CN210348859U - Examination paper modifying all-in-one machine - Google Patents
Examination paper modifying all-in-one machine Download PDFInfo
- Publication number
- CN210348859U CN210348859U CN201822036224.4U CN201822036224U CN210348859U CN 210348859 U CN210348859 U CN 210348859U CN 201822036224 U CN201822036224 U CN 201822036224U CN 210348859 U CN210348859 U CN 210348859U
- Authority
- CN
- China
- Prior art keywords
- test paper
- module
- question
- topic
- scanning file
- 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.)
- Active
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The utility model provides a test paper correcting all-in-one machine, wherein a scanning module is used for scanning standard test paper to obtain standard test paper scanning files and scanning student test paper to obtain student test paper scanning files; the identification module is used for identifying the question stem and the answer of each question in the standard test paper scanning file and identifying the question stem and the answer of each question in the student test paper scanning file; the storage module is used for storing the question stem and the answer of each question in the standard test paper scanning file identified by the identification module; and the correcting module is used for correcting each question in the student test paper scanning file according to the question stem and the answer of each question in the standard test paper scanning file stored in the storage module and the question stem and the answer of each question in the student test paper scanning file identified by the identification module. The utility model discloses can solve the problem that prior art pilot scale paper wholesale is inefficient, make mistakes easily.
Description
Technical Field
The utility model belongs to the technical field of the operation is criticized and is changed, a paper is criticized and is changed all-in-one is related to.
Background
At present, teachers mostly arrange assignments for students in the form of test paper and test the learning effect of the students, which results in that the teachers need to amend a large number of student test paper. However, in the prior art, the way that the teacher corrects the test paper is more traditional, usually based on handwriting, which is generally inefficient and error-prone, for example, if a teacher corrects 60 students' test papers every day, each student corrects for 5 minutes, and works for 5 hours every day.
Therefore, how to improve the efficiency of the teacher correcting the test paper and reduce the error is a problem to be solved urgently.
SUMMERY OF THE UTILITY MODEL
An object of the utility model is to provide a paper is criticized and is changed all-in-one to solve the problem that prior art paper is criticized and is changed inefficiency, make mistakes easily.
In order to solve the technical problem, the utility model provides a paper revises all-in-one, include: the device comprises a scanning module, an identification module connected with the scanning module, a storage module connected with the identification module and a correction module, wherein the storage module is connected with the correction module;
the scanning module is used for scanning standard test paper to obtain a standard test paper scanning file and scanning student test paper to obtain a student test paper scanning file;
the identification module is used for identifying the question stem and the answer of each question in the standard test paper scanning file and identifying the question stem and the answer of each question in the student test paper scanning file;
the storage module is used for storing the question stem and the answer of each question in the standard test paper scanning file identified by the identification module;
and the correcting module is used for correcting each question in the student test paper scanning file according to the question stem and the answer of each question in the standard test paper scanning file stored in the storage module and the question stem and the answer of each question in the student test paper scanning file identified by the identification module.
Optionally, the examination paper approval all-in-one machine further includes: and the printing module is connected with the correcting module and is used for printing the student test paper scanning file corrected by the correcting module.
Optionally, the examination paper approval all-in-one machine further includes: the paper assembling module is connected with the storage module and is used for selecting questions from the questions stored in the storage module to form test papers or directly selecting the test papers from the test paper templates stored in the storage module;
the examination paper approval all-in-one machine further comprises: and the printing module is respectively connected with the correcting module and the paper assembling module and is used for printing the student test paper scanning file corrected by the correcting module and printing test papers formed by the paper assembling module or selected test papers.
Optionally, the correcting module is further configured to count the scores of the student test papers according to the correcting result of the student test paper scanning file.
Optionally, the identification module is further configured to identify the school number and/or name on the student test paper scanning file, and output the school number and/or name and a corresponding score.
Optionally, when the standard test paper is a single sheet, the correcting unit is specifically configured to compare the identified answers in the standard test paper scanning file and the student test paper scanning file according to the position or the serial number of the question, and correct each question in the student test paper scanning file.
Optionally, when the standard test paper is a plurality of different test papers, the correcting module includes: a search submodule and a modification submodule;
the searching sub-module is used for searching in the questions stored in the storage module according to the question stem of each question in the student test paper scanning file identified by the identification module and determining a standard test paper matched with the student test paper;
and the correcting submodule is used for correcting the answers of the questions according to the answers of the questions matched with the questions on the determined standard test paper aiming at each question in the student test paper scanning file.
Optionally, the search submodule includes:
the first obtaining unit is used for inputting the text content of the question stem of each question in the student test paper scanning file into a pre-trained question stem vectorization model to obtain a feature vector of the question stem of each question as the feature vector of each question, wherein the question stem vectorization model is a model based on a neural network;
the searching unit is used for searching in the titles stored in the storage module aiming at each title, searching for a feature vector matched with the feature vector of the title, and determining the title corresponding to the feature vector matched in the titles stored in the storage module as the title closest to the title;
and the determining unit is used for summarizing the test paper of the closest subject of all the searched subjects, and determining the test paper meeting the preset conditions as the standard test paper matched with the student test paper.
Optionally, in a case that the subjects in the student test paper include pictures, the search sub-module further includes:
a second obtaining unit, configured to input a picture in a topic including the picture into a pre-trained picture vectorization model to obtain a feature vector of the picture including the topic of the picture, where the picture vectorization model is a model based on a neural network;
and a third obtaining unit, configured to, for a topic that does not include a picture, directly use the feature vector of the topic stem of the topic as the feature vector of the topic, and for a topic that includes a picture, splice the feature vector of the topic picture and the feature vector of the topic stem to use as the feature vector of the topic.
Optionally, the search sub-module further includes:
the preprocessing unit is used for establishing an index information table for the feature vectors of all the titles stored in the storage module in advance;
the searching unit is specifically configured to search, for each topic, a feature vector matched with the feature vector of the topic in the index information table; and determining the corresponding topic of the matched feature vector in the index information table as the topic closest to the topic.
Optionally, the preprocessing unit is further configured to group the feature vectors with different lengths according to length before establishing the index information table;
the searching unit is specifically configured to search, for each topic to be searched, a feature vector matched with the feature vector of the topic to be searched in a group of the index information table, where the length of the group of the feature vectors is the same as or similar to the length of the feature vector of the topic to be searched.
Optionally, the determining unit is specifically configured to determine the test paper with the largest frequency of occurrence and larger than a first preset threshold as a standard test paper matched with the student test paper.
Optionally, the identification module includes: a detection submodule and an identification submodule;
the detection submodule is used for detecting the standard test paper scanning file by using a pre-trained detection model, detecting the area of each subject in the standard test paper scanning file, detecting the student test paper scanning file by using a pre-trained detection model, and detecting the area of each subject in the student test paper scanning file, wherein the detection model is a model based on a neural network;
the recognition submodule is used for recognizing the stem and the answer text content in the area of each subject in the standard test paper scanning file by using a pre-trained recognition model, and recognizing the stem and the answer text content in the area of each subject in the student test paper scanning file by using the pre-trained recognition model, wherein the recognition model is a model based on a neural network.
Compared with the prior art, the utility model provides a test paper correcting and amending integrated machine, which comprises a scanning module, an identification module connected with the scanning module, a storage module connected with the identification module and a correcting module, wherein the storage module is connected with the correcting module, only the standard test paper needs to be scanned by the scanning module to obtain a standard test paper scanning file, then the identification module identifies the question stem and answer of each question in the standard test paper scanning file, the storage module stores the question stem and answer of each question in the standard test paper scanning file, when a student test paper needs to be corrected, the student test paper is scanned by the scanning module to obtain the standard test paper scanning file, then the identification module identifies the question stem and answer of each question in the student test paper scanning file, and finally the correcting module scans the question stem and answer of each question in the standard test paper scanning file according to the standard test paper stored by the storage module, and correcting each question in the student test paper scanning file. Therefore, the test paper correcting all-in-one machine is used for correcting the test paper, so that a teacher does not need to spend a large amount of time and energy to manually correct each student test paper, the time of the teacher can be greatly saved, the efficiency of correcting the test paper is improved, and errors are reduced.
Drawings
Fig. 1 is a schematic structural diagram of an examination paper batching and modifying integrated machine provided by an embodiment of the present invention.
Detailed Description
The following describes the examination paper batching and reforming integrated machine provided by the present invention in further detail with reference to the accompanying drawings and specific embodiments. The advantages and features of the present invention will become more apparent from the claims and the following description.
Fig. 1 is a schematic view of a test paper batching and modifying integrated machine provided by an embodiment of the present invention. Referring to fig. 1, an examination paper approval and correction machine may include: the device comprises a scanning module 1, an identification module 2 connected with the scanning module 1, a storage module 3 and a correction module 4 connected with the identification module 2, wherein the storage module 3 is connected with the correction module 4.
The scanning module 1 is used for scanning standard test paper to obtain standard test paper scanning files and scanning student test paper to obtain student test paper scanning files. The standard test paper can be a test paper with standard answers filled in each question, and the student test paper can be a test paper for students to answer. The obtained standard test paper scanning file and the student test paper scanning file can be files in image or PDF format and the like.
The identification module 2 is used for identifying the question stem and the answer of each question in the standard test paper scanning file and identifying the question stem and the answer of each question in the student test paper scanning file.
The storage module 3 is configured to store the question stem and the answer of each question in the standard test paper scan file identified by the identification module 2. The storage module 3 may further store the standard test paper scanning file and the student test paper scanning file.
The correcting module 4 is configured to correct each question in the student test paper scanning file according to the question stem and the answer of each question in the standard test paper scanning file stored in the storage module 3 and the question stem and the answer of each question in the student test paper scanning file identified by the identifying module 2. The way of correction may be to mark √ or ×, at the answer of each topic.
Utilize the utility model provides a paper is criticized and is changed all-in-one and carry out the operation and criticize and change, only need scan the standard paper by scanning module 1 and obtain standard paper scanning file, the problem of every topic is done and the answer in the standard paper scanning file is discerned by identification module 2 again, and the problem of every topic is done and the answer in the standard paper scanning file is stored by storage module 3, when will criticize and change a certain student's paper, scan the student's paper by scanning module 1 and obtain standard paper scanning file, the problem of every topic is done and the answer in the student's paper scanning file is discerned by identification module 2 again, the problem of every topic is done and the answer in the standard paper scanning file of storage module 3 storage is done and is changed by criticizing module 4 at last, the problem of every topic is criticized and is changed in the student's paper scanning file. Therefore, the test paper correcting all-in-one machine is used for correcting the test paper, so that a teacher does not need to spend a large amount of time and energy to manually correct each student test paper, the time of the teacher can be greatly saved, the efficiency of correcting the test paper is improved, and errors are reduced.
Specifically, the identification module 2 may include a detection sub-module and an identification sub-module;
the detection submodule is used for detecting the standard test paper scanning file by using a pre-trained detection model, detecting the area of each subject in the standard test paper scanning file, detecting the student test paper scanning file by using a pre-trained detection model, and detecting the area of each subject in the student test paper scanning file, wherein the detection model is a model based on a neural network;
the recognition submodule is used for recognizing the stem and the answer text content in the area of each subject in the standard test paper scanning file by using a pre-trained recognition model, and recognizing the stem and the answer text content in the area of each subject in the student test paper scanning file by using the pre-trained recognition model, wherein the recognition model is a model based on a neural network.
The detection model may be obtained by training samples in a test paper sample training set based on a deep Convolutional Neural Network (CNN), for example. Extracting a two-dimensional characteristic vector from a scanning file by using a trained detection model, generating anchor points with different shapes in each grid of the two-dimensional characteristic vector, labeling the detected regions of each topic by using a labeling frame (group ports), and performing regression (regression) processing on the labeling frame and the generated anchor points to enable the labeling frame to be closer to the actual position of the topic. After the topic areas are detected, each topic is cut into a single area, or the topic areas are not actually cut, each topic area is distinguished during processing and is used as a single area for processing, and sequencing is carried out according to the position information of the topics.
After the areas of the questions are detected, the recognition sub-module can recognize the question stems and answers in the areas of the questions by using a pre-trained recognition model, wherein the recognition model is a model based on a neural network. Firstly, each component in the question is marked, wherein the component can comprise a question stem, an answer and/or a picture, and then the text content of the question stem and the text content of the answer in the question are identified through an identification model. The recognition model can be established based on a hole convolution and an attention model, specifically, the hole convolution is adopted to extract features of labeling frames corresponding to the question stem, the answer and/or the picture, and then the extracted features are decoded into characters through the attention model. Further, the recognition model may include a recognition model for a stem and a recognition model for an answer, wherein the stem is composed of printing fonts, the answer is composed of handwriting fonts, and the recognition model for the stem and the recognition model for the answer are respectively formed by independent training.
In one case, the standard test paper is a single sheet, that is, the scanning module 1 scans a single standard test paper to obtain a standard test paper scanning file. At this time, the correcting module 4 is specifically configured to compare the identified answers in the standard test paper scanning file and the student test paper scanning file according to the positions or serial numbers of the questions, and correct each question in the student test paper scanning file. Specifically, because the positions or serial numbers of the subjects in the standard test paper and the student test paper are the same, the answers of the subjects with the same subject positions or serial numbers in the standard test paper scanning file and the student test paper scanning file can be compared, and the purpose of correcting each subject in the student test paper scanning file is achieved. For example, the answer of the question 1 or the first position question identified by the standard test paper is A, the answer of the question 1 or the first position question identified by the student test paper is B, and the answer A and the answer B are compared and corrected.
In another case, the standard test paper is a plurality of different test papers, that is, the scanning module 1 scans a plurality of different standard test papers to obtain a plurality of different standard test paper scanning files. The modification module 4 may include: the search submodule is used for searching in the questions stored in the storage module 3 according to the question stems of all the questions in the student test paper scanning file identified by the identification module 2, and determining standard test papers matched with the student test papers; and the correcting submodule is used for correcting the answers of the questions according to the answers of the questions matched with the questions on the determined standard test paper aiming at each question in the student test paper scanning file.
In practical applications, the search sub-module may include: the first obtaining unit is used for inputting the text content of the question stem of each question in the student test paper scanning file into a pre-trained question stem vectorization model to obtain a feature vector of the question stem of each question as the feature vector of each question, wherein the question stem vectorization model is a model based on a neural network; the searching unit is used for searching in the titles stored in the storage module 3 for each title, searching for a feature vector matched with the feature vector of the title, and determining the title corresponding to the feature vector matched in the titles stored in the storage module 3 as the title closest to the title; and the determining unit is used for summarizing the test paper of the closest subject of all the searched subjects, and determining the test paper meeting the preset conditions as the standard test paper matched with the student test paper.
The question stem vectorization model can be a neural network-based model, such as a CNN model, and can be obtained by training through the following steps: labeling each topic sample in the first topic sample training set to label the text content of the topic stem in each topic sample; and performing two-dimensional feature vector extraction on the text content of the question stem in each question sample by using a neural network model, thereby training to obtain the question stem vectorization model. The specific training process belongs to the prior art, and is not described herein.
Preferably, in order to facilitate the search of the feature vector, an index information table may be further established in advance for the feature vectors of the respective topics stored in the storage module 3. The index information table can store the feature vector of each topic in the topic library, the specific content of the topic, the ID of the test paper where the topic is located, and the like. Specifically, the search sub-module may further include: the preprocessing unit is used for establishing an index information table for the feature vectors of all the titles stored in the storage module 3 in advance; the searching unit is specifically configured to search, for each topic, a feature vector matched with the feature vector of the topic in the index information table; and determining the corresponding topic of the matched feature vector in the index information table as the topic closest to the topic.
In addition, the feature vectors with different lengths can be grouped according to the length so as to be convenient for searching and improve the searching speed, and the preprocessing unit is also used for grouping the feature vectors with different lengths according to the length before the index information table is established; the searching unit is specifically configured to search, for each topic to be searched, a feature vector matched with the feature vector of the topic to be searched in a group of the index information table, where the length of the group of the feature vectors is the same as or similar to the length of the feature vector of the topic to be searched.
Specifically, the determining unit is specifically configured to determine the test paper with the largest occurrence frequency and larger than a first preset threshold as the standard test paper matched with the student test paper.
Furthermore, under the condition that the questions in the student test paper contain pictures, searching is carried out by combining the question stems in the questions and the pictures when the questions are searched, and the accuracy of searching the questions can be further improved. Therefore, the search sub-module may further include: a second obtaining unit, configured to input a picture in a topic including the picture into a pre-trained picture vectorization model to obtain a feature vector of the picture including the topic of the picture, where the picture vectorization model is a model based on a neural network; and a third obtaining unit, configured to, for a topic that does not include a picture, directly use the feature vector of the topic stem of the topic as the feature vector of the topic, and for a topic that includes a picture, splice the feature vector of the topic picture and the feature vector of the topic stem to use as the feature vector of the topic.
The picture vectorization model may be a neural network-based model, such as a CNN model, and may be obtained by training: labeling each topic sample in the second topic sample training set to label a picture in each topic sample; and extracting two-dimensional feature vectors of the pictures in each topic sample by using a neural network model, thereby training to obtain the picture vectorization model. The specific training process belongs to the prior art and is not described herein.
After the search sub-module determines the standard test paper matched with the student test paper, the correction sub-module can correct the answer of each question in the student test paper scanning file according to the answer of the question matched with the question on the determined standard test paper. For example, for the topic A in the student test paper scanning file, if the topic A has the same topic stem as the topic A ' on the determined standard test paper, the topic A ' is matched with the topic A, and the answer to the topic A is modified according to the answer to the topic A '. It should be noted that, for the subjects with the types of oral calculation, selection, blank filling, judgment and simple calculation (such as four-rule mixing operation, discrete calculation, vertical calculation, unit conversion, score calculation, division with remainder, estimation, etc.), the examination paper batching and reforming integrated machine can be used for automatic batching and reforming, while the subjective subjects can be batched and reformed manually by teachers.
It should be noted that, in practical application, the standard test paper may also be a blank test paper with no answer filled in, a teacher puts the blank test paper into the test paper modifying all-in-one machine, after the scanning module 1 scans the blank test paper, the identification module 2 identifies the question stem in the blank test paper, searches corresponding questions and test papers in the question bank according to the question stem, fills the standard answer of the corresponding question in the blank test paper, and the test paper with the answer filled in is the standard test paper, so that the standard test paper can be stored in the storage module 3 as a reference test paper when modifying the student test paper, so that the modifying module 4 can modify the student test paper. If the corresponding questions cannot be searched in the question bank, the identification module can prompt a teacher to manually fill in standard answers of the unsearched questions in the editable test paper scanning file. Or after the scanning module 1 scans to obtain a blank test paper scanning file, the processing module converts the scanning file into an editable file format, and then a teacher manually fills standard answers in answering positions of corresponding questions directly so as to be used as a reference test paper at a later stage. The test paper filled with the standard answer may be stored in the storage module 3 by a transmission method (e.g., a transmission method through a wired network, a wireless network, or a data interface (e.g., USB, etc.)). In addition, a blank test paper with the filled-in answers may be printed as a standard test paper for the teacher.
Further, the examination paper correction all-in-one machine can further comprise a printing module 5, wherein the printing module 5 is connected with the correction module 4 and used for printing the student examination paper scanning file corrected by the correction module 4 so that teachers and students can check the correction results of the examination paper conveniently.
In practical application, the test paper batching and changing all-in-one machine further comprises a test paper grouping module 6, wherein the test paper grouping module 6 is connected with the storage module 3 and is used for selecting subjects from the subjects stored in the storage module 3 to form test papers or directly selecting test papers from test paper templates stored in the storage module 3.
It can be understood that the storage module 3 can be used as an item bank to store the test paper templates and a large number of items, so that the teacher can select the items from the item bank to form the test paper to be arranged for the student to answer according to the requirement. At this time, the printing module 5 is respectively connected to the correcting module 4 and the paper assembling module 6, and is configured to print out the student test paper scanning file corrected by the correcting module 4, and print out the test paper or the selected test paper composed by the paper assembling module 6. In this case, since the questions (including the question stem and the answer) of the test paper composed by the paper composing module 6 or the selected test paper are already stored in the storage module 3, the teacher does not need to fill in the standard answer to obtain the standard test paper and scan the standard test paper, but can directly scan the test paper answered by the student, after the identification module 2 identifies the question stem and the answer of each question in the student test paper scanning file, and the correction module 4 corrects each question in the student test paper scanning file according to the question stem and the answer of the question in the question bank stored in the storage module 3, so that the workload of the teacher can be further reduced.
Optionally, the correcting module 4 may be further configured to count the scores of the student test papers according to the correcting results of the student test paper scanning files. Specifically, the correct questions in the correction results and the corresponding scores are added, and the scores of the student test paper are obtained through statistics.
Further, the identification module 4 may be further configured to identify the school number and/or name on the student test paper scan file, and output the school number and/or name and the corresponding score. For example, the school number and/or name of the student test paper scanning file is recognized, the school number and/or name and the corresponding score can be output to a computer connected with the test paper batching and changing machine, and the teacher can conveniently check the school number and/or name.
The above description is only for the purpose of describing the preferred embodiments of the present invention, and is not intended to limit the scope of the present invention, and any variations and modifications made by those skilled in the art based on the above disclosure are within the scope of the appended claims.
Claims (9)
1. The utility model provides a paper wholesale all-in-one which characterized in that includes: the device comprises a scanning module, an identification module connected with the scanning module, a storage module connected with the identification module and a correction module, wherein the storage module is connected with the correction module;
the scanning module is used for scanning standard test paper to obtain a standard test paper scanning file and scanning student test paper to obtain a student test paper scanning file;
the identification module is used for identifying the question stem and the answer of each question in the standard test paper scanning file and identifying the question stem and the answer of each question in the student test paper scanning file;
the storage module is used for storing the question stem and the answer of each question in the standard test paper scanning file identified by the identification module;
the correcting module is used for correcting each question in the student test paper scanning file according to the question stem and the answer of each question in the standard test paper scanning file stored in the storage module and the question stem and the answer of each question in the student test paper scanning file identified by the identification module;
wherein the identification module comprises: a detection submodule and an identification submodule;
the detection submodule is used for detecting the standard test paper scanning file by using a pre-trained detection model, detecting the area of each subject in the standard test paper scanning file, detecting the student test paper scanning file by using a pre-trained detection model, and detecting the area of each subject in the student test paper scanning file, wherein the detection model is a model based on a neural network;
the recognition submodule is used for recognizing the stem and the answer text content in the area of each subject in the standard test paper scanning file by using a pre-trained recognition model, and recognizing the stem and the answer text content in the area of each subject in the student test paper scanning file by using the pre-trained recognition model, wherein the recognition model is a model based on a neural network.
2. The test paper batching and changing all-in-one machine according to claim 1, further comprising: and the printing module is connected with the correcting module and is used for printing the student test paper scanning file corrected by the correcting module.
3. The test paper batching and changing all-in-one machine according to claim 1, further comprising: the paper assembling module is connected with the storage module and is used for selecting questions from the questions stored in the storage module to form test papers or directly selecting the test papers from the test paper templates stored in the storage module;
further comprising: and the printing module is respectively connected with the correcting module and the paper assembling module and is used for printing the student test paper scanning file corrected by the correcting module and printing test papers formed by the paper assembling module or selected test papers.
4. The test paper correcting all-in-one machine as claimed in claim 1, wherein when the standard test paper is a single sheet, the correcting module is specifically configured to compare the identified answers in the standard test paper scanning file and the student test paper scanning file according to the position or sequence number of the question, and correct each question in the student test paper scanning file.
5. The examination paper correction integrated machine according to claim 1, wherein when the standard examination paper is a plurality of different examination papers, the correction module comprises: a search submodule and a modification submodule;
the searching sub-module is used for searching in the questions stored in the storage module according to the question stem of each question in the student test paper scanning file identified by the identification module and determining a standard test paper matched with the student test paper;
and the correcting submodule is used for correcting the answers of the questions according to the answers of the questions matched with the questions on the determined standard test paper aiming at each question in the student test paper scanning file.
6. The examination paper approval all-in-one machine of claim 5, wherein the search submodule comprises:
the first obtaining unit is used for inputting the text content of the question stem of each question in the student test paper scanning file into a pre-trained question stem vectorization model to obtain a feature vector of the question stem of each question as the feature vector of each question, wherein the question stem vectorization model is a model based on a neural network;
the searching unit is used for searching in the titles stored in the storage module aiming at each title, searching for a feature vector matched with the feature vector of the title, and determining the title corresponding to the feature vector matched in the titles stored in the storage module as the title closest to the title;
and the determining unit is used for summarizing the test paper of the closest subject of all the searched subjects, and determining the test paper meeting the preset conditions as the standard test paper matched with the student test paper.
7. The examination paper batching and changing all-in-one machine as claimed in claim 6, wherein in case that the subject in the student examination paper contains a picture, the search sub-module further comprises:
a second obtaining unit, configured to input a picture in a topic including the picture into a pre-trained picture vectorization model to obtain a feature vector of the picture including the topic of the picture, where the picture vectorization model is a model based on a neural network;
and a third obtaining unit, configured to, for a topic that does not include a picture, directly use the feature vector of the topic stem of the topic as the feature vector of the topic, and for a topic that includes a picture, splice the feature vector of the topic picture and the feature vector of the topic stem to use as the feature vector of the topic.
8. The test paper batching and changing all-in-one machine as claimed in claim 6, wherein the search sub-module further comprises:
the preprocessing unit is used for establishing an index information table for the feature vectors of all the titles stored in the storage module in advance;
the searching unit is specifically configured to search, for each topic, a feature vector matched with the feature vector of the topic in the index information table; and determining the corresponding topic of the matched feature vector in the index information table as the topic closest to the topic.
9. The examination paper batching and changing all-in-one machine as claimed in claim 8, wherein the preprocessing unit is further configured to group the feature vectors of different lengths according to length before establishing the index information table;
the searching unit is specifically configured to search, for each topic to be searched, a feature vector matched with the feature vector of the topic to be searched in a group of the index information table, where the length of the group of the feature vectors is the same as or similar to the length of the feature vector of the topic to be searched.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201822036224.4U CN210348859U (en) | 2018-12-05 | 2018-12-05 | Examination paper modifying all-in-one machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201822036224.4U CN210348859U (en) | 2018-12-05 | 2018-12-05 | Examination paper modifying all-in-one machine |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN210348859U true CN210348859U (en) | 2020-04-17 |
Family
ID=70172778
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201822036224.4U Active CN210348859U (en) | 2018-12-05 | 2018-12-05 | Examination paper modifying all-in-one machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN210348859U (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109326161A (en) * | 2018-12-05 | 2019-02-12 | 杭州大拿科技股份有限公司 | A kind of paper corrects all-in-one machine |
-
2018
- 2018-12-05 CN CN201822036224.4U patent/CN210348859U/en active Active
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109326161A (en) * | 2018-12-05 | 2019-02-12 | 杭州大拿科技股份有限公司 | A kind of paper corrects all-in-one machine |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110008933B (en) | Universal intelligent marking system and method | |
| US11508251B2 (en) | Method and system for intelligent identification and correction of questions | |
| US11790641B2 (en) | Answer evaluation method, answer evaluation system, electronic device, and medium | |
| CN109271401B (en) | Topic searching and correcting method and device, electronic equipment and storage medium | |
| CN109326161A (en) | A kind of paper corrects all-in-one machine | |
| CN110659584B (en) | Intelligent mark-remaining paper marking system based on image recognition | |
| CN110110581B (en) | Test paper correcting method and system based on artificial intelligence | |
| CN109817046B (en) | Learning auxiliary method based on family education equipment and family education equipment | |
| CN109710590A (en) | A kind of wrong answer list generation method and device | |
| CN113590956B (en) | Knowledge point recommendation method, knowledge point recommendation device, knowledge point recommendation terminal and computer readable storage medium | |
| CN109284355B (en) | Method and device for correcting oral arithmetic questions in test paper | |
| CN106227773B (en) | Data processing method and system | |
| CN110222678B (en) | Topic analysis method, system, readable storage medium and electronic device | |
| CN112990180B (en) | Question judging method, device, equipment and storage medium | |
| CN106846961A (en) | The treating method and apparatus of electronic test paper | |
| CN107038438A (en) | It is a kind of that method is read and appraised based on image recognition | |
| CN111144079A (en) | Method and device for intelligently acquiring learning resources, printer and storage medium | |
| CN112347997A (en) | Test question detection and identification method and device, electronic equipment and medium | |
| CN108153915A (en) | A kind of educational information fast acquiring method based on internet | |
| CN110837793A (en) | Intelligent recognition handwriting mathematical formula reading and amending system | |
| CN111767307A (en) | Object processing method and system, and mathematic subject correcting method and system | |
| CN116303871A (en) | Exercise book reading method | |
| CN111126486A (en) | Test statistical method, device, equipment and storage medium | |
| CN210348859U (en) | Examination paper modifying all-in-one machine | |
| CN113569112A (en) | Tutoring strategy providing method, system, device and medium based on question |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant | ||
| GR01 | Patent grant |