CN105224665B - Wrong question management method and system - Google Patents

Abstract

The invention discloses a wrong question management method and system. The method comprises the following steps: acquiring a picture of a wrong question; performing character recognition on the picture to acquire keywords of the wrong question; matching the test questions in the cloud question bank by using the keywords; if the same or similar test questions are matched, inputting the test questions into an electronic wrong question bank; and if no matched test questions exist, the wrong questions are fed back to the background server, and the background server records the wrong questions into the electronic wrong question bank. The invention obtains the wrong question from the cloud question bank or the background server to be recorded into the electronic wrong question book by performing character recognition on the wrong question picture and utilizing the keywords of the wrong question, improves the wrong question recording efficiency, and is favorable for classifying, sorting and reviewing and recommending the wrong question.

Description

Method and system for managing wrong questions

technical field

The invention relates to the technical field of intelligent learning, in particular to a method and system for managing wrong questions.

Background technique

During the learning process, students often need to sort out the wrong test questions and make corresponding wrong question books, and conduct targeted learning on the contents of the wrong question books during review, so as to better grasp what they have not mastered in the learning process. Knowledge points, problem-solving skills, etc., in order to improve academic performance, but sorting out the correct and wrong question books is a very time-consuming and cumbersome task for students. The most original method of making wrong question books is to manually copy paper wrong questions on test papers, books, exercise books, etc. into the wrong question book. This method has a large workload, low efficiency, and is difficult to persist; In the wrong question book management method, the wrong question book is made by inputting the wrong question pictures on paper, but the wrong question pictures cannot be classified and sorted after the wrong question is entered. The wrong question book is messy and the input efficiency is low. . The present invention recognizes the characters of the wrong question pictures, uses the keywords of the wrong questions, obtains the wrong questions from the cloud question bank or the background server and enters the electronic wrong question book, improves the efficiency of wrong question entry, and is beneficial to classify the wrong questions Organize and review recommendations.

Contents of the invention

The present invention proposes a method and system for managing wrong questions. The method uses keywords of wrong questions to obtain wrong questions from a cloud question bank or a background server and enter them into an electronic wrong question book, thereby improving the efficiency of wrong question entry.

For reaching this purpose, the present invention adopts following technical scheme:

In a first aspect, the present invention proposes a method for managing wrong questions, including:

Get the picture of the wrong question;

Carrying out character recognition on the picture to obtain the keyword of the wrong question;

Using the keywords to match the test questions in the cloud question bank;

If the same or similar test questions are matched, the test questions will be entered into the electronic wrong question bank; if there is no matching test question, the wrong questions will be fed back to the backend server, and the backend server will enter the wrong questions into the electronic wrong question bank .

The technology adopted for the character recognition of the picture is optical character recognition technology, and the optical character recognition program is used to scan and analyze the picture of the wrong question, and automatically analyze the keywords in the picture of the wrong question. The same or similar test questions are specifically the test questions that are the same as the photographed wrong question and the test questions whose keywords are synonyms of the key of the wrong question.

Wherein, the matching of the test questions in the cloud test bank by using the keyword is specifically: matching the test questions in the cloud test bank by using the keyword and a synonym of the keyword.

Wherein, after the matching, it also includes: recording the reason of the wrong question, counting the frequency of the wrong question, and classifying the wrong question;

The categorization of the wrong questions is specifically: classifying according to the attributes of the wrong questions;

The attributes include grades, subjects, and knowledge points.

Wherein, after said inputting the test questions into the electronic wrong question bank, it also includes: obtaining related content of the wrong question from the cloud question bank, and entering the related content into the electronic wrong question bank.

If the same or similar test questions are matched, the test questions entered into the electronic wrong question bank are the same or similar test questions. Test questions or questions with the same error reason;

Wherein, after the background server enters the wrong question into the electronic wrong question bank, it further includes: entering the wrong question into the cloud question bank.

Wherein, after the classification of wrong questions, further includes: recommending review of wrong questions.

Among them, the recommended review questions include:

If the frequency of the wrong question exceeds the preset first threshold, it is recommended to review the wrong question, or recommend reviewing the wrong question and related content;

If the frequency of wrong questions for the same knowledge point exceeds the preset second threshold, it is recommended to review the wrong questions for this knowledge point;

If the frequency of wrong questions with the same reason for wrong questions exceeds the preset third threshold, it is recommended to review the wrong questions for that reason.

Here, there are no specific regulations on the preset first threshold, preset second threshold and preset third threshold, and the preset first threshold, preset second threshold and preset third threshold may all be the same, or may be They are all different, or both of them can be the same. Different settings can be made according to the user's learning ability, learning frequency, and learning time each time. For example, the preset first threshold can be 3 times, 5 times, 7 times or 10 times. The preset second threshold can be 2 times, 4 times, 6 times or 8 times, and the preset third threshold can be 3 times, 5 times, 7 times or 10 times.

In a second aspect, the present invention proposes a wrong question management system, including:

Acquisition module: used to obtain pictures of wrong questions;

Recognition module: used to perform character recognition on the picture to obtain the keyword of the wrong question;

Matching module: used to use the keywords to match the test questions in the cloud question bank;

Input module: used to input the same or similar test questions into the electronic wrong question database; if there is no matching test question, feed back the wrong questions to the backend server, and the backend server will enter the wrong questions to the electronic wrong question bank.

Wherein, it also includes a classification module: used for recording the reasons for the wrong questions, counting the frequency of the wrong questions, and classifying the wrong questions.

Among them, it also includes a review recommendation module: used to recommend review questions.

A method and system for managing wrong questions provided by the present invention, the method comprising: obtaining pictures of wrong questions; performing character recognition on the pictures to obtain keywords of the wrong questions; Perform matching; if the same or similar test questions are matched, the test questions will be entered into the electronic wrong question bank; if there is no matching test question, the wrong questions will be fed back to the background server, and the background server will enter the wrong questions into the In the wrong question bank. The present invention recognizes the characters of the wrong question pictures, uses the keywords of the wrong questions, obtains the wrong questions from the cloud question bank or the background server and enters the electronic wrong question book, improves the efficiency of wrong question entry, and is beneficial to classify the wrong questions Organize and review recommendations.

Description of drawings

Fig. 1 is a method flow chart of Embodiment 1 of a wrong question management method provided by the present invention.

Fig. 2 is a method flow chart of Embodiment 2 of a wrong question management method provided by the present invention.

Fig. 3 is a functional block diagram of Embodiment 1 of a wrong title management system provided by the present invention.

Fig. 4 is a functional block diagram of Embodiment 2 of a wrong question management system provided by the present invention.

Detailed ways

The technical solutions of the present invention will be further described through specific embodiments below in conjunction with the accompanying drawings.

Embodiment one

With reference to Fig. 1, a kind of wrong question management method comprises the following steps:

S101. Acquiring pictures of wrong questions.

The user takes pictures of the paper wrong questions on the paper books, exercise books, test papers, etc. through the camera equipment, and obtains the pictures corresponding to the wrong questions.

S102. Perform character recognition on the picture to acquire keywords of the wrong question.

In the present invention, the optical character recognition technology is preferably used for character recognition of the pictures, and the optical character recognition program is used to scan and analyze the wrong picture, and automatically analyze the keywords in the wrong picture.

Optical character recognition technology can process the picture information of photos obtained by scanning and taking pictures and convert them into text information, and analyze the keywords of wrong questions in the converted text information. This step can effectively convert wrong pictures into text The information is beneficial to help improve the entry efficiency of wrong questions. In addition, other character recognition technologies may be used to perform character recognition on the picture, and this embodiment does not specifically limit the character recognition technology used.

S103. Using the keyword to match the test questions in the cloud test bank.

The use of the keywords to match the test questions in the cloud test bank is specifically: using the keywords and synonyms of the keywords to match the test questions in the cloud test bank. When matching, not only use keywords to match, but also Synonyms of keywords are used for matching. Users can also modify keywords during the matching process, and use modified keywords to match test questions in the cloud question bank. Using multiple keyword matching methods improves the matching from the cloud question bank. The probability of finding the same or similar test questions as the wrong questions can effectively improve the efficiency of wrong question entry.

S104. If the same or similar test questions are matched, record the test questions into the electronic wrong question bank.

The similar test questions refer to the test questions whose keywords are synonymous with the keywords, and may be test questions with the same knowledge points as the wrong question, or test questions with the same problem-solving skills as the wrong question.

Preferably, after S103, it also includes:

S105. If there is no matching test question, feed back the wrong question to the background server, and the background server will enter the wrong question into the electronic wrong question database.

Wherein, one of S104 and S105 is performed.

In S104 and S105, the test questions entered from the cloud question bank or the background server are all electronic test questions, and the electronic test questions are entered into the wrong question book to generate the electronic wrong question book.

The present invention recognizes the characters of the wrong question pictures, uses the keywords of the wrong questions, obtains the wrong questions from the cloud question bank or the background server and enters the electronic wrong question book, improves the efficiency of wrong question entry, and is beneficial to classify the wrong questions Organize and review recommendations.

Embodiment two

In this embodiment, several steps are added on the basis of Embodiment 1, refer to FIG. 2 .

A method for managing wrong questions, comprising the following steps:

S201. Acquiring pictures of wrong questions.

S202. Perform character recognition on the picture to acquire keywords of the wrong question.

S203. Using the keyword to match the test questions in the cloud test bank.

S204. If the same or similar test questions are matched, record the test questions into the electronic wrong question bank.

S205. Obtain the related content of the wrong question from the cloud question bank, and enter the related content into the electronic wrong question bank;

The associated content of the wrong question includes test questions related to the knowledge points of the wrong question, test questions with the same problem-solving skills as the wrong question, the content of the relevant knowledge points of the wrong question, the content analysis of the relevant knowledge points, and all problem-solving skills. After entering the matching same or similar test questions into the electronic wrong question book, continue to obtain the related content of the wrong question from the cloud question bank, and also enter the related content of the wrong question into the electronic wrong question book. When learning the wrong questions in the wrong question book, the related content of the wrong questions can be learned together, which effectively improves the learning efficiency of the user.

S206. Recording the reasons for the wrong questions, counting the frequency of the wrong questions, and classifying the wrong questions.

The categorizing the wrong questions specifically includes: classifying according to the attributes of the wrong questions; the attributes of the wrong questions include grades, subjects, and knowledge points. Classify the wrong questions according to their grade, subject, and knowledge points. For example, the user has entered a geometry proof question, and the geometric proof question belongs to the second semester of junior high school, mathematics subject, unit 2 of Chapter 3, For the test questions about congruent triangles, classify the test questions into the category of the second grade of junior high school, the second semester of the semester, the subject of mathematics, the chapter of the second unit of Chapter 3, and the knowledge point of congruent triangles , if the re-entered test questions are still related to congruent triangles, they will also be classified under the category directory, and the user can quickly find all the test questions of the corresponding grade or all the test questions of the corresponding subject of the corresponding grade under the category directory , or all the test questions of the corresponding knowledge points of the corresponding subjects of the corresponding grade, and read the selected test questions according to the user's choice for review and recommendation. At this time, the wrong question rate of the entered test questions can also be obtained from the cloud question bank. The wrong question rate refers to the ratio of the number of times the same wrong question is matched by all users in the cloud question bank and entered into the wrong question book and the number of users. Users can know the error rate of the wrong questions entered by themselves among all users, so as to selectively learn the wrong questions entered. If the wrong question rate of entered wrong questions is high, for example, the user thinks that the wrong question rate is higher than 30% of the test questions are generally error-prone and need to be reviewed, and this part of the test questions will be marked according to the needs of users.

S207. It is recommended to review wrong questions.

The recommended review questions include:

If the frequency of the wrong question exceeds the preset first threshold, it is recommended to review the wrong question, or recommend reviewing the wrong question and related content;

If the frequency of wrong questions for the same knowledge point exceeds the preset second threshold, it is recommended to review the wrong questions for this knowledge point;

If the frequency of wrong questions with the same reason for wrong questions exceeds the preset third threshold, it is recommended to review the wrong questions for that reason.

Here, there is no specific regulation on the values of the preset first threshold, the preset second threshold and the preset third threshold, and the preset first threshold, the preset second threshold and the preset third threshold may all be the same, Both may be different, or both may be the same.

Users can set the preset first threshold, preset second threshold and preset third threshold according to their own learning intensity, learning frequency, learning ability, etc. For example, when the user is learning a certain unit, if the user sets The preset first threshold is 3 times, 5 times, 7 times or 10 times. When the frequency of the wrong question exceeds 3 times, 5 times, 7 times or 10 times, it is considered that the frequency of the wrong question is too high. If the frequency of wrong questions is too high and the related content of wrong questions will be repeated for review and recommendation; if the preset second threshold is set to 2, 4, 6 or 8 times, when the frequency of wrong questions on the same knowledge point exceeds 2, 4, 6 or 8 times means that the knowledge point is not firmly grasped, and the review recommendation will be repeated for the knowledge point and the test questions corresponding to the knowledge point; if the preset third threshold is set to 3 times, 5 times times, 7 times or 10 times, when the frequency of wrong questions with the same reason for wrong questions exceeds 3 times, 5 times, 7 times or 10 times, it is considered that the frequency is too high. The associated content will be repeatedly reviewed and recommended. In addition, it is also possible to periodically review and recommend the electronic error book according to the user's needs, such as once a week or once a month, and the specific recommendation period can be set according to the user's needs.

Obtain the associated content of the wrong question from the cloud question bank, enter the related content into the electronic wrong question bank, and can also associate the wrong question when the user studies the wrong question in the electronic wrong question book Content learning can effectively improve the user's learning efficiency; record the wrong reasons for the wrong questions, count the frequency of the wrong questions, and classify them according to the attributes of the wrong questions, which is more convenient for users to use when learning. Classification for targeted learning, in addition to recommending review of wrong questions, according to the frequency of wrong questions, the frequency of wrong questions for the same knowledge point, or the frequency of wrong questions for the same reason for wrong questions, targeted review recommendations to further improve improve the user's learning efficiency.

The present invention recognizes the characters of the wrong question pictures, uses the keywords of the wrong questions, obtains the wrong questions from the cloud question bank or the background server and enters the electronic wrong question book, improves the efficiency of wrong question entry, and is beneficial to classify the wrong questions Organize and review recommendations.

Preferably, after S203, it also includes:

S208. If there is no matching test question, feed back the wrong question to the background server, and the background server will enter the wrong question into the electronic wrong question database.

S209. Enter the wrong question into the cloud question bank.

In actual operation, there may be wrong questions that are made by the teacher himself. At this time, the wrong questions may not match the same or similar test questions in the cloud question bank. At this time, the wrong questions will be fed back to the backend server, and the backend server will complete the test. The input of the wrong questions avoids the situation that the wrong questions cannot be entered when the same or similar test questions cannot be matched in the cloud question bank. After the wrong question is entered into the electronic question bank, the wrong question is entered into the cloud question bank, which can further expand the amount of test questions in the cloud question bank. Match the same or similar test questions from the cloud question bank.

S206. Recording the reasons for the wrong questions, counting the frequency of the wrong questions, and classifying the wrong questions.

Since the wrong question entered at this time is an unmatched test question in the cloud question bank, the wrong question attributes of the wrong question cannot be obtained from the cloud question bank; the attributes include grade, subject, and knowledge points. At this time, the classification of the wrong question is based on the reason of the wrong question, and when the wrong question is entered again, the frequency of the wrong question is counted.

S207. It is recommended to review wrong questions.

The recommended review of wrong questions specifically includes: if the frequency of wrong questions for the same reason for wrong questions exceeds a preset third threshold, recommending review of wrong questions for this reason for wrong questions.

Here, there is no specific regulation on the value of the preset third threshold, which can be set differently according to the user's learning ability, learning frequency, and learning time each time.

Users can set the preset third threshold according to their own learning intensity, learning frequency, each learning time, learning ability, etc. For example, when the user is learning a certain unit, if the preset third threshold is set to 3 times, 5 times, 7 times or 10 times, when the frequency of wrong questions with the same reason for wrong questions exceeds 3 times, 5 times, 7 times or 10 times, it is considered that the frequency is too high. The associated content will be repeatedly reviewed and recommended. In addition, it is also possible to periodically review and recommend the electronic error book according to the user's needs, such as once a week or once a month, and the specific recommendation period can be set according to the user's needs.

This embodiment effectively solves the situation that when the wrong question cannot be matched with the same or similar test questions in the cloud question bank, the wrong question is fed back to the backend server, and the backend server enters the wrong question into the electronic wrong question bank , and enter the wrong questions into the cloud question bank, which improves the efficiency of wrong question entry and expands the amount of test questions in the cloud question bank.

The present invention recognizes the characters of the wrong question pictures, uses the keywords of the wrong questions, obtains the wrong questions from the cloud question bank or the background server and enters the electronic wrong question book, improves the efficiency of wrong question entry, and is beneficial to classify the wrong questions Organize and review recommendations.

Embodiment Three

The wrong question management system provided in this embodiment corresponds to Embodiment 1. Please refer to Embodiment 1 for details not detailed in this embodiment.

As shown in Figure 3, a wrong question management system includes:

Acquisition module: used to obtain pictures of wrong questions;

Recognition module: used to perform character recognition on the picture to obtain the keyword of the wrong question;

Matching module: used to use the keywords to match the test questions in the cloud question bank;

Input module: used to input the same or similar test questions into the electronic wrong question database; if there is no matching test question, feed back the wrong questions to the backend server, and the backend server will enter the wrong questions to the electronic wrong question bank.

Embodiment four

The wrong question management system provided in this embodiment corresponds to Embodiment 2. Please refer to Embodiment 2 for details not detailed in this embodiment.

A wrong question management system, comprising:

Acquisition module: used to obtain pictures of wrong questions;

Recognition module: used to perform character recognition on the picture to obtain the keyword of the wrong question;

Matching module: used to use the keywords to match the test questions in the cloud question bank;

Input module: used to input the same or similar test questions into the electronic wrong question database; if there is no matching test question, feed back the wrong questions to the backend server, and the backend server will enter the wrong questions to the electronic wrong question bank.

Classification module: used to record the reasons for the wrong questions, count the frequency of the wrong questions, and classify the wrong questions.

Review recommendation module: used to recommend review wrong questions.

The present invention recognizes the characters of the wrong question pictures, uses the keywords of the wrong questions, obtains the wrong questions from the cloud question bank or the background server and enters the electronic wrong question book, improves the efficiency of wrong question entry, and is beneficial to classify the wrong questions Organize and review recommendations.

The technical principles of the embodiments of the present invention have been described above in conjunction with specific embodiments. These descriptions are only for explaining the principles of the embodiments of the present invention, and cannot be interpreted in any way as limiting the scope of protection of the embodiments of the present invention. Those skilled in the art do not need to Other specific implementation manners of the embodiments of the present invention can be conceived with creative efforts, and these manners will all fall within the protection scope of the embodiments of the present invention.

Claims (9)

1. a kind of wrong topic management method, which is characterized in that including：

Obtain the picture of wrong topic；

Character recognition is carried out to the picture, obtains the keyword of the wrong topic；

The examination question of high in the clouds exam pool is matched using the keyword；

It, will be in the test questions input to electronics mistake exam pool if matching same or similar examination question；If, will without matched examination question The wrong topic feeds back to background server, and the wrong topic is entered into electronics mistake exam pool by background server；

It is described that the examination question of high in the clouds exam pool is matched using the keyword, specially：Use the keyword and the pass The synonym of key word matches the examination question of high in the clouds exam pool.

2. wrong topic management method according to claim 1, which is characterized in that after the matching, further include：Described in record The wrong topic reason of mistake topic, counts the frequency of the wrong topic, sorts out to mistake topic；

It is described that mistake topic is sorted out, specially：Sorted out according to the wrong attribute of inscribing；

The attribute includes grade, subject, knowledge point.

3. wrong topic management method according to claim 1, which is characterized in that described to inscribe the test questions input to electronics mistake After in library, further include：The association content is entered into electronics mistake by the association content that the wrong topic is obtained from the exam pool of high in the clouds In exam pool.

4. wrong topic management method according to claim 1, which is characterized in that it is described by background server by the wrong questions record After entering into electronics mistake exam pool, further include：The wrong topic is entered into the high in the clouds exam pool.

5. wrong topic management method according to claim 2, which is characterized in that it is described mistake topic is sorted out after, also wrap It includes：Recommend to review wrong topic.

6. wrong topic management method according to claim 5, which is characterized in that wrong topic is reviewed in the recommendation, including：

If the frequency of mistake topic is more than preset first threshold, recommends to review mistake topic or recommend to review mistake topic and association content；

If the frequency of the wrong topic of same knowledge point is more than preset second threshold, recommend the wrong topic for reviewing the knowledge point；

If the frequency of the wrong topic of same wrong topic reason is more than preset third threshold value, recommend the wrong topic for reviewing mistake topic reason.

7. a kind of wrong topic management system, which is characterized in that including：

Acquisition module：Picture for obtaining wrong topic；

Identification module：For carrying out character recognition to the picture, the keyword of the wrong topic is obtained；

Matching module：For using the keyword to match the examination question of high in the clouds exam pool；

Recording module：If, will be in the test questions input to electronics mistake exam pool for matching same or similar examination question；If no The wrong topic is fed back to background server by matched examination question, and the wrong topic is entered into electronics mistake exam pool by background server In；

The matching module is specifically used for：Using the synonym of the keyword and the keyword to the examination question of high in the clouds exam pool into Row matching.

8. wrong topic management system according to claim 7, which is characterized in that further include classifying module：It is described for recording The wrong topic reason of mistake topic, counts the frequency of the wrong topic, sorts out to mistake topic.

9. wrong topic management system according to claim 8, which is characterized in that further include reviewing recommending module：For recommending Review wrong topic.