CN109977214A - A kind of online knotty problem answer recommendation interactive approach and system of education of middle and primary schools - Google Patents
A kind of online knotty problem answer recommendation interactive approach and system of education of middle and primary schools Download PDFInfo
- Publication number
- CN109977214A CN109977214A CN201910248955.1A CN201910248955A CN109977214A CN 109977214 A CN109977214 A CN 109977214A CN 201910248955 A CN201910248955 A CN 201910248955A CN 109977214 A CN109977214 A CN 109977214A
- Authority
- CN
- China
- Prior art keywords
- answer
- knotty
- answers
- keyword
- online
- 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
Abstract
The invention belongs to the online knotty problems of education of middle and primary schools to answer technical field, more particularly to a kind of online knotty problem answer of education of middle and primary schools recommends interactive approach, simultaneously, it is answered the present invention also provides a kind of online knotty problem of education of middle and primary schools and recommends interaction systems, this method includes inputting difficult problem, identify problem category, keyword extraction, problem answers extract, relatedness computation, the present invention is solved the problems, such as that the prior art there are problems that lacking and is answered user's proposition using natural language recognition technology using Internet resources as the knowledge base of system, have to have filled up answers the problem of user proposes this blank using natural language recognition technology on the internet, convenient for the answer of knotty problem, improve matching efficiency, the recall rate of raising system and the advantageous effects of accuracy.
Description
Technical field
The invention belongs to the online knotty problem of education of middle and primary schools answer technical field more particularly to a kind of educations of middle and primary schools
Online knotty problem answer recommend interactive approach, meanwhile, the present invention also provides a kind of online knotty problems of education of middle and primary schools
Answer recommends interaction systems.
Background technique
During education of middle and primary schools, online knotty problem answer especially calculates some mathematical problems or understands
Relevant problem concerning study although can easily convert various linear modules by Google search engine, and carries out some letters
Single mathematical computations, but for the answer of other problems, intelligence seeks answer by the webpage of search result, and existing system needs
This problem is solved, and be that search engine is added to can be automatically request-answering system that student furnishes an answer with regard to many problems,
Question answering system is a kind of new intelligence system being born in the case where all need urgent need, it uses natural language processing technique,
On the one hand the analysis processing for completing user's query, on the other hand completes the generation of accurate answer, allows people in rambling net
The useful answer that really oneself is desired is quickly and accurately obtained in the network world, the prior art exists to lack is with Internet resources
The knowledge base of system answers the problem of the problem of user proposes using natural language recognition technology.
Summary of the invention
The online knotty problem answer that the present invention provides a kind of education of middle and primary schools recommends interactive approach and system, on solving
It states and proposes the prior art in background technique and exist that lack using Internet resources be the knowledge base of system using natural language recognition skill
Art is come the problem of answering the problem of user proposes.
Technical problem solved by the invention is realized using following technical scheme: a kind of online difficulty of education of middle and primary schools
Interactive approach is recommended in answer, comprising:
Inputting difficult problem；
Identification problem category: the interrogative of knotty problem is identified into the heat studied in classification by the method for template matching
Point；
Keyword extraction: the clause of knotty problem is divided into the keyword of several determinations by Weight algorithm；
Problem answers extract: the semantic frame of the corresponding candidate documents of knotty problem is extracted by comparing by corresponding position
Paragraph as problem answers；
Relatedness computation: several candidate answers of problem answers are subjected to relatedness computation and are judged to the degree of correlation and related
Degree sequence.
Further, the inputting difficult problem include the interrogative of knotty problem what includes, who, how, how, how
When, be how many and other related interrogatives.
Further, if the identification problem category further includes that a no less than template matches with knotty problem, basis
Greedy algorithm is matched according to maximum match principle.
Further, the matching template in the method for the template matching includes:
| Type | The interrogative of knotty problem | With reference to sentence |
| Ask people | Who, who | Whom the author of this article is |
| Ask place | Where | Where is the provincial capital in Hunan Province |
| Ask title | What is | What the abbreviation in Hunan is |
| Ask time and date | It when is | When your birthday is |
| Ask money | How much | This computer how much |
| Ask number | How much is long how many, width | How much is this tree height |
| Ask other | How, what if | How to improve the health of the people |
Further, it includes removing knotty problem other than interrogative that the keyword extraction, which includes the determination principle of keyword,
Using noun or verb or adjective and the word of other parts of speech as keyword.
Further, the Weight algorithm of the keyword extraction includes:
It is described to contain sexual keyword and be determined as strong restricted key if keyword is that must contain sexual keyword
Word；
If keyword be it is nonessential contain sexual keyword, it is described to contain sexual keyword and be determined as weak restricted key
Word.
Further, described problem answer extracting includes that the semantic frame by the corresponding candidate documents of knotty problem passes through
Comparing the paragraph for extracting corresponding position as problem answers includes all sentences for traversing the corresponding candidate documents of knotty problem, will
All sentences of traversal obtain semantic frame by semantic analysis, by the semantic frame of the semantic frame of candidate documents and knotty problem
Frame is compared, if successful match, the word and sentence for extracting the corresponding position of candidate documents are as problem answers.
Further, the specific algorithm of described problem answer extracting includes:
A, knotty problem keyword extraction is determined treated problem category；
B, all paragraphs of all candidate documents are scanned and find out the candidate answers to match with knotty problem；
C, each candidate answers are distributed to an initial score；
D, positive or negative points operation is carried out according to Weight algorithm；
E, the possibility answer of weight limit score is selected；
F, will likely answer score be set as zero；
G, repeat selection weight limit score possibility answer and will likely answer score be set as zero, up to select it is several
A suitable candidate answers are as problem answers.
Further, the relatedness computation include by several candidate answers of problem answers by semantic knowledge resources bank with
Knotty problem matching primitives simultaneously obtain relevance degree；
The degree of correlation judges
If all relevance degrees are respectively less than relevance threshold, it is determined that several candidate answers of problem answers not with difficulty
Problem matching；Otherwise the highest candidate answers of relevance degree are determined as most matching problem answer；
The relevancy ranking includes being ranked up several candidate answers of problem answers according to relevance degree size, will
The highest Railway Project answer of the degree of correlation is determined as final result and is showed in the form of list collection.
Meanwhile being answered the present invention also provides a kind of online knotty problem of education of middle and primary schools and recommending interaction systems, it is included in
Line knotty problem answers platform, and the online knotty problem answer platform is used for:
The interrogative of knotty problem is identified the heat studied in classification by the method for template matching by inputting difficult problem
The clause of knotty problem is divided into the keyword of several determinations by point by Weight algorithm, by the corresponding candidate text of knotty problem
The semantic frame of shelves by comparing extracting the paragraph of corresponding position as problem answers, by several candidate answers of problem answers into
Row relatedness computation and degree of correlation judgement and relevancy ranking.
Advantageous effects:
This patent uses inputting difficult problem, and the interrogative of knotty problem is identified in classification by the method for template matching
The clause of knotty problem is divided into the keyword of several determinations by the hot spot of research by Weight algorithm, and knotty problem is corresponding
Candidate documents semantic frame by comparing extract corresponding position paragraph as problem answers, by several times of problem answers
Answer is selected to carry out relatedness computation and degree of correlation judgement and relevancy ranking, due to being proposed with the mode of assembly line based on interconnection
The answer of network technology problem, and propose and introduce natural language understanding and asked using natural language based on Internet technology
It answers and is pre-processed, since knotty problem Internet-based answer can be applied in Web education, can establish student's
The progress of record tracks student excavates the mode of student, accomplishes to teach students in accordance with their aptitude, and the problem of student is answered and shared, can
Mutually to inspire, fills up and answered the problem of user proposes this blank using natural language recognition technology on the internet.
Detailed description of the invention
Fig. 1 is that a kind of online knotty problem answer of education of middle and primary schools of the present invention recommends interactive approach general flow chart；
Fig. 2 is that a kind of online knotty problem answer of education of middle and primary schools of the present invention recommends the specific flow chart of interactive approach；
Specific embodiment
The present invention is described further below in conjunction with attached drawing:
In figure:
S101- inputting difficult problem；
S102- identifies problem category；
S103- keyword extraction；
S104- problem answers extract；
S105- relatedness computation；
The interrogative of S201- knotty problem identifies the hot spot studied in classification by the method for template matching；
The clause of knotty problem is divided into the keyword of several determinations by S202- by Weight algorithm；
The paragraph of corresponding position is made the semantic frame of the corresponding candidate documents of knotty problem by comparing extraction by S203-
For problem answers；
Several candidate answers of problem answers are carried out relatedness computation and degree of correlation judgement to S204- and the degree of correlation is arranged
Sequence；
Embodiment:
The present embodiment: as shown in Figure 1, 2, a kind of online knotty problem answer recommendation interactive approach of education of middle and primary schools, packet
It includes:
Inputting difficult problem S101；
Identification problem category S102: the interrogative of knotty problem identified by the method for template matching and is studied in classification
Hot spot S201；
Keyword extraction S103: the clause of knotty problem is divided into the keyword of several determinations by Weight algorithm
S202；
Problem answers extract S104: will be corresponding by comparing extraction by the semantic frame of the corresponding candidate documents of knotty problem
The paragraph of position is as problem answers S203；
Relatedness computation S105: several candidate answers of problem answers are subjected to relatedness computation S105 and the degree of correlation judges
And relevancy ranking S204.
Due to using inputting difficult problem, the interrogative of knotty problem is identified in classification by the method for template matching and is ground
The clause of knotty problem is divided into the keyword of several determinations by the hot spot studied carefully by Weight algorithm, and knotty problem is corresponding
The semantic frame of candidate documents extracts the paragraph of corresponding position as problem answers, by several candidates of problem answers by comparing
Answer carries out relatedness computation and degree of correlation judgement and relevancy ranking, due to being proposed with the mode of assembly line based on internet
The answer of technical problem, and propose and introduce natural language understanding and using natural language to the question and answer based on Internet technology
It is pre-processed, since knotty problem Internet-based answer can be applied in Web education, can establish the shelves of student
Case tracks the progress of student, excavates the mode of student, accomplishes to teach students in accordance with their aptitude, the problem of student is answered and shared, can be with
It mutually inspires, has filled up and answered the problem of user proposes this blank using natural language recognition technology on the internet.
The inputting difficult problem S101 include the interrogative of knotty problem what includes, who, how, how, when, be
How many and other related interrogatives.
Due to using inputting difficult problem, the inputting difficult problem include the interrogative of knotty problem what includes, who,
How, how, when, be how many and other related interrogatives, due to general problem types can be roughly divided into " asking people ",
" asking place ", " asking title ", " asking time and date ", " asking money ", " asking number and other ", and general problem is usually
" what ", " who ", " how ", " how ", " when ", " being how many " etc., can be by problem reduction, convenient for doubting by above-mentioned classification
The answer of difficult problem.
If the identification problem category S102 further includes that a no less than template matches with knotty problem, according to greed
Algorithm is matched according to maximum match principle.
Matching template in the method for the template matching includes:
| Type | The interrogative of knotty problem | With reference to sentence |
| Ask people | Who, who | Whom the author of this article is |
| Ask place | Where | Where is the provincial capital in Hunan Province |
| Ask title | What is | What the abbreviation in Hunan is |
| Ask time and date | It when is | When your birthday is |
| Ask money | How much | This computer how much |
| Ask number | How much is long how many, width | How much is this tree height |
| Ask other | How, what if | How to improve the health of the people |
Due to using identification problem category: the interrogative of knotty problem being identified in classification by the method for template matching and is ground
The hot spot studied carefully；If the identification problem category further includes that a no less than template matches with knotty problem, calculated according to greed
Method is matched according to maximum match principle, since the method based on template matching is the hot spot studied in identification classification, if
There is multiple template to match with problem, is matched according to the maximum matched principle of greedy algorithm selection, the difficulty such as proposed is asked
Topic is " whom current men's 100-metre dash world record holder is? " comprising " whom is ", then who matches with problem, this feelings
Under condition, problem types just for Whom, the characteristics of being identified by the method for template matching be it is simple and direct, this improves matchings
Efficiency.
The keyword extraction S103 includes that the determination principle of keyword includes removing knotty problem other than interrogative
Using noun or verb or adjective and the word of other parts of speech as keyword.
The Weight algorithm of the keyword extraction S103 includes:
It is described to contain sexual keyword and be determined as strong restricted key if keyword is that must contain sexual keyword
Word；
If keyword be it is nonessential contain sexual keyword, it is described to contain sexual keyword and be determined as weak restricted key
Word.
Due to using keyword extraction: the clause of knotty problem is divided into the key of several determinations by Weight algorithm
Word, the keyword extraction include the determination principle of keyword include knotty problem is removed other than interrogative with noun or
Verb or adjective and the word of other parts of speech are as keyword, if the Weight algorithm of the keyword extraction includes: keyword
It is for sexual keyword must be contained, then described to contain sexual keyword and be determined as strong restricted keyword；If keyword be it is non-must
Sexual keyword must be contained, then it is described to contain sexual keyword and be determined as weak restricted keyword, due to being closed to the problem of proposition
The accuracy that keyword extracts influences subsequent search result, and in general, keyword is mostly with noun, verb, adjective
Etc. parts of speech composition, in real application systems, can removing interrogative, unexpectedly most of word can mention in this way as keyword
The precision of height retrieval, keyword is endowed different weights, to problem by limiting by force very much as " must contain " property keyword
Qualitative effect, it is very small if a possibility that answer obtained without containing " must contain " property keyword is correct, thus may be used
See, the accuracy of retrieval can be improved by the effect of keyword, and certain words are not the keyword of original problem in answer,
These can be considered the extension of keyword, the recall rate and accuracy using extended mode to improve system.
Described problem answer extracting S104 includes that the semantic frame by the corresponding candidate documents of knotty problem passes through ratio
It, will be all over compared with all sentences for extracting the paragraph of corresponding position as problem answers and include the traversal corresponding candidate documents of knotty problem
All sentences gone through obtain semantic frame by semantic analysis, by the semantic frame of the semantic frame of candidate documents and knotty problem
It is compared, if successful match, the word and sentence for extracting the corresponding position of candidate documents are as problem answers.
The specific algorithm of described problem answer extracting S104 includes:
A, knotty problem keyword extraction S103 is determined treated problem category；
B, all paragraphs of all candidate documents are scanned and find out the candidate answers to match with knotty problem；
C, each candidate answers are distributed to an initial score；
D, positive or negative points operation is carried out according to Weight algorithm；
E, the possibility answer of weight limit score is selected；
F, will likely answer score be set as zero；
G, repeat selection weight limit score possibility answer and will likely answer score be set as zero, up to select it is several
A suitable candidate answers are as problem answers.
Due to using problem answers extract: by the semantic frame of the corresponding candidate documents of knotty problem extract by comparing by
The paragraph of corresponding position is as problem answers；Described problem answer extracting includes described by the corresponding candidate documents of knotty problem
Semantic frame includes the traversal corresponding candidate documents of knotty problem as problem answers by comparing the paragraph for extracting corresponding position
All sentences, by all sentences of traversal by semantic analysis obtain semantic frame, by the semantic frame of candidate documents and doubt
The semantic frame of difficult problem is compared, if successful match, the word and sentence for extracting the corresponding position of candidate documents, which are used as, is asked
Inscribe answer, the specific algorithm of described problem answer extracting includes: a, determines knotty problem keyword extraction treated problem class
Not；B, all paragraphs of all candidate documents are scanned and find out the candidate answers to match with knotty problem；C, each candidate is answered
Case distributes an initial score；D, positive or negative points operation is carried out according to Weight algorithm；E, the possibility answer of weight limit score is selected；
F, will likely answer score be set as zero；G, repeat selection weight limit score possibility answer and will likely answer score be arranged
It is zero, until selecting several suitable candidate answers as problem answers, since answer is during extraction, traverses time
All sentences in selection shelves, to current sentence carry out semantic analysis, obtain semantic frame, then by obtained semantic frame with
Inquiry sentence semantic frame is compared, and the word or sentence that can extract corresponding position if matching are commonly answered as answer
Case extraction algorithm is as follows: 1, determining the type of answer after issue handling；2, all paragraphs of scanning are found out answers with 1 matched candidate
Case；3, an initial score is distributed to each answer；4, positive or negative points calculating is carried out according to weight；5, that selects largest score can
It can answer；6, zero is set by the answer score obtained in 5；7,5,6 processes are repeated, are answered until selecting several suitable candidates
Case；
The relatedness computation S105 includes that several candidate answers of problem answers are passed through to semantic knowledge resources bank and are doubted
Difficult problem matching primitives simultaneously obtain relevance degree；
The degree of correlation judges
If all relevance degrees are respectively less than relevance threshold, it is determined that several candidate answers of problem answers not with difficulty
Problem matching；Otherwise the highest candidate answers of relevance degree are determined as most matching problem answer；
The relevancy ranking includes being ranked up several candidate answers of problem answers according to relevance degree size, will
The highest Railway Project answer of the degree of correlation is determined as final result and is showed in the form of list collection.
Due to using relatedness computation: several candidate answers of problem answers being carried out relatedness computation and the degree of correlation judges
And relevancy ranking, the relatedness computation include by several candidate answers of problem answers by semantic knowledge resources bank with
Knotty problem matching primitives simultaneously obtain relevance degree, if degree of correlation judgement includes: that all relevance degrees are respectively less than the degree of correlation
Threshold value, it is determined that several candidate answers of problem answers are not matched with knotty problem；Otherwise by the highest candidate of relevance degree
Answer is determined as most matching problem answer；The relevancy ranking includes several candidate answers by problem answers according to the degree of correlation
Value size is ranked up, and the highest Railway Project answer of the degree of correlation is determined as final result and is showed in the form of list collection,
Method due to calculating the degree of correlation calculates language of all the problems in user query and knowledge base including the use of semantic knowledge resources bank
The adopted degree of correlation can think that do not have user to be proposed in knowledge base asks if the value of semantic relevancy is less than threshold value M
Topic is improved to user, is calculated otherwise just by answer corresponding in knowledge base and the problem of user query semantic relevancy highest
It out after the degree of correlation of all candidate answers, is ranked up according to the size of the degree of correlation, the highest several candidates of the degree of correlation is taken to answer
Case is showed as final result with tabular form.
Meanwhile being answered the present invention also provides a kind of online knotty problem of education of middle and primary schools and recommending interaction systems, it is included in
Line knotty problem answers platform, and the online knotty problem answer platform is used for:
The interrogative of knotty problem is identified by the method for template matching and to be studied in classification by inputting difficult problem S101
The clause of knotty problem is divided into the keyword S202 of several determinations by hot spot S201 by Weight algorithm, by knotty problem phase
The semantic frame for the candidate documents answered extracts the paragraph of corresponding position as problem answers, by the several of problem answers by comparing
Candidate answers carry out relatedness computation S105 and degree of correlation judgement and relevancy ranking S204.
Working principle:
This patent is identified in classification by inputting difficult problem, by the interrogative of knotty problem by the method for template matching
The clause of knotty problem is divided into the keyword of several determinations by the hot spot of research by Weight algorithm, and knotty problem is corresponding
Candidate documents semantic frame by comparing extract corresponding position paragraph as problem answers, by several times of problem answers
Answer is selected to carry out relatedness computation and degree of correlation judgement and relevancy ranking, due to being proposed with the mode of assembly line based on interconnection
The answer of network technology problem, and propose and introduce natural language understanding and asked using natural language based on Internet technology
It answers and is pre-processed, since knotty problem Internet-based answer can be applied in Web education, can establish student's
The progress of record tracks student excavates the mode of student, accomplishes to teach students in accordance with their aptitude, and the problem of student is answered and shared, can
Mutually to inspire, the present invention solve the prior art exist lack using Internet resources as the knowledge base of system using natural language knowledge
Other technology utilizes natural language recognition technology back and forth on the internet the problem of answering the problem of user proposes, to have to have filled up
Answer user propose the problem of this blank, convenient for knotty problem answer, improve matching efficiency, improve system recall rate and
The advantageous effects of accuracy.
Using technical solution of the present invention or those skilled in the art under the inspiration of technical solution of the present invention, design
Similar technical solution out, and reach above-mentioned technical effect, it is to fall into protection scope of the present invention.
Claims (10)
1. a kind of online knotty problem answer of education of middle and primary schools recommends interactive approach characterized by comprising
Inputting difficult problem；
Identification problem category: the interrogative of knotty problem is identified into the hot spot studied in classification by the method for template matching；
Keyword extraction: the clause of knotty problem is divided into the keyword of several determinations by Weight algorithm；
Problem answers extract: the semantic frame of the corresponding candidate documents of knotty problem is extracted by comparing by the section of corresponding position
It falls as problem answers；
Relatedness computation: several candidate answers of problem answers are subjected to relatedness computation and degree of correlation judgement and the degree of correlation is arranged
Sequence.
2. a kind of online knotty problem answer of education of middle and primary schools according to claim 1 recommends interactive approach, feature
Be, the inputting difficult problem include the interrogative of knotty problem what includes, who, how, how, when, be how much with
And other related interrogatives.
3. a kind of online knotty problem answer of education of middle and primary schools according to claim 1 recommends interactive approach, feature
It is, if the identification problem category further includes that a no less than template matches with knotty problem, according to greedy algorithm root
It is matched according to maximum match principle.
4. a kind of online knotty problem answer of education of middle and primary schools according to claim 3 recommends interactive approach, feature
It is, the matching template in the method for the template matching includes:
5. a kind of online knotty problem answer of education of middle and primary schools according to claim 1 recommends interactive approach, feature
Be, the keyword extraction include the determination principle of keyword include by knotty problem remove interrogative other than with noun,
Or verb or adjective and the word of other parts of speech are as keyword.
6. a kind of online knotty problem answer of education of middle and primary schools according to claim 5 recommends interactive approach, feature
It is, the Weight algorithm of the keyword extraction includes:
It is described to contain sexual keyword and be determined as strong restricted keyword if keyword is that must contain sexual keyword；
If keyword be it is nonessential contain sexual keyword, it is described to contain sexual keyword and be determined as weak restricted keyword.
7. a kind of online knotty problem answer of education of middle and primary schools according to claim 1 recommends interactive approach, feature
It is, described problem answer extracting includes that the semantic frame by the corresponding candidate documents of knotty problem extracts phase by comparing
It includes all sentences for traversing the corresponding candidate documents of knotty problem that the paragraph of position, which is answered, as problem answers, by all of traversal
Sentence obtains semantic frame by semantic analysis, and the semantic frame of the semantic frame of candidate documents and knotty problem is compared
Compared with if successful match, the word and sentence for extracting the corresponding position of candidate documents are as problem answers.
8. a kind of online knotty problem answer of education of middle and primary schools according to claim 7 recommends interactive approach, feature
It is, the specific algorithm of described problem answer extracting includes:
A, knotty problem keyword extraction is determined treated problem category；
B, all paragraphs of all candidate documents are scanned and find out the candidate answers to match with knotty problem；
C, each candidate answers are distributed to an initial score；
D, positive or negative points operation is carried out according to Weight algorithm；
E, the possibility answer of weight limit score is selected；
F, will likely answer score be set as zero；
G, repeat the possibility answer of selection weight limit score and will likely answer score be set as zero, up to selecting several conjunctions
Suitable candidate answers are as problem answers.
9. a kind of online knotty problem answer of education of middle and primary schools according to claim 1 recommends interactive approach, feature
It is, the relatedness computation includes that several candidate answers of problem answers are passed through semantic knowledge resources bank and knotty problem
With calculating and obtain relevance degree；
The degree of correlation judges
If all relevance degrees are respectively less than relevance threshold, it is determined that several candidate answers of problem answers not with knotty problem
Matching；Otherwise the highest candidate answers of relevance degree are determined as most matching problem answer；
The relevancy ranking includes being ranked up several candidate answers of problem answers according to relevance degree size, will be related
Highest Railway Project answer is spent to be determined as final result and show in the form of list collection.
10. a kind of online knotty problem answer of education of middle and primary schools recommends interaction systems, which is characterized in that asked including online difficulty
The key to exercises answers platform, and the online knotty problem answer platform is used for:
The interrogative of knotty problem is identified the hot spot studied in classification by the method for template matching by inputting difficult problem, will
The clause of knotty problem is divided into the keyword of several determinations by Weight algorithm, by the language of the corresponding candidate documents of knotty problem
Adopted frame, as problem answers, several candidate answers of problem answers is carried out related by comparing the paragraph for extracting corresponding position
Degree calculates and the degree of correlation judges and relevancy ranking.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910248955.1A CN109977214A (en) | 2019-03-29 | 2019-03-29 | A kind of online knotty problem answer recommendation interactive approach and system of education of middle and primary schools |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910248955.1A CN109977214A (en) | 2019-03-29 | 2019-03-29 | A kind of online knotty problem answer recommendation interactive approach and system of education of middle and primary schools |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109977214A true CN109977214A (en) | 2019-07-05 |
Family
ID=67081599
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910248955.1A Pending CN109977214A (en) | 2019-03-29 | 2019-03-29 | A kind of online knotty problem answer recommendation interactive approach and system of education of middle and primary schools |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109977214A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112395397A (en) * | 2019-08-13 | 2021-02-23 | 北京国双科技有限公司 | Automatic response method, device, electronic equipment and storage medium |
| CN114385830A (en) * | 2022-01-14 | 2022-04-22 | 中国建设银行股份有限公司 | Operation and maintenance knowledge online question and answer method and device, electronic equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103425640A (en) * | 2012-05-14 | 2013-12-04 | 华为技术有限公司 | Multimedia questioning-answering system and method |
| CN103902652A (en) * | 2014-02-27 | 2014-07-02 | 深圳市智搜信息技术有限公司 | Automatic question-answering system |
| CN104536991A (en) * | 2014-12-10 | 2015-04-22 | 乐娟 | Answer extraction method and device |
| CN105824933A (en) * | 2016-03-18 | 2016-08-03 | 苏州大学 | Automatic question-answering system based on theme-rheme positions and realization method of automatic question answering system |
| CN106777232A (en) * | 2016-12-26 | 2017-05-31 | 上海智臻智能网络科技股份有限公司 | Question and answer abstracting method, device and terminal |
| CN108804529A (en) * | 2018-05-02 | 2018-11-13 | 深圳智能思创科技有限公司 | A kind of question answering system implementation method based on Web |
-
2019
- 2019-03-29 CN CN201910248955.1A patent/CN109977214A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103425640A (en) * | 2012-05-14 | 2013-12-04 | 华为技术有限公司 | Multimedia questioning-answering system and method |
| CN103902652A (en) * | 2014-02-27 | 2014-07-02 | 深圳市智搜信息技术有限公司 | Automatic question-answering system |
| CN104536991A (en) * | 2014-12-10 | 2015-04-22 | 乐娟 | Answer extraction method and device |
| CN105824933A (en) * | 2016-03-18 | 2016-08-03 | 苏州大学 | Automatic question-answering system based on theme-rheme positions and realization method of automatic question answering system |
| CN106777232A (en) * | 2016-12-26 | 2017-05-31 | 上海智臻智能网络科技股份有限公司 | Question and answer abstracting method, device and terminal |
| CN108804529A (en) * | 2018-05-02 | 2018-11-13 | 深圳智能思创科技有限公司 | A kind of question answering system implementation method based on Web |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112395397A (en) * | 2019-08-13 | 2021-02-23 | 北京国双科技有限公司 | Automatic response method, device, electronic equipment and storage medium |
| CN114385830A (en) * | 2022-01-14 | 2022-04-22 | 中国建设银行股份有限公司 | Operation and maintenance knowledge online question and answer method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107329949B (en) | Semantic matching method and system | |
| CN109213999B (en) | Subjective question scoring method | |
| CN105989040B (en) | Intelligent question and answer method, device and system | |
| CN103425635B (en) | Method and apparatus are recommended in a kind of answer | |
| KR101923650B1 (en) | System and Method for Sentence Embedding and Similar Question Retrieving | |
| CN108595525B (en) | Lawyer information processing method and system | |
| CN104137102A (en) | Non-factoid question answering system and computer program | |
| CN107247751B (en) | LDA topic model-based content recommendation method | |
| Abdi et al. | A question answering system in hadith using linguistic knowledge | |
| CN108681548B (en) | Lawyer information processing method and system | |
| CN111221962A (en) | Text emotion analysis method based on new word expansion and complex sentence pattern expansion | |
| CN113342958B (en) | Question-answer matching method, text matching model training method and related equipment | |
| Santhanavijayan et al. | Automatic generation of multiple choice questions for e-assessment | |
| Citawan et al. | Automatic essay scoring in E-learning system using LSA method with N-gram feature for Bahasa Indonesia | |
| CN112417119A (en) | Open domain question-answer prediction method based on deep learning | |
| CN109977214A (en) | A kind of online knotty problem answer recommendation interactive approach and system of education of middle and primary schools | |
| Kwankajornkiet et al. | Automatic multiple-choice question generation from Thai text | |
| Omran et al. | Automatic essay grading system for short answers in English language | |
| Yogish et al. | Survey on trends and methods of an intelligent answering system | |
| JP6942759B2 (en) | Information processing equipment, programs and information processing methods | |
| Riza et al. | Natural language processing and levenshtein distance for generating error identification typed questions on TOEFL | |
| Dündar et al. | A Hybrid Approach to Question-answering for a Banking Chatbot on Turkish: Extending Keywords with Embedding Vectors. | |
| Asprino et al. | A large visual question answering dataset for cultural heritage | |
| CN116362331A (en) | Knowledge point filling method based on man-machine cooperation construction knowledge graph | |
| Blšták | Automatic Question Generation Based on Sentence Structure Analysis. |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190705 |