CN106021288A - Method for rapid and automatic classification of classroom testing answers based on natural language analysis - Google Patents
Method for rapid and automatic classification of classroom testing answers based on natural language analysis Download PDFInfo
- Publication number
- CN106021288A CN106021288A CN201610283931.6A CN201610283931A CN106021288A CN 106021288 A CN106021288 A CN 106021288A CN 201610283931 A CN201610283931 A CN 201610283931A CN 106021288 A CN106021288 A CN 106021288A
- Authority
- CN
- China
- Prior art keywords
- answer
- word
- natural language
- keyword
- language analysis
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Electrically Operated Instructional Devices (AREA)
Abstract
The invention provides a method for rapid and automatic mobile terminal classification of classroom testing answers based on natural language analysis. The method comprises the steps including (1) collection of classroom testing answer texts, (2) pre-processing of the classroom testing answer texts, (3) word division of the answer texts, (4) filtering of stop words, (5) synonym replacement, (6) keyword extraction, (7) establishment of a keyword set and (8) text classification. The method provided by the invention solves the problem in the rapid and automatic classification for a few of short texts in the classroom testing answer results of students during teacher teaching. With the method, a teacher can obtain a summarization report of question answer situations in time; former collection of answers written on paper and subsequent manual summarization are avoided; and teacher-student interaction efficiency in class is increased.
Description
Technical field
The invention belongs to field of computer technology, relate to natural language text analytical technology, for a kind of based on natural language analysis
Mobile terminal is with the method for the fast automatic classification of hall test answers.During solving teachers ' teaching, test learner answering questions result with hall
The problem of the fast automatic classification of a small amount of short text.Make teacher can obtain the summary report of question answering situation in time, remove from
Need to collect the work that papery answer the most manually collects in the past, improve the efficiency of classroom Ability of Normal School Students interactive event.
Background technology
Mobile terminal with hall test refer to that teacher is applied by mobile phone, by the Internet by student organization in a virtual class.
Teacher issues to students in this class and tests with hall, and students participates in and answers the problem in testing with hall, and teacher just may be used
To obtain the result with hall test in time.True-False, individual event/multiple-choice question, these 3 kinds of shapes of question-and-answer problem are had with the problem of hall test
Formula.Wherein True-False, individual event/multiple-choice question, using one or more numerical value as the question-and-answer problem of answer, computer can be very
Easily answer is classified.But for the question-and-answer problem using a bit of text as answer, it is automatically classified and is still a difficult problem,
This is also the problem that the present invention solves.
Chinese natural language process refer to that Chinese text is converted to computer it will be appreciated that form, and it is processed, and
Result being stored in a computer, be computer science, artificial intelligence, computer and Human Natural Language are paid close attention in linguistics
Between the field of interaction.Technology based on machine learning, by the text of natural language by a series of participles, key word
The technology such as extraction understand.The present invention uses Chinese analysis, keyword extraction, the answer in testing with hall is carried out automatically
Classification, facilitates teacher to obtain the feedback with hall test result in time.
Summary of the invention
The problem to be solved in the present invention is: propose a kind of based on natural language analysis will with hall test learner answering questions result the shortest
The method of the fast automatic classification of text so that teacher can obtain the summary report of question answering situation in time, eliminates and needed in the past
The work that papery answer to be collected the most manually collects.
The technical scheme is that based on natural language analysis with the hall fast automatic sorting technique of test answers, use nature
Linguistic analysis, the question-and-answer problem answer submitted students on classroom to is classified automatically.Concretely comprise the following steps:
1) collect and pretreatment is answered
Use the mode of mobile terminal application, the problem issuing classroom test to student, and collect the answer of students.Filter also
Remove and there is the text that form is abnormal.The situation of form exception has: the loss of learning in answer;The essential information of answered exercise question
Exercise question gone out with teacher does not mates;Answer topic exceeds the time that teacher specifies object time;Answer in the text of exercise question and comprise injection
Attack code.The type of automatic clustering is needed according to content of text automatic decision.All answers are to the feelings of Chinese answer
Condition, uses the mode of Chinese natural language analysis to classify;When all answers all comprise numerical value, use numerical value
Join and classify;When all answers all comprise and only comprise English word, use the mode of English natural linguistic analysis
Classify.
2) natural language analysis
After obtaining the pre-processed results answered, if the classification of values match form, only the numerical value in answer need to be extracted
Out, being mutually matched and classify, this type is relatively simple.If Chinese form or the answer of English form,
Need to use natural language analysis it is processed and classifies.Basically identical, only for English and Chinese automatic classification method
Only there are differences in the selection of segmenting method, the selection of dictionary.Five steps it are divided into again in natural language analysis: participle,
Stop-word filtration, synonym replacement, keyword extraction, structure keyword set.
Participle: the part of the natural language description in learner answering questions is carried out participle operation, every part of answer is divided into independent word
Language.For the answer of Chinese form, use Chinese natural language to process engine, Chinese text is divided into several phrases, and
It is labelled with part of speech for these phrases;For the answer of English form, use English natural language processing engine, English text is drawn
It is divided into several phrases, and is labelled with part of speech for these phrases.
Stop-word filters: for the result of participle, chooses the dictionary of Chinese and English stop-word, removes stop-word therein.Stop-word
But being that in language, the frequency of occurrences is the highest does not has influential word for the meaning of one's words, these words are extremely universal, but compared with other words,
These words do not have any physical meaning.
Synonym is replaced: generating a synon dictionary, the different terms that will refer to same object is classified as a set, uses
A word in this set is as representing word, and replaces other words in this set contained in learner answering questions with representing word.
Use synonym can reduce the ambiguousness in classification results after replacing so that final result is the most accurate.
Keyword extraction: by natural language analysis, by the keyword extraction in every part of learner answering questions out, is dropped by weight size
Sequence arranges, and the value of weight represents that this key word can represent the degree of statement implication, k the key word that weighting weight is maximum.
Weight refers to: assesses a word for the significance level of a copy of it text in whole answer, quantifies this significance level,
Weight as this word.Conventional technology is TF-IDF, and its main thought is if certain word or phrase are at an article
Frequency TF of middle appearance is high, and seldom occurs in other articles, then it is assumed that this word or phrase have good class discrimination
Ability, is adapted to classification.
TFIDF refers to TF with IDF and is multiplied obtained result.Wherein TF is word frequency (Term Frequency);IDF generation
The reverse document-frequency of table (Inverse Document Frequency), main thought is: if the document comprising entry is the fewest,
IDF is the biggest.
Formula 1:
Formula 1 Middle molecule is this word occurrence number hereof, and denominator is then the occurrence number sum of the most all words.
Formula 2:
In formula 2, the molecule in logarithm is general act number, and denominator is then the number of the file of this word.
Formula 3:tfidfI, j=tfI, j×idfi
In formula 3, TFIDF is multiplied gained by TF with IDF.
Structure keyword set: the number of times occurred in all error reportings according to key word, selects occurrence number and exceedes setting threshold
The key word of value a, by occurrence number descending, and chooses m the key word that occurrence number is most, constitutes keyword set.
Key word in set and stem are contrasted, the word removal setting threshold value b and occurring in stem will be exceeded.
3) classification is answered:
Selecting n the keyword that occurrence number is most, if comprising one or more keyword in single answer, weight selection is
The representative keyword that high keyword is answered as this.Being classified as a class by representing the identical answer of keyword, n keyword will
Answer is divided into n class, and the answer that n keyword does not comprises individually is classified as a class.
Accompanying drawing explanation
Fig. 1 is a kind of based on natural language analysis the flow process with the hall fast automatic sorting technique of test answers of the embodiment of the present invention
Figure
Detailed description of the invention
In order to know more about the technology contents of the present invention, especially exemplified by specific embodiment and coordinate institute's accompanying drawings to be described as follows.
Fig. 1 is a kind of based on natural language analysis the flow process with the hall fast automatic sorting technique of test answers of the embodiment of the present invention
Figure.
A kind of based on natural language analysis with the hall fast automatic sorting technique of test answers, it is characterised in that to comprise the following steps:
In flow process the 1. step is to collect from Mobile solution or wechat public number to answer.What the present invention solved is that teachers and students are mutual in classroom
How disorder of internal organs, use mobile terminal to be efficiently completed the problem with hall test.So teacher needs to use the mode of mobile terminal application,
The problem issuing classroom test to student, the answer of systematic collection students, use for subsequent step.
In flow process the 2. step is pretreatment.The reference format of answer collected comprises 3 main informations: the exercise question of textual form
Answer;The student's essential information participated in;The essential information (comprising stem, class, teacher, time) of the exercise question answered.
If form is undesirable, then abandoned.The undesirable situation of form has: the loss of learning in answer;Answered
The essential information exercise question gone out with teacher of exercise question does not mates;Answer topic exceeds the time that teacher specifies object time;Answer exercise question
Text comprises the attack code of injection.In the case of form is normal, analyzes the text answered, be divided three classes.Use canonical
Answer is classified by expression formula.When all answers all comprise numerical value, it is classified as numeric form;Equal for all answers
Comprise and only comprise the situation of English word, be classified as English form;When all answers all comprise Chinese answer, it is classified as
Chinese form.
In flow process the 3. step is that Chinese word segmentation/English string segmentation/numerical value extracts.According to the result of pretreatment, for numeric form
Answer, extract concrete numerical value therein, using concrete numerical value as keyword, by mating concrete numerical value, it is classified;Right
In the answer of Chinese form, use Chinese natural language to process engine, Chinese text is divided into several phrases, and is these
Phrase is labelled with part of speech.Ansj_seg is have employed as natural language processing engine in being embodied as flow process.This be one based on
The java of the Chinese word segmentation of *** semantic model+conditional random field models realizes, and efficiency maintains the leading position in similar tools,
Participle speed reaches (to test under mac air) about word each second about 2,000,000, and accuracy rate can reach more than 96%.ansj_seg
Being an open source projects, just have submitted up-to-date version author in January, 2016, the last renewal of distance has been separated by two
Year, in this version, ansj_seg can carry out participle to English.So, language natural for the Chinese in pretreatment
Speech processes and English natural Language Processing can use ansj_seg as participle instrument.
In flow process the 4. step is off word and filters.But stop-word be in language the frequency of occurrences the highest the meaning of one's words is not had influential
Word, these words are extremely universal, but compared with other words, what physical meaning these words do not have.Can run into a lot inside English
A, the, or etc. use the word or word that frequency is a lot, are often article, preposition, adverbial word or conjunction etc..It is similar to, in Chinese
" ", " the inside ", " also ", " ", " it ", " being " these words be all off word.By these stop-words from
Word segmentation result filters off.
In flow process the 5. step is that synonym is replaced.Being that different classmates submits to owing to collecting the answer come, they are for same
Individual things may use different phrases to go to describe, or uses the word of synonym to go to answer same problem.Such as " Tang is too
Ancestor " and " Li Shih-min " referred to same personage, the two word just should be considered synonym, in natural language processing
In should be by as same word processing.Generating a synon dictionary in being embodied as flow process, it is same right to will refer to
The different terms of elephant is classified as a set, and a word in gathering with this is as representing word, and returns with representing word replacement student
Answer containing this set in other words.
In flow process the 6. step is to extract key word.Use is the keyword extraction instrument in ansj_seg.Key in ansj_seg
Word extracting method is based on TF-IDF method.TF-IDF (term frequency-inverse document frequency) is
A kind of conventional weighting technique for information retrieval Yu data mining.Specifically, TF-IDF method in order to assess a word for
The significance level of a copy of it file in one file set or a corpus.One word weight for the text at its place
The property wanted, the number of times occurred in the text at its place along with it is directly proportional rising, the frequency simultaneously occurred in other texts along with it
Rate is inversely proportional to decline.Taking k the key word that wherein weight is maximum, k value is the biggest, the select key representativeness to statement
The best, but accordingly, time cost is the biggest.For the text of student's answer, owing to being most hundreds of of dozens of
Short text, number of words is generally tens the most hundreds of, and for this situation, the efficiency of TF-IDF is the highest.If the property of server
Can not be good, it is possible to use the frequency of occurrences of statistics same words to replace its weight.
In flow process the 7. step is to build keyword set.The number of times occurred in all answers according to key word, selects out occurrence
Number exceedes the key word setting threshold value a, by occurrence number descending, and chooses m the key word that occurrence number is most, structure
Become keyword set.Key word in set and stem are contrasted, the word setting threshold value b and occurring in stem will be exceeded
Remove.
In flow process the 8. step is text classification.Select n the keyword that occurrence number is most, if single answer comprises one
Individual or multiple keywords, the representative keyword that the keyword that weight selection is the highest is answered as this.Identical by representing keyword
Answer is classified as a class, and answer is divided into n class by n keyword, and the answer that n keyword does not comprises individually is classified as a class.Will knot
Fruit is stored in data base, checks for relevant teacher.
In sum, the present invention divides automatically based on natural language analysis, the question-and-answer problem answer submitted students on classroom to
Class, during solving teachers ' teaching, with the problem of the fast automatic classification of a small amount of short text of hall test learner answering questions result.Make
Obtain teacher and can obtain the summary report of question answering situation in time, eliminate and need to collect what papery answer the most manually collected in the past
Work, improves the efficiency that classroom Ability of Normal School Students is interactive.
Claims (7)
1. based on natural language analysis with the hall fast automatic sorting technique of test answers, it is characterized in that using natural language
Analyzing, the question-and-answer problem answer submitted students on classroom to carries out participle, extracts effective keyword therein, to submitted to
Text answers is classified automatically so that teacher can obtain the summary report of question answering situation in time, eliminates needs in the past
Collect the work that papery answer the most manually collects, the problem solving the fast automatic classification of a small amount of short text.
Natural language analysis the most according to claim 1 with the fast automatic sorting technique of hall test answers, it specifically walks
Suddenly it is:
1) collect and pretreatment is answered
Collect the question answering that student submits on classroom, receive the question answering that user sends, and according to the error format set
Standard, filters wherein there is abnormal text, prevents from causing mistake in subsequent step.Automatically sentence according to content of text
The disconnected type needing automatic clustering.It is specifically divided into Chinese form, English form, numeric form;
2) natural language analysis
To 1) in result carry out natural language analysis.The classification answered for numeric form, extracts the numerical value in answer
Come, be mutually matched and classify;The classification answered for Chinese and English form, needs to use natural language analysis at it
Manage and classify.English and Chinese automatic classification method be there are differences in the selection of segmenting method, the selection of dictionary.Bag
Include following sub-step:
2a) participle:
The part of the natural language description in learner answering questions is carried out participle operation, every part of answer is divided into independent word.Make
Use natural language processing engine, text is divided into several phrases, and is labelled with part of speech for these phrases;
2b) stop-word filters:
Definition 1: but stop-word is that in language, the frequency of occurrences is the highest does not has influential word for the meaning of one's words, and these words are extremely universal,
But compared with other words, these words do not have any physical meaning;
For 2a) result, remove stop-word therein;
2c) synonym is replaced:
Generating a TongYiCi CiLin, the different terms that will refer to same object is classified as a set, in gathering with this
Individual word is as representing word, and replaces other words in this set contained in learner answering questions with representing word, reduces final result
Ambiguousness;
2d) keyword extraction:
By natural language analysis, by the keyword extraction in every part of learner answering questions out, by weight size descending, weight
Value represent that this key word can represent the degree of statement implication, k the key word that weighting weight is maximum;
2e) build keyword set:
The number of times occurred in all error reportings according to key word, selects occurrence number and exceedes the key word setting threshold value a, press
Occurrence number descending, and choose m the key word that occurrence number is most, constitute keyword set.Key word in gathering
Contrast with stem, the word removal setting threshold value b and occurring in stem will be exceeded;
3) classification is answered:
Selecting n the keyword that occurrence number is most, if comprising one or more keyword in single answer, weight selection is
The representative keyword that high keyword is answered as this.Being classified as a class by representing the identical answer of keyword, n keyword will
Answer is divided into n class, and the answer that n keyword does not comprises individually is classified as a class.
The concrete steps with the fast automatic sorting technique of hall test answers of natural language analysis the most according to claim 2,
It is characterized in that, step 1) in, use mobile terminal application or wechat public's account to accept the question answering that user sends, and
It is stored in data base.
The concrete steps with the fast automatic sorting technique of hall test answers of natural language analysis the most according to claim 2,
It is characterized in that, step 1) in, use regular expression that answer is classified.All answers are all comprised to the feelings of numerical value
Condition, is classified as numeric form;When all answers all comprise and only comprise English word, it is classified as English form;For institute
Have and answer the situation all comprising Chinese answer, be classified as Chinese form.
The concrete steps with the fast automatic sorting technique of hall test answers of natural language analysis the most according to claim 2,
It is characterized in that, step 2a) in, for the answer of Chinese form, use Chinese natural language to process engine, by Chinese literary composition
Originally it is divided into several phrases, and is labelled with part of speech for these phrases;For the answer of English form, use English natural language
Process engine, English text is divided into several phrases, and is labelled with part of speech for these phrases.
The concrete steps with the fast automatic sorting technique of hall test answers of natural language analysis the most according to claim 2,
It is characterized in that, step 2d) in, assess the words significance level for a copy of it text in whole answer, quantify
This significance level, as the weight of this words.The technology being most frequently with is TF-IDF, its main thought be if certain
Frequency TF that individual word or phrase occur in an article is high, and seldom occurs in other articles, then it is assumed that this word or short
Language has good class discrimination ability, is adapted to classification;
TFIDF refers to TF*IDF;TF is word frequency (Term Frequency);IDF represents reverse document-frequency (Inverse
Document Frequency), IDF main thought is: if the document comprising entry is the fewest, IDF is the biggest;
Formula 1:
Formula 1 Middle molecule is this word occurrence number hereof, and denominator is then the occurrence number sum of the most all words;
Formula 2:
In formula 2, the molecule in logarithm is general act number, and denominator is then the number of the file of this word;
Formula 3:tfidfI, j=tfI, j×idfi
In formula 3, TFIDF is just multiplied gained by TF with IDF.
The concrete steps with the fast automatic sorting technique of hall test answers of natural language analysis the most according to claim 2,
It is characterized in that, step 2d) in, the k value chosen is the biggest, and select key is the best to the representativeness of statement, but phase
Answering, time cost is the biggest.It is smaller that the scene that the present invention is directed to requires time for cost, it is considered to actual scene, and k value chooses 2-5
The most suitable.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610283931.6A CN106021288A (en) | 2016-04-27 | 2016-04-27 | Method for rapid and automatic classification of classroom testing answers based on natural language analysis |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610283931.6A CN106021288A (en) | 2016-04-27 | 2016-04-27 | Method for rapid and automatic classification of classroom testing answers based on natural language analysis |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106021288A true CN106021288A (en) | 2016-10-12 |
Family
ID=57081437
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610283931.6A Pending CN106021288A (en) | 2016-04-27 | 2016-04-27 | Method for rapid and automatic classification of classroom testing answers based on natural language analysis |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106021288A (en) |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106599054A (en) * | 2016-11-16 | 2017-04-26 | 福建天泉教育科技有限公司 | Method and system for title classification and push |
CN106846956A (en) * | 2017-03-27 | 2017-06-13 | 深圳信息职业技术学院 | Teaching control system and method |
CN106940788A (en) * | 2017-03-07 | 2017-07-11 | 百度在线网络技术(北京)有限公司 | Intelligent scoring method and device, computer equipment and computer-readable medium |
CN107590124A (en) * | 2017-09-06 | 2018-01-16 | 陈飞 | The method replaced to synonym by scene and compared according to the standard phrase sorted out by scene |
CN108363694A (en) * | 2018-02-23 | 2018-08-03 | 北京窝头网络科技有限公司 | Keyword extracting method and device |
CN108614860A (en) * | 2018-03-27 | 2018-10-02 | 成都律云科技有限公司 | A kind of lawyer's information processing method and system |
CN108734153A (en) * | 2018-07-18 | 2018-11-02 | 深圳迪普乐宁科技有限公司 | A kind of method and system of efficient computer marking |
CN108777141A (en) * | 2018-05-31 | 2018-11-09 | 康键信息技术(深圳)有限公司 | Test device, the method for test and storage medium |
CN108829663A (en) * | 2018-05-21 | 2018-11-16 | 宁波薄言信息技术有限公司 | A kind of article appraisal procedure and system |
CN109299865A (en) * | 2018-09-06 | 2019-02-01 | 西南大学 | Psychological assessment system and method, information data processing terminal based on semantic analysis |
CN110196893A (en) * | 2019-05-05 | 2019-09-03 | 平安科技(深圳)有限公司 | Non- subjective item method to go over files, device and storage medium based on text similarity |
CN112101005A (en) * | 2020-04-02 | 2020-12-18 | 上海迷因网络科技有限公司 | Method for generating and dynamically adjusting quick expressive force test questions |
CN112232072A (en) * | 2020-10-27 | 2021-01-15 | 上海安畅网络科技股份有限公司 | Quality analysis method, quality analysis device, electronic apparatus, and computer-readable storage medium |
CN112906895A (en) * | 2021-02-09 | 2021-06-04 | 柳州智视科技有限公司 | Method for imitating question object |
CN113920794A (en) * | 2021-09-09 | 2022-01-11 | 江西台德智慧科技有限公司 | Network interactive education method and intelligent system |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030095544A1 (en) * | 2001-11-20 | 2003-05-22 | Chu Keith T. | Method and apparatus for low-latency, semantic conveyance of an answer tone over a data network |
CN1790332A (en) * | 2005-12-28 | 2006-06-21 | 刘文印 | Display method and system for reading and browsing problem answers |
CN103578311A (en) * | 2012-08-10 | 2014-02-12 | 俞晓鸿 | System and method aimed at realizing preconcept capture and analysis in effective teaching |
CN104536881A (en) * | 2014-11-28 | 2015-04-22 | 南京慕测信息科技有限公司 | Public testing error report priority sorting method based on natural language analysis |
-
2016
- 2016-04-27 CN CN201610283931.6A patent/CN106021288A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030095544A1 (en) * | 2001-11-20 | 2003-05-22 | Chu Keith T. | Method and apparatus for low-latency, semantic conveyance of an answer tone over a data network |
CN1790332A (en) * | 2005-12-28 | 2006-06-21 | 刘文印 | Display method and system for reading and browsing problem answers |
CN103578311A (en) * | 2012-08-10 | 2014-02-12 | 俞晓鸿 | System and method aimed at realizing preconcept capture and analysis in effective teaching |
CN104536881A (en) * | 2014-11-28 | 2015-04-22 | 南京慕测信息科技有限公司 | Public testing error report priority sorting method based on natural language analysis |
Cited By (25)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106599054B (en) * | 2016-11-16 | 2019-12-24 | 福建天泉教育科技有限公司 | Method and system for classifying and pushing questions |
CN106599054A (en) * | 2016-11-16 | 2017-04-26 | 福建天泉教育科技有限公司 | Method and system for title classification and push |
CN106940788A (en) * | 2017-03-07 | 2017-07-11 | 百度在线网络技术(北京)有限公司 | Intelligent scoring method and device, computer equipment and computer-readable medium |
WO2018161917A1 (en) * | 2017-03-07 | 2018-09-13 | 百度在线网络技术(北京)有限公司 | Intelligent scoring method and apparatus, computer device, and computer-readable medium |
CN106940788B (en) * | 2017-03-07 | 2020-05-29 | 百度在线网络技术(北京)有限公司 | Intelligent scoring method and device, computer equipment and computer readable medium |
CN106846956A (en) * | 2017-03-27 | 2017-06-13 | 深圳信息职业技术学院 | Teaching control system and method |
CN107590124A (en) * | 2017-09-06 | 2018-01-16 | 陈飞 | The method replaced to synonym by scene and compared according to the standard phrase sorted out by scene |
CN107590124B (en) * | 2017-09-06 | 2020-12-04 | 耀灵人工智能(浙江)有限公司 | Method for replacing synonyms according to scenes and comparing standard phrases classified according to scenes |
CN108363694B (en) * | 2018-02-23 | 2021-08-24 | 北京窝头网络科技有限公司 | Keyword extraction method and device |
CN108363694A (en) * | 2018-02-23 | 2018-08-03 | 北京窝头网络科技有限公司 | Keyword extracting method and device |
CN108614860A (en) * | 2018-03-27 | 2018-10-02 | 成都律云科技有限公司 | A kind of lawyer's information processing method and system |
CN108829663B (en) * | 2018-05-21 | 2022-04-29 | 宁波薄言信息技术有限公司 | Article evaluation method and system |
CN108829663A (en) * | 2018-05-21 | 2018-11-16 | 宁波薄言信息技术有限公司 | A kind of article appraisal procedure and system |
CN108777141B (en) * | 2018-05-31 | 2022-01-25 | 康键信息技术(深圳)有限公司 | Test apparatus, test method, and storage medium |
CN108777141A (en) * | 2018-05-31 | 2018-11-09 | 康键信息技术(深圳)有限公司 | Test device, the method for test and storage medium |
CN108734153A (en) * | 2018-07-18 | 2018-11-02 | 深圳迪普乐宁科技有限公司 | A kind of method and system of efficient computer marking |
CN109299865B (en) * | 2018-09-06 | 2021-12-17 | 西南大学 | Psychological evaluation system and method based on semantic analysis and information data processing terminal |
CN109299865A (en) * | 2018-09-06 | 2019-02-01 | 西南大学 | Psychological assessment system and method, information data processing terminal based on semantic analysis |
CN110196893A (en) * | 2019-05-05 | 2019-09-03 | 平安科技(深圳)有限公司 | Non- subjective item method to go over files, device and storage medium based on text similarity |
CN112101005A (en) * | 2020-04-02 | 2020-12-18 | 上海迷因网络科技有限公司 | Method for generating and dynamically adjusting quick expressive force test questions |
CN112101005B (en) * | 2020-04-02 | 2022-08-30 | 上海迷因网络科技有限公司 | Method for generating and dynamically adjusting quick expressive force test questions |
CN112232072A (en) * | 2020-10-27 | 2021-01-15 | 上海安畅网络科技股份有限公司 | Quality analysis method, quality analysis device, electronic apparatus, and computer-readable storage medium |
CN112906895A (en) * | 2021-02-09 | 2021-06-04 | 柳州智视科技有限公司 | Method for imitating question object |
CN112906895B (en) * | 2021-02-09 | 2022-12-06 | 柳州智视科技有限公司 | Method for imitating question object |
CN113920794A (en) * | 2021-09-09 | 2022-01-11 | 江西台德智慧科技有限公司 | Network interactive education method and intelligent system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106021288A (en) | Method for rapid and automatic classification of classroom testing answers based on natural language analysis | |
Pal et al. | An approach to automatic text summarization using WordNet | |
CN109376251A (en) | A kind of microblogging Chinese sentiment dictionary construction method based on term vector learning model | |
CN101520802A (en) | Question-answer pair quality evaluation method and system | |
CN106202584A (en) | A kind of microblog emotional based on standard dictionary and semantic rule analyzes method | |
CN106446147A (en) | Emotion analysis method based on structuring features | |
Pal et al. | An approach to automatic text summarization using simplified lesk algorithm and wordnet | |
Banik et al. | Survey on text-based sentiment analysis of bengali language | |
CN113934814B (en) | Automatic scoring method for subjective questions of ancient poems | |
CN116362591A (en) | Multidimensional teacher evaluation auxiliary method and system based on emotion analysis | |
Ali et al. | Sentiment analysis using unlabeled email data | |
CN111522948A (en) | Method and system for intelligently processing official document | |
Flor et al. | Text mining and automated scoring | |
CN112711666B (en) | Futures label extraction method and device | |
CN109241276B (en) | Word classification method in text, and speech creativity evaluation method and system | |
Ma et al. | Multi-resolution annotations for emoji prediction | |
Zhu et al. | YUN111@ Dravidian-CodeMix-FIRE2020: Sentiment Analysis of Dravidian Code Mixed Text. | |
Atapattu et al. | Evaluation of concept importance in concept maps mined from lecture notes | |
Bessou et al. | Efficient measuring of readability to improve documents accessibility for arabic language learners | |
Krilavičius et al. | News media analysis using focused crawl and natural language processing: case of Lithuanian news websites | |
Shankar et al. | Novice Retroaction Report | |
Daba | Improving Afaan Oromo question answering system: definition, list and description question types for non-factoid questions | |
Chung et al. | An annotated news corpus of Malaysian Malay | |
Sathiyamurthy et al. | Automatic question generation from documents for e-learning | |
Talwar et al. | Intelligent Classroom System for Qualitative Analysis of Students' Conceptual Understanding |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
DD01 | Delivery of document by public notice |
Addressee: NANJING MUCE INFORMATION TECHNOLOGY CO., LTD. Document name: Notification of Passing Preliminary Examination of the Application for Invention |
|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20161012 |