A test question precision duplicate checking method and system

Abstract

The application relates to the technical field of test question duplication checking, and specifically discloses a test question precision duplication checking method and system, which comprises the following steps: constructing a test question database, collecting and storing test question title information and answer information; processing the test question information, obtaining a duplication checking test question, preprocessing the duplication checking test question, and then extracting the text information by sentence decomposition with punctuation marks as intervals; performing traversal duplication checking comparison between the duplication checking test question and the test question database title, obtaining a repeated signal; marking and displaying the repeated signal in the database title in the database obtained through the traversal duplication checking, comparing the number of words of the duplication checking test question with the number of words of the database title, preliminarily judging the coincidence degree of the test question, and when the keyword coincidence degree is high, it is indicated that the meanings of the duplication checking test question and the target test question are similar; and based on the total value difference of the test question text sentences of the duplication checking test question and the similar text sentence meaning, the duplication checking test question and the target test question are calculated and judged to be repeated or not.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of test paper duplication checking, and particularly relates to a test paper accurate duplication checking method and system. BACKGROUND

[0002] There are two kinds of existing test paper duplication checking methods: one is a method of checking duplication by using inverted index, which has fast search speed, but has the following three shortcomings: first, it does not support duplication checking when part of the document is updated; second, if new documents are to be searched, the entire inverted index needs to be updated, which limits the amount of data contained in the index and the frequency of index updates; third, it does not support duplication checking based on test paper semantics. The other commonly used test paper duplication checking method is a method of checking duplication by using algorithm.

[0003] The above-mentioned duplication checking methods can only check duplication according to the same text during the duplication checking process, and the duplication checking effect is not good: the accuracy of the duplication checking result is low and the entire duplication checking speed is too slow. SUMMARY

[0004] The present application aims to provide a test paper accurate duplication checking method and system to solve the problems in the background.

[0005] The object of the present application can be achieved by the following technical solutions:

[0006] A test paper accurate duplication checking method comprises the following steps:

[0007] Step 1: Construct a test paper database to collect and store the question information and answer information of test papers;

[0008] Process the test paper information as follows: perform text stripping on the test paper information to obtain text information; perform sentence decomposition extraction on the text information with punctuation marks as intervals; and perform keyword extraction on each decomposed sentence;

[0009] Step 2: Obtain a duplication checking test paper, and pre-process the duplication checking test paper, including: performing text cleaning on the test paper information to obtain text information, and then performing sentence decomposition extraction on the text information with punctuation marks as intervals;

[0010] Step 3: Traverse and compare the duplication checking test paper with the question of the test paper database to obtain a repeated signal;

[0011] Step 4: Mark and display the repeated signal in the database.

[0012] As a further scheme of the present application: in the step one, the text information includes: the title text, i.e. the text information of each question in the database, including: the total value of the question text sentence KZ; also including the number of words in each text sentence KS i , and the number of keywords in each question text sentence is obtained and marked as KG i ; wherein i is 1, 2, 3, …, and i is a positive integer; i is used for marking the sentence in the question text.

[0013] As a further scheme of the present application: in the step two, the text information is decomposed and extracted: including obtaining the total value of the duplicate checking text sentence and the number of words in each sentence in the duplicate checking question text; and marked as CCZ and CCS j ; wherein j is 1, 2, 3, …, and j is a positive integer; j is used for marking the sentence in the duplicate checking question text;

[0014] At the same time, the keywords of the duplicate checking question are extracted, including the keywords and the number of keywords in each question text sentence in the text information, and marked as CCG j .

[0015] As a further scheme of the present application: in the step three, the duplicate checking question is compared with the question in the database, including:

[0016] A1: obtaining the data information of the title of the question in the database, calculating the total value difference ZZC of the question text sentence, and obtaining the first-level duplicate checking signal;

[0017] A2: based on the primary duplicate checking signal, comparing the keywords in each question text sentence in the duplicate checking question with the keywords in the question text, and obtaining the number of keyword matches in each question text sentence, and marked as PP i ;

[0018] The meaning similarity ratio WBS i of each question text sentence in the duplicate checking question is calculated;

[0019] The average value WSBJ of the meaning similarity ratio of the text sentence is calculated;

[0020] The average value WSBJ of the meaning similarity ratio of the text sentence is compared with the threshold value WSBY of the meaning similarity ratio of the text sentence; the second-level duplicate checking signal is obtained;

[0021] A3: based on the second-level duplicate checking signal: obtaining the number of words CCS j in the text sentence in the duplicate checking question and the number of words KS i in the question text; calculating the duplicate checking comprehensive ratio CZHB, and then obtaining the duplicate signal.

[0022] As a further scheme of the present application: in the A1, the total value difference value ZZC of the test text sentence is compared with the total value difference threshold ZZCy of the test text sentence;

[0023] If 0≤the total value difference value ZZC of the test text sentence≤the total value difference threshold ZZCy of the test text sentence; a first-level duplicate checking signal is generated;

[0024] If the total value difference value ZZC of the test text sentence>the total value difference threshold ZZCy of the test text sentence; a difference signal is generated.

[0025] As a further scheme of the present application: in the A2, if the text sentence meaning similarity ratio average WSBJ is greater than or equal to the text sentence meaning similarity ratio threshold WSBY, a second-level duplicate checking signal is generated;

[0026] If the text sentence meaning similarity ratio average WSBJ is less than the text sentence meaning similarity ratio threshold WSBY, a difference signal is generated.

[0027] As a further scheme of the present application: the calculation process of the duplicate checking comprehensive ratio CZHB includes:

[0028] The meaning deviation value HP of each text sentence in the duplicate checking test is obtained by calculation j ;

[0029] The total meaning deviation value HPZ of the text sentence in the duplicate checking test is obtained by calculation;

[0030] The duplicate checking comprehensive ratio CZHB is obtained by calculation, wherein a is a sentence difference coefficient, b is a meaning similarity coefficient, and c is a meaning deviation coefficient.

[0031] As a further scheme of the present application: the duplicate checking comprehensive ratio CZHB is compared with the duplicate checking comprehensive ratio threshold CZHBy;

[0032] If the duplicate checking comprehensive ratio CZHB is greater than or equal to the duplicate checking comprehensive ratio threshold CZHBy, a duplicate signal is generated;

[0033] If the duplicate checking comprehensive ratio CZHB is less than the duplicate checking comprehensive ratio threshold CZHBy, a difference signal is generated.

[0034] As a further scheme of the present application: in the step four, based on the duplicate signal, the answer information of the target test can also be displayed.

[0035] As a further scheme of the present application: a test accurate duplicate checking system includes:

[0036] A database construction module: used for constructing a test database, collecting and storing the question information and answer information of the test;

[0037] The process of processing the test question information is as follows: text stripping is performed on the test question information to obtain text information; the text information is divided into sentences by punctuation marks; and key words are extracted from each divided sentence;

[0038] The data acquisition module is used for acquiring the duplicate test questions and pre-processing the duplicate test questions, and the processing process includes: text cleaning is performed on the test question information to obtain text information, and then the text information is divided into sentences by punctuation marks;

[0039] The duplicate checking module is used for comparing the duplicate test questions with the test questions in the test question database to obtain a duplicate signal;

[0040] The display module is used for marking and displaying the titles of the test questions in the database with the duplicate signal obtained by the duplicate checking.

[0041] The beneficial effects of the present application are as follows:

[0042] In the present application, the number of words of the duplicate test question is compared with the number of words of the title of the test question in the database to preliminarily judge the coincidence degree of the test question, and then the key words in each sentence of the duplicate test question are compared with the key words in the target test question; when the key words have a high coincidence degree, it indicates that the duplicate test question and the target test question have similar meanings, and based on this, the sentences of the duplicate test question and the target test question are compared, and the total value difference of the test question text sentences of the duplicate test question and the similarity of the text sentence meanings are calculated and judged to determine whether the duplicate test question and the target test question are repeated, and then it is determined whether the test question database contains a similar test question to the duplicate test question; the accuracy is higher and the effect is better;

[0043] The duplicate checking of the test questions in the database can be used for the update of the test question database by teachers, and can be used for test question retrieval in the learning process of students to ensure that the learning process of students can be better based on the same type of test questions for consolidation learning; and can also be used for searching for test question answer analysis. BRIEF DESCRIPTION OF DRAWINGS

[0044] The present application will be further described below with reference to the accompanying drawings.

[0045] Figure 1 is a schematic diagram of the method flow structure of the present application;

[0046] Figure 2 is a schematic diagram of the duplicate checking method flow in the present application;

[0047] Figure 3 is a system block diagram in the present application. DETAILED DESCRIPTION

[0048] With reference to the accompanying drawings, the technical solutions in the embodiments of the present application will be clearly and completely described below, obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work belong to the protection scope of the present application.

[0049] Embodiment one

[0050] Please refer to Figure 1 The present application is a kind of test item accurate duplicate checking method, comprising the following steps:

[0051] Step one: build a test item database, collect and store the test item information and answer information;

[0052] The test item information is processed, the process is as follows: the text information is obtained by text stripping of the test item information; the text information is extracted by sentence decomposition with punctuation marks as interval; and the keywords of each decomposed sentence are extracted, it should be noted that in the process of keyword extraction of text information, the keywords of each sentence are extracted by using Word2Vec technology;

[0053] The text information includes the text of the title, that is, the text information of each test item in the database, including: the total value KZ of the text of the title; also including the word value KS of each text sentence in the title text i , and the number of keywords of each title text sentence is obtained and marked as KG i ; wherein i is 1, 2, 3…, and i is a positive integer; i is used for the sentence mark in the title text;

[0054] Step two: obtain the duplicate checking test item, and pretreat the duplicate checking test item, the processing process includes: text cleaning of the test item information to obtain text information, and then text information is extracted by sentence decomposition with punctuation marks as interval; the total value of the duplicate checking text and the word value of each sentence in the duplicate checking test item text are obtained; and marked as CCZ and CCS j ; wherein j is 1, 2, 3…, and j is a positive integer; j is used for the sentence mark in the duplicate checking title text;

[0055] At the same time, the keywords of the duplicate checking test item are extracted, including the keywords and the number of keywords of each title text sentence in the text information, and marked as CCG j ;

[0056] Step three: traverse duplicate checking comparison between the duplicate checking test item and the test item database title; as Figure 2 shown, specifically including:

[0057] A1: obtaining the data information of the target title from the database, and comparing the data information of the target title with the data information of the target title in the database through calculating the total value difference value ZZC of the text sentences of the test questions;

[0058] comparing the total value difference value ZZC of the text sentences of the test questions with the total value difference threshold ZZCy of the text sentences of the test questions;

[0059] if 0≤total value difference value ZZC of the text sentences of the test questions≤total value difference threshold ZZCy of the text sentences of the test questions; generating a primary duplicate checking signal;

[0060] if total value difference value ZZC of the text sentences of the test questions>total value difference threshold ZZCy of the text sentences of the test questions; generating a difference signal;

[0061] A2: based on the primary duplicate checking signal, comparing the key words in each text sentence of the duplicate checking test questions with the key words in the target title text, obtaining the key word matching number value of each text sentence of the test questions, and marking it as PP i ;

[0062] calculating the meaning similarity ratio WBS of each text sentence of the duplicate checking test questions i ;

[0063] calculating the average meaning similarity ratio WSBJ of the text sentences of the duplicate checking test questions; comparing the average meaning similarity ratio WSBJ of the text sentences with the text sentence meaning similarity ratio threshold WSBY;

[0064] if the average meaning similarity ratio WSBJ of the text sentences is greater than or equal to the text sentence meaning similarity ratio threshold WSBY, a secondary duplicate checking signal is generated;

[0065] if the average meaning similarity ratio WSBJ of the text sentences is less than the text sentence meaning similarity ratio threshold WSBY, a difference signal is generated;

[0066] A3: based on the secondary duplicate checking signal: obtaining the text sentence word number value CCS of the duplicate checking test questions j and the word number value KS of the text sentences of the target title i ;

[0067] calculating the meaning deviation value HP of each text sentence of the duplicate checking test questions j ;

[0068] calculating the total meaning deviation value HPZ of the text sentences of the duplicate checking test questions through HPZ=∑HP j ;

[0069]

[0070] calculating the total meaning deviation value HPZ of the text sentences of the duplicate checking test questions through HPZ=∑HP j ; ​Calculate the duplicate comprehensive ratio CZHB, wherein a is the statement difference coefficient, b is the meaning similarity coefficient, and c is the meaning deviation coefficient;

[0071] Compare the duplicate comprehensive ratio CZHB with the duplicate comprehensive ratio threshold CZHBy;

[0072] If the duplicate comprehensive ratio CZHB is greater than or equal to the duplicate comprehensive ratio threshold CZHBy, a duplicate signal is generated;

[0073] If the duplicate comprehensive ratio CZHB is less than the duplicate comprehensive ratio threshold CZHBy, a difference signal is generated;

[0074] It should be noted that when the test question database is traversed for duplicate checking, after obtaining the test question under the duplicate signal, the remaining test questions in the test question database still need to be subsequently checked for duplicates, so that all test questions similar to the duplicate test questions in the test question database are displayed for the duplicate checker to view;

[0075] Step four: marking and displaying the target test question in the database with the duplicate signal obtained by the traversal duplicate checking; meanwhile, the answer information of the target test question can also be displayed;

[0076] First, compare the number of words of the duplicate test question with the number of words of the target test question to preliminarily judge the coincidence degree of the test question, and then compare the keywords in each sentence of the duplicate test question with the keywords in the target test question; when the keywords have a high coincidence degree, it means that the duplicate test question and the target test question have similar meanings, etc. Based on this, the sentences of the duplicate test question and the target test question are compared, and whether the duplicate test question and the target test question are duplicated is further calculated and judged based on the total value difference of the test question text sentences of the duplicate test question and the text sentence meaning similarity ratio, so as to judge whether the test question database contains a test question similar to the duplicate test question; the accuracy is higher and the effect is better;

[0077] Duplicate checking of the database test question can be used for updating of the test question library by teachers, and can be used for test question retrieval in the learning process of students, so that the learning process of students can be better based on the same type of test question for consolidation learning; it can also be used for searching of test question answer analysis, etc.

[0078] Embodiment two

[0079] As shown in Figure 3 A test question accurate duplicate checking system, comprising:

[0080] A database construction module for constructing a test question database, collecting and storing test question information and answer information;

[0081] The process of processing the test question information is as follows: text stripping is performed on the test question information to obtain text information; the text information is divided into sentences with punctuation marks as intervals; and key words are extracted from each divided sentence;

[0082] The data acquisition module is configured to acquire the duplicate test question and pre-process the duplicate test question, and the processing process includes: text cleaning is performed on the test question information to obtain text information, and then the text information is divided into sentences with punctuation marks as intervals;

[0083] The duplicate detection module is configured to compare the duplicate test question with the test question database to obtain a duplicate signal;

[0084] The display module is configured to mark and display the title of the database question with the duplicate signal.

[0085] The number of words of the duplicate test question is compared with the number of words of the title of the database question to preliminarily judge the coincidence degree of the test question, and then the key words in each sentence of the duplicate test question are compared with the key words in the target test question; when the key words have a high coincidence degree, it means that the duplicate test question and the target test question have similar meanings, based on which, the sentences of the duplicate test question and the target test question are compared, and the total value difference of the test question text sentences of the duplicate test question and the similarity of the text sentence meanings are calculated and judged to determine whether the duplicate test question and the target test question are repeated, and whether the test question database contains a similar question to the duplicate test question is determined;

[0086] The duplicate detection of the database test question can be used for the teacher to update the test question database, can be used for the test question retrieval in the student learning process, so that the student learning process can be better based on the same type of test question for consolidation learning, and can also be used for searching the test question answer analysis.

[0087] In the description of the present specification, the description of the terms "one embodiment", "example", "specific example" and the like means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are contained in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the described specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0088] The above formulas are all dimensionless values calculated, the formulas are obtained by collecting a large amount of data to simulate a formula of the most recent real situation, and preset parameters in the formulas are set by a person skilled in the art according to actual conditions. The preferred embodiments of the application disclosed above are only used to help explain the application. The preferred embodiments do not describe all the details and limit the application to the specific implementation. Obviously, according to the content of the specification, many modifications and changes can be made. The specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the application, so that those skilled in the art can well understand and use the application. The application is limited by the claims and the entire scope and equivalents thereof.

[0089] The above describes one embodiment of the application in detail, but the content described is only a preferred embodiment of the application and cannot be considered as limiting the implementation scope of the application. Any equivalent changes and improvements made according to the scope of the application should still belong to the patent coverage of the application.

Claims

1. A method for accurately checking duplicate test questions, characterized in that, It comprises the following steps: Step one: build a test database, collect and store test information and answer information of the test; Process the test information as follows: text stripping is performed on the test information to obtain text information; the text information is divided into sentences with punctuation marks as intervals for extraction; And the key word extraction is carried out to each decomposed sentence; the text information includes: the test text, i.e. the text information of each test in the database, including: the total value KZ of the test text sentence; also including the test text sentence word value KS in each text sentence i At the same time, the number of keywords of each test text sentence is obtained and marked as KG i ; wherein i is 1, 2, 3…, and i is a positive integer; i is used for marking the sentence in the test text Step two: obtain the duplicate checking test paper, and pre-process the duplicate checking test paper, the processing process including: text cleaning of the duplicate checking test paper information to obtain text information, and then performing sentence decomposition extraction on the duplicate checking test paper text information with punctuation marks as intervals; decomposing and extracting the duplicate checking test paper text information: including obtaining the total value of the duplicate checking text sentences and the value of the number of words in each sentence in the duplicate checking test paper text; and marking CCZ and CCS respectively j ; wherein, j is 1, 2, 3…, and j is a positive integer; j is used for marking the sentences in the duplicate checking test paper text; Meanwhile, the key words of the duplicate checking test questions are extracted, the key word information of each test text sentence in the duplicate checking test question text information is extracted, including the key words and the number of key words, and is marked as CCG j ; Step three: compare the duplicate test with the test database test for traversal duplicate comparison to obtain a duplicate signal; Step four: mark and display the target test in the database with the duplicate signal obtained by traversal duplicate comparison; In step three, the duplicate test is compared with the test database test for traversal duplicate comparison, which comprises: A1: obtain the data information of the target test from the database, calculate the total value difference value ZZC of the test text sentence, and obtain the first-level duplicate signal; A2: Based on the primary duplicate detection signal, compare the keywords in each test question text in the duplicate detection test question with the keywords in the target test question text, and get the keyword matching quantity value of each test question text, and mark it as PP i ; By The similarity of the meaning of each text sentence in the duplicate checking test paper is calculated as WBS i ; Further by The similarity ratio of the meaning of the text sentence is calculated as WSBJ. Compare the average value WSBJ of the text sentence meaning similarity ratio with the text sentence meaning similarity ratio threshold WSBY; obtain the second-level duplicate signal; A3: Based on the secondary duplicate checking signal: obtain the text sentence character number value CCS in the duplicate checking test question j and the character number value KS of the target test question text sentence i ; calculate the duplicate checking comprehensive ratio CZHB, and then obtain the duplicate signal; In A1, compare the total value difference value ZZC of the test text sentence with the total value difference threshold ZZCy of the test text sentence; If 0≤total value difference value ZZC of the test text sentence≤total value difference threshold ZZCy of the test text sentence; generate a first-level duplicate signal; If the total value difference value ZZC of the test text sentence is greater than the total value difference threshold ZZCy of the test text sentence; generate a difference signal; If the average value WSBJ of the text sentence meaning similarity ratio is greater than or equal to the text sentence meaning similarity ratio threshold WSBY, generate a second-level duplicate signal; If the average value WSBJ of the text sentence meaning similarity ratio is less than the text sentence meaning similarity ratio threshold WSBY, generate a difference signal; The calculation process of the duplicate comprehensive ratio CZHB comprises: By The meaning deviation value HP of each text sentence in the duplicate checking test question is calculated j ; Further through The meaning deviation total value HPZ of the text sentence in the duplicate checking test question is calculated. By The duplicate checking comprehensive ratio CZHB is calculated, wherein a is a statement difference coefficient, b is a meaning similarity coefficient, and c is a meaning deviation coefficient. Compare the duplicate comprehensive ratio CZHB with the duplicate comprehensive ratio threshold CZHBy; If the duplicate comprehensive ratio CZHB is greater than or equal to the duplicate comprehensive ratio threshold CZHBy, a duplicate signal is generated; If the duplicate comprehensive ratio CZHB is less than the duplicate comprehensive ratio threshold CZHBy, a difference signal is generated.

2. The test item precision duplicate detection method according to claim 1, characterized in that, In step four, based on the duplicate signal, the answer information of the target test is displayed.

3. A test question precision duplicate checking system, characterized in that, The system is used to perform the test accurate duplicate checking method according to any one of claims 1-2, which comprises: A database building module for building a test database, collecting and storing test information and answer information of the test; Process the test information as follows: text stripping is performed on the test information to obtain text information; the text information is divided into sentences with punctuation marks as intervals for extraction; and key words are extracted from each decomposed sentence; A data acquisition module for acquiring a duplicate test, pre-processing the duplicate test, and the processing process comprising: text cleaning is performed on the test information to obtain text information, and then the text information is divided into sentences with punctuation marks as intervals for extraction; A duplicate checking module for comparing the duplicate test with the test database test for traversal duplicate comparison to obtain a duplicate signal; A display module for marking and displaying the target test in the database with the duplicate signal obtained by traversal duplicate comparison.

Images