109857835A (en) * | 2018-12-28 | 2019-06-07 | 北京红山瑞达科技有限公司 | A kind of adaptive network security knowledge assessment method based on cognitive diagnosis theory |
| CN111047515A (en) * | 2019-12-29 | 2020-04-21 | 兰州理工大学 | Cavity convolution neural network image super-resolution reconstruction method based on attention mechanism |
| CN111460128A (en) * | 2019-11-14 | 2020-07-28 | 临沂市拓普网络股份有限公司 | Computerized self-adaptive test method based on cognitive diagnosis |
| CN113284011A (en) * | 2021-04-27 | 2021-08-20 | 西安邮电大学 | Personalized cognitive diagnosis method based on learning evaluation data |
| CN113344635A (en) * | 2021-06-24 | 2021-09-03 | 深圳市小摩豆网络科技有限公司 | Customer consumption behavior habit management and analysis method |
-
2022
- 2022-02-24 CN CN202210170060.2A patent/CN114238546A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109857835A (en) * | 2018-12-28 | 2019-06-07 | 北京红山瑞达科技有限公司 | A kind of adaptive network security knowledge assessment method based on cognitive diagnosis theory |
| CN111460128A (en) * | 2019-11-14 | 2020-07-28 | 临沂市拓普网络股份有限公司 | Computerized self-adaptive test method based on cognitive diagnosis |
| CN111047515A (en) * | 2019-12-29 | 2020-04-21 | 兰州理工大学 | Cavity convolution neural network image super-resolution reconstruction method based on attention mechanism |
| CN113284011A (en) * | 2021-04-27 | 2021-08-20 | 西安邮电大学 | Personalized cognitive diagnosis method based on learning evaluation data |
| CN113344635A (en) * | 2021-06-24 | 2021-09-03 | 深圳市小摩豆网络科技有限公司 | Customer consumption behavior habit management and analysis method |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116056074A (en) * | 2023-04-03 | 2023-05-02 | 微网优联科技(成都)有限公司 | Wireless communication control method based on multiple verification and wireless router applying same |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108960409B (en) | Method and device for generating annotation data and computer-readable storage medium | |
| CN109165692B (en) | User character prediction device and method based on weak supervised learning | |
| CN108052577A (en) | A kind of generic text content mining method, apparatus, server and storage medium | |
| CN108229478A (en) | Image, semantic segmentation and training method and device, electronic equipment, storage medium and program | |
| Loy et al. | Model choice and diagnostics for linear mixed-effects models using statistics on street corners | |
| CN111581966A (en) | Context feature fusion aspect level emotion classification method and device | |
| CN109902912B (en) | Personalized image aesthetic evaluation method based on character features | |
| CN111881262A (en) | Text emotion analysis method based on multi-channel neural network | |
| AU2011254040A1 (en) | Method, apparatus and system for determining a saliency map for an input image | |
| Koo et al. | Integrating spatial data analysis functionalities in a GIS environment: Spatial Analysis using ArcGIS Engine and R (SAAR) | |
| CN113449787B (en) | Chinese character stroke structure-based font library completion method and system | |
| CN112949628B (en) | Track data enhancement and track identification method based on embedding-mixing | |
| Kang | Combining grey relationship analysis and neural network to develop attractive automobile booth design | |
| CN113298780A (en) | Child bone age assessment method and system based on deep learning | |
| CN113283336A (en) | Text recognition method and system | |
| CN114612714A (en) | Curriculum learning-based non-reference image quality evaluation method | |
| CN118133689B (en) | Teaching scene-oriented simulation platform data processing method and device and electronic equipment | |
| CN117540104A (en) | Learning group difference evaluation method and system based on graph neural network | |
| CN114238546A (en) | Self-adaptive cognitive diagnosis test equipment and method based on cavity convolution | |
| Fieberg | Statistics for Ecologists: A Frequentist and Bayesian Treatment of Modern Regression Models. | |
| CN113409157B (en) | Cross-social network user alignment method and device | |
| CN114511759A (en) | Method and system for identifying categories and determining characteristics of skin state images | |
| CN113609355B (en) | Video question-answering system, method, computer and storage medium based on dynamic attention and graph network reasoning | |
| KR20190082453A (en) | Method, apparatus and computer program for analyzing new learning contents for machine learning modeling | |
| Quadri | Constructing Frameworks for Task-Optimized Visualizations |
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: 20220325 |
|
| RJ01 | Rejection of invention patent application after publication |