CN110083808A - Cheating determination method, device, equipment and storage medium based on user's answer - Google Patents
Cheating determination method, device, equipment and storage medium based on user's answer Download PDFInfo
- Publication number
- CN110083808A CN110083808A CN201910203513.5A CN201910203513A CN110083808A CN 110083808 A CN110083808 A CN 110083808A CN 201910203513 A CN201910203513 A CN 201910203513A CN 110083808 A CN110083808 A CN 110083808A
- Authority
- CN
- China
- Prior art keywords
- paper
- detected
- similarity
- answer
- cheating
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000001514 detection method Methods 0.000 claims abstract description 54
- 238000012360 testing method Methods 0.000 claims abstract description 53
- 238000013179 statistical model Methods 0.000 claims abstract description 48
- 238000012550 audit Methods 0.000 claims description 14
- 239000000178 monomer Substances 0.000 claims description 3
- 230000006870 function Effects 0.000 description 6
- 238000007726 management method Methods 0.000 description 6
- 241000009298 Trigla lyra Species 0.000 description 5
- 230000002159 abnormal effect Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 239000000203 mixture Substances 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 235000013399 edible fruits Nutrition 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000005194 fractionation Methods 0.000 description 2
- 230000007115 recruitment Effects 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/10—Text processing
- G06F40/194—Calculation of difference between files
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/105—Human resources
- G06Q10/1053—Employment or hiring
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Artificial Intelligence (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Data Mining & Analysis (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Electrically Operated Instructional Devices (AREA)
Abstract
The invention discloses a kind of cheating determination method, device, equipment and storage mediums based on user's answer.The described method includes: the paper to be detected comprising subjective item is transferred from database according to paper unique number;Obtain single topic similarity corresponding with each subjective item in paper to be detected;Single topic similarity of each subjective item is input to preset statistical model, receives the overall similarity of the paper to be detected of statistical model output;The cheating testing result of paper to be detected, and the preset detection zone by overall similarity and cheating testing result associated storage into paper to be detected are obtained according to overall similarity.The present invention realizes automation, efficient the detection and plagiarizing of practising fraud is assert, and reduces human input.
Description
Technical field
The present invention relates to intelligent decision fields, and in particular to a kind of cheating determination method based on user's answer, is set device
Standby and storage medium.
Background technique
Currently, especially in the written examination link of recruitment examination, the user submitted for subjective item is answered in recruitment
Case, in user's answer there are when similar situation, due to manual reviews' heavy workload, and there are subjective factor influences, therefore are difficult
By manually distinguishing the similitude of user's answer, and then judges user there are acts of plagiarism.And existing examination system, by adopting
Collecting the monitoring image at examination hall scene, detection user whether there is cheating, and in paper scoring process, it is examined without cheating
It surveys.
Based on this, it is necessary to provide a kind of cheating determination method, solve the problems, such as manually to assert and plagiarize difficulty, and reduce
Human input.
Summary of the invention
The embodiment of the present invention provides a kind of cheating determination method, device, equipment and storage medium based on user's answer, answers
For intelligent examination system, a reliable cheating detection instrument is provided for the intelligent examination system, realizes automation, efficient
Change ground cheating detection, reduces human input.
A kind of cheating determination method based on user's answer, comprising:
The paper to be detected comprising subjective item is transferred from database according to paper unique number;
Obtain single topic similarity corresponding with subjective item described in each in the paper to be detected;
Single topic similarity of each subjective item is input to preset statistical model, receives the statistical model output
The overall similarity of the paper to be detected;
Obtain the cheating testing result of the paper to be detected according to the overall similarity, and by the overall similarity
With preset detection zone of the cheating testing result associated storage into the paper to be detected.
A kind of cheating decision maker based on user's answer, comprising:
Module is transferred, for transferring the paper to be detected comprising subjective item from database according to paper unique number;
Single topic detection module, it is similar for obtaining single topic corresponding with subjective item described in each in the paper to be detected
Degree;
Single topic statistical module is received for single topic similarity of each subjective item to be input to preset statistical model
The overall similarity of the paper to be detected of the statistical model output;
Cheating determination module, for obtaining the cheating testing result of the paper to be detected according to the overall similarity,
And the preset detection zone by the overall similarity and the cheating testing result associated storage into the paper to be detected
Domain.
A kind of computer equipment, including memory, processor and storage are in the memory and can be in the processing
The computer-readable instruction run on device, the processor is realized when executing the computer-readable instruction above-mentioned to be answered based on user
The cheating determination method of case.
A kind of computer readable storage medium, the computer-readable recording medium storage have computer-readable instruction, institute
It states and realizes the above-mentioned cheating determination method based on user's answer when computer-readable instruction is executed by processor.
Cheating determination method, device, equipment and storage medium provided by the invention based on user's answer, what be will acquire is each
Single topic similarity is input in preset statistical model, the overall similarity of the statistical model output ready to receive, real
Automation is showed, efficient the detection and plagiarizing of practising fraud is assert, and reduces human input;It simultaneously will cheating testing result and inspection
Measured data is added in the detection zone of the paper to be detected, so that cheating testing result and detection data are evidence-based, is had
Conducive to the unified management of data.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below by institute in the description to the embodiment of the present invention
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the invention
Example, for those of ordinary skill in the art, without creative efforts, can also obtain according to these attached drawings
Obtain other attached drawings.
Fig. 1 is an application environment schematic diagram of the cheating determination method in one embodiment of the invention based on user's answer;
Fig. 2 is a flow chart of the cheating determination method in one embodiment of the invention based on user's answer;
Fig. 3 is a flow chart of the cheating determination method in one embodiment of the invention based on user's answer;
Fig. 4 is a flow chart of the cheating determination method in one embodiment of the invention based on user's answer;
Fig. 5 is a flow chart of the cheating determination method in one embodiment of the invention based on user's answer;
Fig. 6 is a flow chart of the cheating determination method in one embodiment of the invention based on user's answer;
Fig. 7 is a functional block diagram of the cheating decision maker in one embodiment of the invention based on user's answer;
Fig. 8 is a functional block diagram of the cheating decision maker in one embodiment of the invention based on user's answer;
Fig. 9 is a schematic diagram of computer equipment in one embodiment of the invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are some of the embodiments of the present invention, instead of all the embodiments.Based on this hair
Embodiment in bright, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, shall fall within the protection scope of the present invention.
Cheating determination method provided by the invention based on user's answer, can be applicable in the application environment such as Fig. 1,
In, client is communicated by network with server.Wherein, client include but is not limited to be various personal computers, pen
Remember this computer, smart phone, tablet computer, camera and portable wearable device.Server can use independent server
The either server cluster of multiple servers composition is realized.
In one embodiment, it as shown in Fig. 2, providing a kind of cheating determination method based on user's answer, answers in this way
It is illustrated for the server in Fig. 1, comprising the following steps:
S10, the paper to be detected comprising subjective item is transferred from database according to paper unique number.
Wherein, the paper to be detected refers to, the answer paper that cheating detection is carried out after applicant is submitted is remembered
Record is paper to be detected;The paper to be detected includes user's answer, applicant's information and paper unique number etc..
Preferably, the database is provided with administrator library, Kao Shengku, answer library, Pipers database etc..Wherein, administrator library
For storage administrator's information;The examinee library is for storing applicant's information;It is recorded for storing by administrator in the answer library
(Key for Reference can be used for paper to the Key for Reference of the subjective item, the subjective item and above-mentioned subjective item that obtain on the net certainly that enter
The scoring stage);The Pipers database (is currently answered for storing history test paper (the scoring paper of previous applicant) and newly-increased paper
The paper to be detected for the person of engaging).
And preferably, in intelligent examination system, the intelligent examination system is set the server application comprising above-mentioned database
Be equipped with login interface, answer interface, check interface etc..Wherein, login interface is used to log in for applicant and administrator logs in;
The answer interface is for showing answer paper, and the answer interface is provided with answer region for typing when applicant answers
User's answer;The interface of checking only has for showing paper to be scored (having been subjected to the paper to be detected of cheating detection)
The administrator for checking permission could carry out paper scoring and user's evaluation to the paper to be detected checked in interface.
In the present embodiment, after detecting the answer paper that applicant submits, the answer paper that will answer
It is recorded as paper to be detected, and cheating detection is carried out to the paper to be detected immediately.
In another embodiment, after same a collection of applicant completes written examination, more parts of tests to be checked are transferred from database
Volume, in the next steps, sequentially carries out cheating detection to all papers to be detected according to paper unique number.
S20, single topic similarity corresponding with subjective item described in each in the paper to be detected is obtained.
Preferably, the topic information and user's answer of the subjective item are obtained, and then is believed according to the topic of the subjective item
Existing answer paper is (comprising the non-history paper with batch of previous applicant or with a batch in the breath detection Pipers database
Other applicants newly-increased paper) whether there is identical subjective item, there are when identical subjective item in the Pipers database, obtain
The comparison answer of all identical subjective items, at this point, can be obtained by detecting user's answer and the comparison answer
Take single topic similarity of the subjective item.Wherein, the topic information includes keyword in topic, belonging to topic examines or check range
Deng.
In another embodiment, it can directly be adjusted from the answer library according to the topic information of the subjective item of acquisition
The Key for Reference of the subjective item is taken, at this point, the master can be obtained by detecting user's answer and the Key for Reference
Single topic similarity of sight topic.Intelligible, if single topic similarity of certain subjective item is higher, there are plagiarism suspicion is bigger for the topic.
S30, single topic similarity of each subjective item is input to preset statistical model, it is defeated receives the statistical model
The overall similarity of the paper to be detected out.
Specifically, obtaining in the paper to be detected shared by the single topic similarity, single topic similarity of each subjective item
The number of weight or single topic similarity, the entirety that the paper to be detected can be obtained by preset statistical model are similar
Degree.Preferably, the statistical model includes the first statistical model based on average algorithm and the second statistical model based on weight.
S40, the cheating testing result that the paper to be detected is obtained according to the overall similarity, and by the whole phase
Preset detection zone like degree with the cheating testing result associated storage into the paper to be detected.
Wherein, the detection zone refer to preset in paper to be detected identification label (such as: comprising " testing result ", " phase
Like degree " etc. characters title) after white space, and the detection zone can be arranged according to demand.
The cheating testing result includes there is cheating and be not present to practise fraud.
In the present embodiment, detect whether the overall similarity is more than preset cheating threshold value, it is similar in the entirety
When degree is more than preset cheating threshold value, confirm that the paper to be detected has cheating, and by the overall similarity and there is work
The corresponding preset detection zone being added in the paper to be detected of the cheating testing result of disadvantage;And in the overall similarity
When being less than preset cheating threshold value, confirm that the paper to be detected there is no cheating, by the overall similarity and is not present
The corresponding preset detection zone being added in the paper to be detected of the cheating testing result of cheating.Wherein, described preset
Cheating threshold value masks the applicant in the presence of cheating for fishing for out the applicant without cheating.
Further, by the overall similarity and the cheating testing result associated storage into the paper to be detected
Preset detection zone, while other detection datas can also be added to the corresponding position of the predeterminable area, such as: it is every
The corresponding list of one subjective item inscribes similarity, there is the specific field etc. plagiarized in user's answer.It is intelligible, the testing number
According to the overall similarity comprising the corresponding single topic similarity of each subjective item, the paper to be detected and there is the specific paragraph plagiarized
Deng.
In conclusion the cheating determination method provided by the invention based on user's answer, complete with a batch of applicant
After written examination, paper to be detected is transferred according to the unique number of paper, and obtains each subjective item in the paper to be detected respectively
Corresponding single topic similarity, and then each single topic similarity is input in preset statistical model, the system can be obtained
Count the overall similarity of model output;Further, cheating detection knot is obtained according to the overall similarity of the paper to be detected
Fruit, and the cheating testing result and detection data are added in the detection zone of the paper to be detected, it realizes automatic
Change, efficient ground is practised fraud detection and plagiarism identification, and reduce human input;Make practise fraud testing result and detection data simultaneously
It is evidence-based, be conducive to the unified management of data.
In one embodiment, after cheating detection, this batch examination is being obtained with the paper to be detected of batch of applicant
The statistical data of volume, and detect the statistical data and reach preset Threshold Alerts value, if the statistical data reaches preset
Threshold Alerts value, then automatic trigger exception handling instruction, and related data is reset according to abnormal data.Such as: subjective item
Cheating probability reach its corresponding Threshold Alerts value, then replace the higher (needle of cheating probability with the stronger subjective item of specific aim
It is on the weak side to property) subjective item.
Wherein, the statistical data includes the cheating probability etc. of the probability of occurrence of subjective item, subjective item.Such as: database
100 parts of papers to be detected are stored with, if number 10 times that certain subjective item occurs in this batch of paper to be detected, the subjective item
Probability of occurrence is 1/10;If having single topic similarity of the subjective item in 8 parts of papers to be detected equal in 10 parts of papers to be detected
More than default cheating threshold value, then the cheating probability of the subjective item is 8/10.
In another embodiment, after cheating detection, this batch is being obtained with the paper to be detected of batch of applicant
The statistical data of paper, and the statistical data is sent to the administrator (such as: director or manager) for checking permission,
And the administrator have found when checking the statistical data abnormal data (such as: cheating problem concentrates on together in difference test paper
One subjective item), the exception handling instruction comprising abnormal data can be sent to the server, and server receives the exception
When process instruction, related data is reset according to the abnormal data.
In one embodiment, as shown in figure 3, before the step S10, i.e., it is described according to paper unique number from database
In transfer the paper to be detected comprising subjective item before, comprising the following steps:
S101, receive include applicant's account the first log-in instruction, data base querying whether there is and the application
The matched applicant's information of person's account.
Wherein, applicant's account includes that user examines in the name of the login interface typing of intelligent examination system and written examination
Number etc..
First log-in instruction refers to, applicant intelligent examination system login interface typing applicant account it
Afterwards, it clicks preset login button and is sent to server.
Preferably, receive include applicant's account the first log-in instruction, according to applicant's account from the examinee
The library inquiry applicant is which applicant being stored in advance in the examinee library, that is, verifying applicant's identity.
S102, exist and applicant's account applicant's information when, randomly select predetermined number from the database
The subjective item of amount and total score generates answer paper.
Preferably, when there is applicant's information matched with applicant's account, the applicant is prompted to be verified,
The subjective item for randomly selecting predetermined quantity and total score from answer library generates answer paper, and the answer paper Pagination Display is existed
The answer interface of intelligent examination system obtains the answer paper comprising user's answer after answering for the applicant.Such as:
Totally 10 subjective item, subjective item total score are 50 points to answer paper, and one page shows 5 subjective items.
Further, the answer library can be divided into multiple word banks according to examination range, at this point, according to default extraction
Ratio chooses subjective item corresponding with ratio from each word bank, and recombinant generates answer paper.Wherein, examination range includes examination
Communication capability, ability to solve problem, cooperation ability, ability of creative, professional knowledge ability of applicant etc..It is appreciated that
, since the subjective item of answer paper is randomly selected from answer library, the corresponding answer paper of different applicants
It is not identical, be conducive to avoid the mutual plagiarism between different applicants, and improve applicant's achievement authenticity.
Such as: the answer paper of intelligent examination system setting includes five most of, amounts to 20 problems, if according to default extraction
Ratio 1:1:1:1:1 extracts subjective item from answer library, then chooses respectively for examining or check communication capability, ability to solve problem, team
Ability to cooperate, ability of creative, subjective item each 4 of professional knowledge ability.
S103, the submission instruction comprising paper unique number is received, the answer paper answered is recorded as to be checked
Test volume, and by the paper unique number and the paper associated storage to be detected into database.
Wherein, the paper unique number is by the written examination number of examining of applicant, paper submission time, current submission serial number
One or more compositions.
Preferably, when confirming that answer is errorless in applicant, and clicking default submitting button, receiving includes paper unique number
Submission instruction, by the paper unique number and comprising applicant's information, the topic information of each subjective item, user's answer to
Paper associated storage is detected into Pipers database.
In another embodiment, when applicant clicks default submitting button, the corresponding answer of each subjective item is detected
The user master is prompted if user's answer is not detected in the answer region of a certain subjective item with the presence or absence of user's answer in region
Non- answer is inscribed in sight, and user is waited to further confirm that submission;Such as: suggestion content is " the topic non-answer of X, if continue to submit ".
In conclusion the cheating determination method provided by the invention based on user's answer, passes through in applicant's authentication
Later, answer paper is generated from the subjective item that database randomly selects preset quantity and total score, after answering for applicant
Obtain user's answer;And after the answer paper for getting same batch applicant, the answer paper answered is recorded as
Test paper to be detected, and by the paper unique number and the paper associated storage to be detected into database, for server
Cheating detection is carried out, is conducive to the unified management of anti-cheating and paper, improves cheating detection efficiency.
In one embodiment, as shown in figure 4, the step S20, i.e., it is described obtain in the paper to be detected with each
The corresponding single topic similarity of the subjective item, specifically includes the following steps:
S201, the topic information and its user's answer for obtaining each subjective item in the paper to be detected, and will
User's answer of the subjective item obtained is recorded as answer to be detected.
S202, according to the topic information of the subjective item, obtain identical subjective item in other papers from the database
User's answer, and in other the described papers that will acquire user's answer of identical subjective item be recorded as comparison answer.
S203, the answer to be detected and the pure words phase compared between answer are detected by preset matching rule
Like degree and/or semantic similarity, and the subjective item is obtained according to the pure words similarity and/or the semantic similarity
Single topic similarity.
Wherein, the matching rule includes character match and semantic matches;Wherein the character match refers to two characters
The identical character string of identical and number of characters is same character string;Semantic matches refer to that the identical character string of two semantemes is same word
Symbol string, such as: " how old are you " and " you the age how much ".
In the one side of the present embodiment, answer to be detected is split by preset ending symbol and number of words threshold value,
Multiple text segments comprising fractionation label are generated, and character is carried out in the comparison answer according to each text segment
With detection, the pure words similarity of the answer to be detected is obtained, and the pure words similarity is recorded as the subjective item
Single topic similarity.Wherein.The form of the text segment are as follows: [text segment+fractionation label];Such as: [text segment a1].
Such as: certain answer to be detected splits into 15 text segments, and each text segment is 10 characters, in the text
There are identical character string, (character is identical and the identical company of number of characters in the comparison answer for the corresponding character string of segment
Continuous character) when, determine the letter plate section to plagiarize, if detect in comparison answer A there are character string and [text segment a1] and
[text segment a2] is identical, compares in answer B that there are character strings is identical with [text segment a1] and [text segment a3], then this to
The pure words similarity for detecting answer is 20%.
In the another aspect of the present embodiment, after getting multiple text segments, and according to each text segment in institute
It states in comparison answer and carries out semantic matches detection, obtain the semantic similarity of the answer to be detected, and by the answer to be detected
Semantic similarity be recorded as single topic similarity of the subjective item
At the another aspect of the present embodiment, according to the pure words similarity, the corresponding weight of the pure words similarity,
The semantic similarity and the corresponding weight of the semantic similarity obtain single topic similarity of the subjective item.
In one embodiment, the step S30, i.e., it is described single topic similarity of each subjective item is input to it is preset
Statistical model receives the overall similarity of the paper to be detected of the statistical model output, comprising the following steps:
Single topic similarity of each subjective item is input to the first statistical model based on average algorithm, receives described the
The overall similarity of the paper to be detected of one statistical model output;First statistical model are as follows:
Wherein, Q is overall similarity;PiSimilarity is inscribed to be single;N is the number of monomer similarity.
It is intelligible, know each single topic similarity and examination question sum, can be calculated and be obtained by the first statistical model
The overall similarity of the paper to be detected.
Such as: known paper to be detected 5 is inscribed totally, and every topic is 20 points, if single topic similarity of the 1st topic is the 15%, the 2nd
It is single topic similarity of the 90%, the 4th topic is the 10%, the 5th topic that single topic similarity of topic, which is single topic similarity of the 95%, the 3rd topic,
Single topic similarity is 85%, then is 59% using the overall similarity that above-mentioned first statistical model can obtain detection test paper.
In another embodiment, the step S30, i.e., it is described single topic similarity of each subjective item is input to it is default
Statistical model, receive the overall similarity of the paper to be detected of statistical model output, comprising the following steps:
Single topic similarity of each subjective item is input to the second statistical model based on weight, receives second system
Count the overall similarity of the paper to be detected of model output;Second statistical model are as follows:
Wherein, KiFor weight shared by single topic similarity;And
It is intelligible, know weight shared by each single topic similarity and single topic similarity, it can be by described
Second statistical model calculates the overall similarity for obtaining the test paper.Wherein, weight shared by single topic similarity with it is described
Score value to each subjective item in answer paper is directly proportional, and score value is higher, and weight shared by single topic similarity of the subjective item is bigger;
Such as: if subjective entitled examination emphasis and examination difficult point, can be set relative to the above-mentioned heavy difficult point topic of conventional topic compared with high score.
Such as: known paper to be detected totally 5, and the score value distribution situation being arranged is the 1st 10 points of topic, the 2nd topic 10 points, the 3rd
20 points of topic, 30 points of the 4th topic and the 5th 30 points of topic, at this point, can know single topic similarity institute of each topic according to the score value distribution situation
The weight accounted for is respectively 1/10,1/10,2/10,3/10 and 3/10, if single topic similarity of the 1st topic is single topic of the 15%, the 2nd topic
Similarity be the 95%, the 3rd topic single topic similarity be the 90%, the 4th topic single topic similarity be the 10%, the 5th topic list topic it is similar
Degree is 85%, then is 57.5% using the overall similarity that above-mentioned second statistical model can obtain detection test paper.
In one embodiment, as shown in figure 5, before the step S40, comprising the following steps:
Whether S501, each single topic similarity of detection are more than preset dynamic threshold.
S502, single topic quantity of the single topic similarity more than the dynamic threshold is obtained, and detects single topic quantity
It whether is more than preset quantity threshold value.
S503, when single topic quantity is more than preset quantity threshold value, according to the preset adjustment model increase entirety
Similarity;The adjustment model are as follows:
QS=Q ± Δ Q=Q ± | n-n0|*S0*F
Wherein, QSFor overall similarity adjusted;Δ Q is adjusted value;n0For preset quantity threshold value;N is single topic quantity;
S0For dynamic threshold;F is regulation coefficient.
Preferably, there are preset corresponding relationships for regulation coefficient, such as are as follows: regulation coefficient can be for preset quantity threshold value
For the Geometric Sequence or arithmetic progression of first item.
When single topic quantity is more than preset quantity threshold value, the probability which has cheating is higher, at this point,
Increase the overall similarity according to preset adjustment formula, i.e., increases an adjusted value in initial overall similarity and adjusted
Overall similarity after whole.Such as: enabling preset quantity threshold value is 3, and initial overall similarity is 59%, preset cheating threshold
Value is 60%, and regulation coefficient is as shown in table 1;If single topic similarity of subjective item is more than preset dynamic in certain part of paper to be detected
Threshold value (such as: single topic quantity 90%) is 5, then can obtain adjusted value according to preset adjustment formula is 5.4%, after adjustment
Overall similarity be 64.4%;In the next steps, if preset cheating threshold value is 60%, confirm the paper to be detected
There are cheatings.
The regulation coefficient table that 1 preset quantity threshold value of table is 3
Item number | Single topic quantity (unit: a) | Regulation coefficient |
1 | 3 | F (0)=1 |
2 | 4 | F (1)=2 |
3 | 5 | F (2)=3 |
…… | …… | …… |
m | n | F (m)=f (0)+d* (m-1) |
S504, when single topic quantity is less than preset quantity threshold value, keep the overall similarity constant or root
Reduce the overall similarity according to preset adjustment model.
Wherein, the overall similarity is reduced according to preset adjustment formula, i.e., reduces one in initial overall similarity
A adjusted value be adjusted after overall similarity, in the next steps, if overall similarity adjusted is still above preset
Cheating threshold value then confirms that the paper to be detected has cheating;If overall similarity adjusted is less than preset cheating threshold
Value, then confirming the paper to be detected, there is no cheatings.Such as: enabling preset quantity threshold value is 3, and initial overall similarity is
64%, preset cheating threshold value is 60%;If single topic similarity of subjective item is more than preset dynamic in certain part of paper to be detected
Threshold value (such as: single topic quantity 90%) is 2, and regulation coefficient is 2 (identical as the regulation coefficient that single topic quantity is 4), then basis
It is 1.8% that preset adjustment formula, which can obtain adjusted value, and overall similarity adjusted is 62.2%;In the next steps,
If preset cheating threshold value is 60%, confirm that the paper to be detected has cheating.
In conclusion whether the cheating determination method provided by the invention based on user's answer, inscribe quantity by the way that detection is single
Overall similarity is adjusted more than preset quantity threshold value, is conducive to the reliability for improving cheating testing result.
In one embodiment, as shown in fig. 6, the cheating testing result is not comprising having cheating and there is cheating, at this point,
The method also includes following steps:
S601, receive include administrator's account the second log-in instruction, data base querying whether there is and the management
Member's matched administrator information of account.
Wherein, administrator's account include the name of login interface typing of the administrator in intelligent examination system, work number,
Administrative power restricting the number etc..
Preferably, receive include administrator's account the second log-in instruction, according to administrator's account from the management
Member's library inquiry administrator is which administrator being stored in advance in the administrator library, namely verifies administrator's identity.
Intelligible, the extent of competence of each administrator is not identical, such as: the administrator with search access right can check
The answer paper of applicant and the corresponding applicant's information of the answer paper, and there is the administrator for checking permission can only may be used
With check matching applicant answer paper (such as: is there is cheating in cheating testing result, but paper score value is more than
The paper to be detected of default audit threshold value), and the answer paper is audited, but can not know that the answer paper is corresponding
Applicant's information etc.;Meanwhile the Permission Levels of each administrator are not identical, and such as: the line manager in non-staff resource portion is only
It can check the answer paper of the applicant of relevant departments, and the responsible person of Human Resources Department can check the answer examination of all applicants
Volume.
S602, when there is administrator information matched with institute administrator's account, acquisition is matched with administrator's account
And its testing result of practising fraud is the paper to be detected in the presence of cheating.
That is, prompt the administrator to be verified when there is user information matched with second log-on message, into
Enter the access interface of intelligent examination system, while acquisition is matched with administrator's account and its testing result of practising fraud is made to exist
The paper to be detected of disadvantage;And when administrator information matched with second log-on message is not present, prompt is not present
The account.
S603, the paper score value that the paper to be detected is obtained according to preset grading parameters, and detect the paper
Whether score value is more than preset audit threshold value.
Wherein, the grading parameters can with rating staff the intelligent examination system access interface typing score value or
Grade can also be using preset Rating Model to the score value or grade obtained after the paper detection to be detected;It is described
Audit threshold value is arranged according to demand;Such as: 100 points.
In the one side of the present embodiment, each subjective item is provided with different score values (such as: 0~10 point), this
When, according to the score value of every one of subjective item in the paper to be detected of acquisition, count and calculate the examination of the paper to be detected
Roll up score value;And in the one side of the present embodiment, for each subjective item be provided with different grades (such as: pass, one
As, it is good and excellent), and different grades be corresponding with a fixed score value (such as: passing is 6 points, generally 7 points, it is good be 8 points,
Excellent is 10 points etc.), at this point, counting and calculating described according to the grade of every one of subjective item in the paper to be detected of acquisition
The paper score value of paper to be detected.Further, after the paper score value for getting the paper to be detected, institute is detected
State whether paper score value reaches preset audit threshold value.
S604, when the paper score value reaches preset audit threshold value, will be corresponding with the paper score value described
Paper to be detected is shown in client with preset display form.
Wherein, the display form includes list, small icon, large icons etc., and can switch each exhibition according to user demand
Show form.
That is, can be to exist to make by cheating testing result when the paper score value reaches preset audit threshold value
The paper to be detected of disadvantage is recorded as doubtful cheating paper, and by the doubtful cheating paper with list display in the access
Interface wait in list of scoring, for administrator audit, to further determine that described paper of the doubtful cheating paper obtains
Whether score value is true, namely determines whether the cheating testing result of the doubtful cheating paper is effective;And in the paper score value
Not up to preset audit threshold value when, be in the presence of the corresponding applicant of the paper to be detected to practise fraud by testing result of practising fraud
Written examination result queue is not pass through written examination.Intelligible, specific display form can be with are as follows: [paper unique number] [paper name
Claim], for example, BS20171023018- integration capability.
Further, if the administrator determine exist cheating cheating testing result be it is effective, by the applicant
Written examination result queue be do not pass through written examination;And if the administrator determine exist cheating cheating testing result be it is invalid,
Send a prompt packet;For example, the paper score value of the applicant is XX, the interview stage can be paid close attention to.It is intelligible, making
Disadvantage testing result be there is cheating, but paper score value be more than default audit threshold value paper to be detected it is more when, the management
Member has permission the paper to be detected that will be audited and transfers other administrators for having and checking permission.
In conclusion the cheating determination method provided by the invention based on user's answer, after detection-phase of practising fraud, also
It is provided with the paper audit stage, can be avoided that overall similarity is higher but the higher paper to be detected of score value is mistaken for existing and make
The paper to be detected of disadvantage effectively avoids missing the preferred talent.
In one embodiment, as shown in fig. 7, providing a kind of cheating decision maker based on user's answer, user should be based on
The cheating determination method based on user's answer corresponds in the cheating decision maker and above-described embodiment of answer.It should be based on user
The cheating decision maker of answer includes transferring module 110, single topic detection module 120, single topic statistical module 130 and cheating to determine mould
Block 140.Detailed description are as follows for each functional module:
Module 110 is transferred, for transferring the paper to be detected comprising subjective item from database according to paper unique number.
Single topic detection module 120, for obtaining single topic corresponding with subjective item described in each in the paper to be detected
Similarity.
Single topic statistical module 130 connects for single topic similarity of each subjective item to be input to preset statistical model
Receive the overall similarity of the paper to be detected of the statistical model output.
Cheating determination module 140, the cheating for obtaining the paper to be detected according to the overall similarity detect knot
Fruit, and the preset detection by the overall similarity and the cheating testing result associated storage into the paper to be detected
Region.
In one embodiment, the cheating decision maker based on user's answer further includes with lower module, and each functional module is detailed
It is described as follows:
Whether the first login module deposits for receiving the first log-in instruction comprising applicant's account in data base querying
With the matched applicant's information of applicant's account.
Answer module, for being selected at random from the database when there is applicant's information with applicant's account
The subjective item of predetermined quantity and total score is taken to generate answer paper.
Memory module instructs for receiving the submission comprising paper unique number, the answer paper answered is remembered
Record is paper to be detected, and by the paper unique number and the paper associated storage to be detected into database.
In one embodiment, it when the cheating testing result includes not there is no cheating and there is cheating, is answered based on user
The cheating decision maker of case further includes with lower module, and detailed description are as follows for each functional module:
Whether the second login module deposits for receiving the second log-in instruction comprising administrator's account in data base querying
With the matched administrator information of administrator's account.
Matching obtains module, for obtaining and the pipe when there is administrator information matched with institute administrator's account
The matching of reason person's account and its testing result of practising fraud are the paper to be detected in the presence of cheating.
Score value detection module, for obtaining the paper score value of the paper to be detected according to preset grading parameters, and
Detect whether the paper score value is more than preset audit threshold value.
Auditing module will be with the paper score value pair for when the paper score value reaches preset audit threshold value
The paper to be detected answered is shown in client with preset display form.
In one embodiment, as shown in figure 8, single topic detection module 120 includes following submodule, each function sub-modules
Detailed description are as follows:
First record sub module 121, for obtaining the topic information of each subjective item in the paper to be detected
And its user's answer, and the user's answer for the subjective item that will acquire is recorded as answer to be detected.
Second record sub module 122 obtains other for the topic information according to the subjective item from the database
User's answer of identical subjective item in paper, and user's answer of identical subjective item is recorded as in other the described papers that will acquire
Compare answer.
Matched sub-block 123, for passing through answer to be detected described in preset character match rule detection and the comparison
Pure words similarity and/or semantic similarity between answer, and it is similar according to the pure words similarity and/or the semanteme
Degree obtains single topic similarity of the subjective item.
In one embodiment, single topic statistical module 130 includes the first output sub-module, and the function sub-modules are specifically
It is bright as follows:
First output sub-module, for single topic similarity of each subjective item to be input to first based on average algorithm
Statistical model receives the overall similarity of the paper to be detected of the first statistical model output;The first statistics mould
Type are as follows:
Wherein, Q is overall similarity;PiSimilarity is inscribed to be single;N is the number of monomer similarity.
In another embodiment, single topic statistical module 130 includes the second output sub-module, and the function sub-modules are detailed
It is described as follows:
Second output unit, for single topic similarity of each subjective item to be input to the second statistics mould based on weight
Type receives the overall similarity of the paper to be detected of the second statistical model output;Second statistical model are as follows:
Wherein, KiFor weight shared by single topic similarity;And
In one embodiment, the cheating decision maker based on user's answer further includes with lower module, and each functional module is detailed
It is described as follows:
Whether first detection module is more than preset dynamic threshold for detecting each single topic similarity.
Second detection module is more than single topic quantity of the dynamic threshold for obtaining single topic similarity, and detects
Whether single topic quantity is more than preset quantity threshold value.
The first adjustment module is used for when single topic quantity is more than preset quantity threshold value, according to preset adjustment model
Increase the overall similarity;The adjustment model are as follows:
QS=Q ± Δ Q=Q ± | n-n0|*S0*F
Wherein, QSFor overall similarity adjusted;Δ Q is adjusted value;n0For preset quantity threshold value;N is single topic quantity;
S0For dynamic threshold;F is regulation coefficient.
Second adjustment module, for keeping the entirety similar when single topic quantity is less than preset quantity threshold value
It spends constant or the overall similarity is reduced according to preset adjustment model.
Specific restriction about the cheating decision maker based on user's answer may refer to answer above for based on user
The restriction of the cheating determination method of case, details are not described herein.Each mould in the above-mentioned cheating decision maker based on user's answer
Block can be realized fully or partially through software, hardware and combinations thereof.Above-mentioned each module can be embedded in the form of hardware or independence
In processor in computer equipment, it can also be stored in a software form in the memory in computer equipment, in order to
Processor, which calls, executes the corresponding operation of the above modules.
In one embodiment, a kind of computer equipment is provided, which can be server, internal junction
Composition can be as shown in Figure 9.The computer equipment include by system bus connect processor, memory, network interface and
Database.Wherein, the processor of the computer equipment is for providing calculating and control ability.The memory packet of the computer equipment
Include non-volatile memory medium, built-in storage.The non-volatile memory medium be stored with operating system, computer-readable instruction and
Database.The built-in storage provides ring for the operation of operating system and computer-readable instruction in non-volatile memory medium
Border.To realize a kind of cheating determination method based on user's answer when the computer-readable instruction is executed by processor.
In one embodiment, a kind of computer equipment is provided, including memory, processor and storage are on a memory
And the computer-readable instruction that can be run on a processor, processor perform the steps of when executing computer-readable instruction
The paper to be detected comprising subjective item is transferred from database according to paper unique number;
Obtain single topic similarity corresponding with subjective item described in each in the paper to be detected;
Single topic similarity of each subjective item is input to preset statistical model, receives the statistical model output
The overall similarity of the paper to be detected;
Obtain the cheating testing result of the paper to be detected according to the overall similarity, and by the overall similarity
With preset detection zone of the cheating testing result associated storage into the paper to be detected.
In one embodiment, a kind of computer readable storage medium is provided, computer-readable instruction is stored thereon with,
It is performed the steps of when computer-readable instruction is executed by processor
The paper to be detected comprising subjective item is transferred from database according to paper unique number;
Obtain single topic similarity corresponding with subjective item described in each in the paper to be detected;
Single topic similarity of each subjective item is input to preset statistical model, receives the statistical model output
The overall similarity of the paper to be detected;
Obtain the cheating testing result of the paper to be detected according to the overall similarity, and by the overall similarity
With preset detection zone of the cheating testing result associated storage into the paper to be detected.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer-readable instruction, it is non-volatile that the computer-readable instruction can be stored in one
Property computer-readable storage medium in, the computer-readable instruction is when being executed, it may include as above-mentioned each method embodiment
Process.Wherein, to memory, storage, database or other media used in each embodiment provided by the present invention
Any reference may each comprise non-volatile and/or volatile memory.Nonvolatile memory may include read-only memory
(ROM), programming ROM (PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory.It is volatile
Property memory may include random access memory (RAM) or external cache.By way of illustration and not limitation, RAM
It is available in many forms, such as static state RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram
(DDRSDRAM), enhanced SDRAM (ESDRAM), synchronization link DRAM (SLDRAM), the direct RAM of memory bus (RDRAM),
Direct memory bus dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..
It is apparent to those skilled in the art that for convenience of description and succinctly, only with above-mentioned each function
Can unit or module division progress for example, in practical application, can according to need and by above-mentioned function distribution by difference
Functional unit or module complete, i.e., the internal structure of described device is divided into different functional unit or module, with complete
All or part of function described above.
Embodiment described above is merely illustrative of the technical solution of the present invention, rather than its limitations;Although referring to aforementioned reality
Applying example, invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each
Technical solution documented by embodiment is modified or equivalent replacement of some of the technical features;And these are modified
Or replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution should all
It is included within protection scope of the present invention.
Claims (10)
1. a kind of cheating determination method based on user's answer characterized by comprising
The paper to be detected comprising subjective item is transferred from database according to paper unique number;
Obtain single topic similarity corresponding with subjective item described in each in the paper to be detected;
Single topic similarity of each subjective item is input to preset statistical model, receives the described of the statistical model output
The overall similarity of paper to be detected;
Obtain the cheating testing result of the paper to be detected according to the overall similarity, and by the overall similarity and institute
State preset detection zone of the cheating testing result associated storage into the paper to be detected.
2. as described in claim 1 based on the cheating determination method of user's answer, which is characterized in that the acquisition is described to be checked
Single topic similarity corresponding with subjective item described in each in test volume, comprising:
The topic information and its user's answer of each subjective item in the paper to be detected are obtained, and will acquire described
User's answer of subjective item is recorded as answer to be detected;
According to the topic information of the subjective item, the user for obtaining identical subjective item in other papers from the database is answered
Case, and user's answer of identical subjective item is recorded as comparison answer in other the described papers that will acquire;
By preset matching rule detect the answer to be detected and it is described comparison answer between pure words similarity and/or
Semantic similarity, and it is similar according to the list topic that the pure words similarity and/or the semantic similarity obtain the subjective item
Degree.
3. as described in claim 1 based on the cheating determination method of user's answer, which is characterized in that described by each subjectivity
Single topic similarity of topic is input to preset statistical model, receives the entirety of the paper to be detected of the statistical model output
Similarity, comprising:
Single topic similarity of each subjective item is input to the first statistical model based on average algorithm, receives first system
Count the overall similarity of the paper to be detected of model output;First statistical model are as follows:
Wherein, Q is overall similarity;PiSimilarity is inscribed to be single;N is the number of monomer similarity;
Or
Single topic similarity of each subjective item is input to the second statistical model based on weight, receives the second statistics mould
The overall similarity of the paper to be detected of type output;Second statistical model are as follows:
Wherein, KiFor weight shared by single topic similarity;And
4. as described in claim 1 based on the cheating determination method of user's answer, which is characterized in that described according to the entirety
Similarity obtains before the cheating testing result of the paper to be detected, comprising:
Detect whether each single topic similarity is more than preset dynamic threshold;
It obtains single topic similarity and is more than single topic quantity of the dynamic threshold, and detect whether single topic quantity is more than pre-
If amount threshold;
When single topic quantity is more than preset quantity threshold value, the overall similarity is increased according to preset adjustment model;Institute
State adjustment model are as follows:
QS=Q ± Δ Q=Q ± | n-n0|*S0*F
Wherein, QSFor overall similarity adjusted;Δ Q is adjusted value;n0For preset quantity threshold value;N is single topic quantity;S0For
Dynamic threshold;F is regulation coefficient;
When single topic quantity is less than preset quantity threshold value, keep the overall similarity constant or according to preset tune
Integral mould reduces the overall similarity.
5. as described in claim 1 based on the cheating determination method of user's answer, which is characterized in that described unique according to paper
Number is transferred from database before the paper to be detected comprising subjective item, comprising:
The first log-in instruction comprising applicant's account is received, whether there is in data base querying and matched with applicant's account
Applicant's information;
When there is applicant's information with applicant's account, predetermined quantity and total score are randomly selected from the database
Subjective item generate answer paper;
It receives the submission comprising paper unique number to instruct, the answer paper answered is recorded as paper to be detected, and
By the paper unique number and the paper associated storage to be detected into database.
6. as described in claim 1 based on the cheating determination method of user's answer, which is characterized in that the cheating testing result
Comprising not there is no cheating and there is cheating;
The cheating testing result that the paper to be detected is obtained according to the overall similarity, and by the overall similarity
After preset detection zone of the cheating testing result associated storage into the paper to be detected, comprising:
The second log-in instruction comprising administrator's account is received, whether there is in data base querying and matched with administrator's account
Administrator information;
When there is administrator information matched with institute administrator's account, acquisition match with administrator's account and it is practised fraud and examines
Surveying result is the paper to be detected in the presence of cheating;
The paper score value of the paper to be detected is obtained according to preset grading parameters, and whether detects the paper score value
More than preset audit threshold value;
It, will the paper to be detected corresponding with the paper score value when the paper score value reaches preset audit threshold value
Client is shown in preset display form.
7. a kind of cheating decision maker based on user's answer characterized by comprising
Module is transferred, for transferring the paper to be detected comprising subjective item from database according to paper unique number;
Single topic detection module, for obtaining single topic similarity corresponding with subjective item described in each in the paper to be detected;
Single topic statistical module, for single topic similarity of each subjective item to be input to preset statistical model, described in reception
The overall similarity of the paper to be detected of statistical model output;
Cheating determination module, for obtaining the cheating testing result of the paper to be detected according to the overall similarity, and will
The preset detection zone of the overall similarity and the cheating testing result associated storage into the paper to be detected.
8. as claimed in claim 7 based on the cheating decision maker of user's answer, which is characterized in that single topic detection mould
Block, comprising:
First record sub module, for obtaining the topic information of each subjective item and its user in the paper to be detected
Answer, and the user's answer for the subjective item that will acquire is recorded as answer to be detected;
Second record sub module obtains in other papers from the database for the topic information according to the subjective item
User's answer of identical subjective item, and in other the described papers that will acquire user's answer of identical subjective item be recorded as comparison answer
Case;
Matched sub-block, it is pure between the answer to be detected and the comparison answer for being detected by preset matching rule
Text similarity and/or semantic similarity, and the master is obtained according to the pure words similarity and/or the semantic similarity
Single topic similarity of sight topic.
9. a kind of computer equipment, including memory, processor and storage are in the memory and can be in the processor
The computer-readable instruction of upper operation, which is characterized in that the processor is realized when executing the computer-readable instruction as weighed
Benefit requires the cheating determination method described in 1 to 6 any one based on user's answer.
10. a kind of computer readable storage medium, the computer-readable recording medium storage has computer-readable instruction, special
Sign is, realizes when the computer-readable instruction is executed by processor and is answered as described in any one of claim 1 to 6 based on user
The cheating determination method of case.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910203513.5A CN110083808B (en) | 2019-03-18 | 2019-03-18 | Cheating judgment method, device, equipment and storage medium based on user answers |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910203513.5A CN110083808B (en) | 2019-03-18 | 2019-03-18 | Cheating judgment method, device, equipment and storage medium based on user answers |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110083808A true CN110083808A (en) | 2019-08-02 |
CN110083808B CN110083808B (en) | 2024-04-02 |
Family
ID=67413172
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910203513.5A Active CN110083808B (en) | 2019-03-18 | 2019-03-18 | Cheating judgment method, device, equipment and storage medium based on user answers |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110083808B (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110674722A (en) * | 2019-09-19 | 2020-01-10 | 浙江蓝鸽科技有限公司 | Test paper splitting method and system |
CN112308417A (en) * | 2020-10-30 | 2021-02-02 | 领扣网络(上海)有限公司 | Anti-cheating method and system for online evaluation |
CN112529091A (en) * | 2020-12-18 | 2021-03-19 | 广州视源电子科技股份有限公司 | Courseware similarity detection method and device and storage medium |
CN113095968A (en) * | 2019-12-23 | 2021-07-09 | 深圳云天励飞技术有限公司 | Cheating identification method and device and electronic equipment |
CN113786625A (en) * | 2021-07-28 | 2021-12-14 | 北京永信至诚科技股份有限公司 | Online anti-cheating monitoring method and device |
CN114912510A (en) * | 2022-04-07 | 2022-08-16 | 央视市场研究股份有限公司 | On-line investigation sample quality evaluation system |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20060125234A (en) * | 2005-06-02 | 2006-12-06 | 헌 김 | Method and apparatus for creating exam sheets to prevent cheating on the exam |
KR20140042440A (en) * | 2012-09-28 | 2014-04-07 | 길석면 | Evaluation security method for online test. |
KR20140052268A (en) * | 2012-10-24 | 2014-05-07 | 에스케이텔레콤 주식회사 | Method for sentence evaluation using web search engine, apparatus thereof and computer recordable medium storing the method |
KR20140052685A (en) * | 2012-10-25 | 2014-05-07 | 에스케이텔레콤 주식회사 | Method for sentence evaluation using similar of sentence, apparatus thereof and computer recordable medium storing the method |
US20150193651A1 (en) * | 2014-01-03 | 2015-07-09 | Gleim Conferencing, Llc | System and method for validating test takers |
CN106170002A (en) * | 2016-09-08 | 2016-11-30 | 中国科学院信息工程研究所 | A kind of Chinese counterfeit domain name detection method and system |
CN108170469A (en) * | 2017-12-20 | 2018-06-15 | 南京邮电大学 | A kind of Git warehouses similarity detection method that history is submitted based on code |
CN108829780A (en) * | 2018-05-31 | 2018-11-16 | 北京万方数据股份有限公司 | Method for text detection, calculates equipment and computer readable storage medium at device |
CN108961122A (en) * | 2018-08-08 | 2018-12-07 | 安徽建筑大学 | A kind of examination anti-cheating method, relevant device and system |
CN108984493A (en) * | 2018-07-19 | 2018-12-11 | 中国联合网络通信集团有限公司 | A kind of Chinese articles duplicate checking method and system |
-
2019
- 2019-03-18 CN CN201910203513.5A patent/CN110083808B/en active Active
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20060125234A (en) * | 2005-06-02 | 2006-12-06 | 헌 김 | Method and apparatus for creating exam sheets to prevent cheating on the exam |
KR20140042440A (en) * | 2012-09-28 | 2014-04-07 | 길석면 | Evaluation security method for online test. |
KR20140052268A (en) * | 2012-10-24 | 2014-05-07 | 에스케이텔레콤 주식회사 | Method for sentence evaluation using web search engine, apparatus thereof and computer recordable medium storing the method |
KR20140052685A (en) * | 2012-10-25 | 2014-05-07 | 에스케이텔레콤 주식회사 | Method for sentence evaluation using similar of sentence, apparatus thereof and computer recordable medium storing the method |
US20150193651A1 (en) * | 2014-01-03 | 2015-07-09 | Gleim Conferencing, Llc | System and method for validating test takers |
CN106170002A (en) * | 2016-09-08 | 2016-11-30 | 中国科学院信息工程研究所 | A kind of Chinese counterfeit domain name detection method and system |
CN108170469A (en) * | 2017-12-20 | 2018-06-15 | 南京邮电大学 | A kind of Git warehouses similarity detection method that history is submitted based on code |
CN108829780A (en) * | 2018-05-31 | 2018-11-16 | 北京万方数据股份有限公司 | Method for text detection, calculates equipment and computer readable storage medium at device |
CN108984493A (en) * | 2018-07-19 | 2018-12-11 | 中国联合网络通信集团有限公司 | A kind of Chinese articles duplicate checking method and system |
CN108961122A (en) * | 2018-08-08 | 2018-12-07 | 安徽建筑大学 | A kind of examination anti-cheating method, relevant device and system |
Non-Patent Citations (2)
Title |
---|
RAZAN BAWARITH ET.AL: "E-exam Cheating Detection System", 《(IJACSA) INTERNATIONAL JOURNAL OF ADVANCED COMPUTER SCIENCE AND APPLICATIONS》, vol. 8, no. 4, pages 176 - 181 * |
常永虎 等: "基于考生在线行为的互抄袭作弊检测算法研究", 重庆工商大学学报(自然科学版), vol. 33, no. 03, pages 51 - 55 * |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110674722A (en) * | 2019-09-19 | 2020-01-10 | 浙江蓝鸽科技有限公司 | Test paper splitting method and system |
CN110674722B (en) * | 2019-09-19 | 2023-04-07 | 浙江蓝鸽科技有限公司 | Test paper splitting method and system |
CN113095968A (en) * | 2019-12-23 | 2021-07-09 | 深圳云天励飞技术有限公司 | Cheating identification method and device and electronic equipment |
CN112308417A (en) * | 2020-10-30 | 2021-02-02 | 领扣网络(上海)有限公司 | Anti-cheating method and system for online evaluation |
CN112308417B (en) * | 2020-10-30 | 2024-01-23 | 领扣网络(上海)有限公司 | Anti-cheating method and system for online evaluation |
CN112529091A (en) * | 2020-12-18 | 2021-03-19 | 广州视源电子科技股份有限公司 | Courseware similarity detection method and device and storage medium |
CN113786625A (en) * | 2021-07-28 | 2021-12-14 | 北京永信至诚科技股份有限公司 | Online anti-cheating monitoring method and device |
CN114912510A (en) * | 2022-04-07 | 2022-08-16 | 央视市场研究股份有限公司 | On-line investigation sample quality evaluation system |
CN114912510B (en) * | 2022-04-07 | 2023-08-15 | 央视市场研究股份有限公司 | Online investigation sample quality evaluation system |
Also Published As
Publication number | Publication date |
---|---|
CN110083808B (en) | 2024-04-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110083808A (en) | Cheating determination method, device, equipment and storage medium based on user's answer | |
Millot | International rankings: Universities vs. higher education systems | |
CN113545026B (en) | Systems and methods for vulnerability assessment and remedial action identification | |
US10121009B2 (en) | Computer system for discovery of vulnerabilities in applications including guided tester paths based on application coverage measures | |
Jahan | The parliament of Bangladesh: representation and accountability | |
Ruggeri et al. | Events data as Bismarck's sausages? Intercoder reliability, coders' selection, and data quality | |
CN110313009A (en) | The trust model and risk tolerance of learning object carry out calculation risk score | |
Aiken et al. | Program targeting with machine learning and mobile phone data: Evidence from an anti-poverty intervention in Afghanistan | |
Davenport et al. | Interpreting the public interest: A survey of professional accountants | |
Holt et al. | An examination of English and Welsh constables’ perceptions of the seriousness and frequency of online incidents | |
Kouatli | People-process-performance benchmarking technique in cloud computing environment: An AHP approach | |
Joo et al. | The influence of information security on the adoption of web-based integrated information systems: an e-government study in Peru | |
CN112702410B (en) | Evaluation system, method and related equipment based on blockchain network | |
LeBlanc et al. | Risk perception of internet-related activities | |
CN107944841A (en) | Emolument computing system and method | |
McVay et al. | Preparers and the financial reporting system | |
Ernita et al. | Strategy to Improve Employee Security Awareness at Information Technology Directorate Bank XYZ | |
Matsieli et al. | Evaluation of E-Government Websites in Lesotho: An Empirical Study | |
Datta et al. | Aspects of inclusion and peoples’ empowerment related to national rural employment Guarantee Scheme in India | |
CN108053510A (en) | Intelligent attendance system and method | |
García-Hernández et al. | Effectiveness of new safeguards and prohibitions to protect audit independence: an empirical research with auditors | |
Soewito et al. | Risk Assessment on Cloud Computing for The Learning System in The Education Environment | |
US20240129284A1 (en) | Method and system for influencing user interactions | |
JP6976477B1 (en) | Information processing equipment and information processing method | |
US20230394609A1 (en) | Machine learning service based on skills graph |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |