CN108170679A - It can recognize that the semantic matching method and system of natural language description based on computer - Google Patents
It can recognize that the semantic matching method and system of natural language description based on computer Download PDFInfo
- Publication number
- CN108170679A CN108170679A CN201711460123.3A CN201711460123A CN108170679A CN 108170679 A CN108170679 A CN 108170679A CN 201711460123 A CN201711460123 A CN 201711460123A CN 108170679 A CN108170679 A CN 108170679A
- Authority
- CN
- China
- Prior art keywords
- words
- natural language
- root
- word
- computer
- 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
Classifications
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/237—Lexical tools
- G06F40/247—Thesauruses; Synonyms
Abstract
The invention belongs to programming technique fields, and in particular to can recognize that the semantic matching method of natural language description and its corresponding semantic matches system based on computer.This can recognize that the semantic matching method of natural language description includes step based on computer：Step S1)：Natural language requirement description as reference, is constrained to the structure for having the step of logic by the logic and step limited with the syntax rule of object language；Step S2)：To fixed clause in the natural language requirement description after constraint, the candidate set of words for including the root in natural language requirement description is obtained；Step S3)：Message name in object language/operation name is segmented, obtains the spare set of words for including the root in message name/operation name；Step S4)：Calculate the matching degree of candidate set of words and spare set of words.The semantic matching method and semantic matches system, energy coordinates user and developer, using upper disagreement, realize the automated programming of machine language for natural language.
Description
Technical field
The invention belongs to programming technique fields, and in particular to a kind of semanteme that can recognize that natural language description based on computer
Matching process and its semantic matches system that can recognize that natural language description based on computer accordingly.
Background technology
Natural language is still the description language of current software requirement document.From the functional requirement of natural language description to stream
The automatically generating of journey, which not only assists in user and developer, rapidly to reach common understanding in demand, moreover it is possible to accelerate flow
Exploitation.
But since user is different with the focus of developer, they also tend to the description of demand different.User
With developer during with natural language description functional requirement, user concern the function that software can be provided and
Attainable performance level of institute etc., developer may then go to portray the demand of software from the angle of technology；Moreover, they are not
Know specific message used in development language and operation naming rule, they to the notional word used in the description of demand simultaneously
It is not necessarily just the same with the word that uses in the message name in development language and operation name.In addition, it in most cases, uses
Family is not familiar with the term and technical problem of those professions.
But current software requirement document is most of or is write with natural language, there is two aspect reasons among these：When
Because user and developer are mostly without the ability of formalized description demand；Second is that because natural language vocabulary is enriched, express
Ability is powerful.But natural language is also inevitably there are shortcoming, including ambiguity, ambiguity and inconsistency.
In order to make up the deficiency of natural language, need a kind of the flow requirement description of natural language expressing can be constrained
With the method for formalization so that computer is it will be appreciated that demand.How coordinates user and developer are for natural language application
On disagreement, become a technical problem to be solved urgently.
Invention content
The technical problems to be solved by the invention are for above-mentioned deficiency in the prior art, and providing one kind can based on computer
It identifies the semantic matching method of natural language description and its can recognize that the semanteme of natural language description based on computer accordingly
Match system can effectively eliminate user and developer for natural language using upper disagreement, realize the automatic volume of machine language
Journey.
Technical solution is that this can recognize that natural language description based on computer used by solving present invention problem
Semantic matching method, including step：
Step S1)：As reference, natural language demand is retouched for the logic and step limited with the syntax rule of object language
It states and is constrained to the structure for having the step of logic；
Step S2)：To fixed clause in the natural language requirement description after constraint, acquisition includes natural language demand and retouches
The candidate set of words of root in stating；
Step S3)：Message name in object language/operation name is segmented, acquisition is included in message name/operation name
The spare set of words of root；
Step S4)：Calculate the matching degree of candidate set of words and spare set of words.
Preferably, step S2) include：
Step S21)：According to the determiner of setting, the statement of requirements of natural language description is obtained, after statement of requirements is segmented
Form primary set of words；
Step S22)：The stop words in primary set of words is removed, is formed and is applicable in set of words；
Step S23)：Synonym extension is carried out to each word being applicable in set of words；
Step S24)：Root reduction is carried out to extension set of words, obtains the root included in natural language requirement description
Candidate set of words.
Preferably, step S21) in, the determiner of object language setting is switched to using prefix as mark for statement of requirements
Know；
Step S22) in, auxiliary word, preposition, conjunction class are prestored as stop words as stop words dictionary；
Step S23) in, synonym extension is carried out to each word being applicable in set of words according to synonym dictionary；
Step S24) in, root retrieving algorithm is Porter algorithms or Lucene algorithms.
Preferably, step S4) including step：
Step S41)：The word of spare set of words is traversed, there are the words of intersection with candidate set of words for screening；
Step S42)：To meeting the word of intersection, matching degree is calculated.
Preferably, step S4) in, the formula of the matching degree of candidate set of words and spare set of words is：
Wherein, count is the word number of the semantic similarity found, | wordsetA| to be segmented in requirement description sentence
Number, | wordsetB| for the participle number in message name/operation name.
A kind of semantic matches system that can recognize that natural language description based on computer, including constraints module, candidate word
Set forms module, spare set of words forms module and matching module, wherein：
The constraints module, for the logic and step that are limited with the syntax rule of object language as reference, by nature
Language needs description is constrained to the structure for having the step of logic；
Candidate's set of words forms module, for fixed clause in the natural language requirement description after constraint,
Obtain the candidate set of words for including the root in natural language requirement description；
The spare set of words forms module, for being segmented to the message name in object language/operation name, obtains
Include the spare set of words of the root in message name/operation name；
The matching module, for calculating the matching degree of candidate set of words and spare set of words.
Preferably, the candidate set of words forms module and includes primary set of words unit, is applicable in set of words list
Member, synonym expanding element and root reduction unit, wherein：
The primary set of words unit, for the determiner according to setting, obtains the statement of requirements of natural language description,
Primary set of words is formed after statement of requirements is segmented；
The applicable set of words unit for removing the stop words in primary set of words, forms and is applicable in set of words；
The synonym expanding element, for carrying out synonym extension to each word being applicable in set of words；
The root reduction unit, for carrying out root reduction to extension set of words, acquisition includes natural language demand
The candidate set of words of root in description.
Preferably, in the primary set of words unit, the determiner of object language setting is switched to for statement of requirements
Using prefix as mark；
In the applicable set of words unit, auxiliary word, preposition, conjunction class are prestored as stop words as stop words word
Library；
In the synonym expanding element, synonym is carried out to each word being applicable in set of words according to synonym dictionary
Extension；
In the root reduction unit, root retrieving algorithm is Porter algorithms or Lucene algorithms.
Preferably, the matching module includes asking presentate member, matching unit, wherein：
Described to ask presentate first, for traversing the word of spare set of words, there are intersections with candidate set of words for screening
Word；
The matching unit to meeting the word of intersection, calculates matching degree.
Preferably, in the matching unit, the formula of the matching degree of candidate set of words and spare set of words is：
Wherein, count is the word number of the semantic similarity found, | wordsetA| to be segmented in requirement description sentence
Number, | wordsetB| for the participle number in message name/operation name.
The beneficial effects of the invention are as follows：This can recognize that the semantic matching method and its phase of natural language description based on computer
The semantic matches system answered on the basis of participle, removal stop words, root reduction and similar calculating, increases synonym extension
It, can coordinates user and developer couple with the matching with message name/operation name suitable for requirement description with the similar calculating of modification
In natural language using upper disagreement, the automated programming of machine language is realized.
Description of the drawings
Fig. 1 is the flow for the semantic matching method that can recognize that natural language description in the embodiment of the present invention based on computer
Figure；
Fig. 2 is to obtain to scheme the step of including the candidate set of words of root in requirement description in the embodiment of the present invention；
Fig. 3 is the structural frames for the semantic matches system that can recognize that natural language description in the embodiment of the present invention based on computer
Figure；
In figure：
1- constraints modules；2- candidates set of words forms module；The spare set of words of 3- forms module；4- matching modules.
Specific embodiment
For those skilled in the art is made to more fully understand technical scheme of the present invention, below in conjunction with the accompanying drawings and specific embodiment party
Formula is to can recognize that the semantic matching method of natural language description the present invention is based on computer and its can be known based on computer accordingly
The semantic matches system of other natural language description is described in further detail.
In order to establish bridge between requirement description and development language, the present invention is based on from the angle of semantic matches
Root and synonym form the dictionary (can be understood as English dictionary dictionary wordnet) with level, propose a kind of based on meter
It is upper that calculation machine can recognize that the semantic matching method of natural language description, energy coordinates user and developer apply natural language
Disagreement realizes the automated programming of machine language, greatly accelerates project process.
As shown in Figure 1, the semantic matching method of natural language description is can recognize that in the present invention based on computer, including as follows
Step：
Step S1)：As reference, natural language demand is retouched for the logic and step limited with the syntax rule of object language
It states and is constrained to the structure for having the step of logic.
The flow functional requirement of natural language description has certain step, and this step passes through the preposition in sentence
It embodies, preposition such as after, if, then, or else, at the same time etc..However, it is this for the mankind very
Simply have the step of logic relationship to be but not easy to be identified and understood by computer.Therefore specified constraint rule are needed
Then, the preparation before natural language is converted to object language is carried out so that user and developer carry out demand with the constraint rule
Description, in order to which requirement description is made directly to embody step to computer.Goal language can be selection programming
Computer language.
In this step, natural language requirement description is constrained, appears as a kind of knot for thering is logic to have step
Structure, moreover, forming the word of these structures for having logic to have step can be integrated to form dictionary wordsetA。
Step S2)：To fixed clause in the natural language requirement description after constraint, acquisition includes natural language demand and retouches
The candidate set of words of root in stating.
User and developer are when with natural language description functional requirement, in this case it is not apparent that specific in program file
Message and operation name information, they to the notional word used in the description of demand might not and program file in message name
It is just the same with the word that uses in operation name.In this step, it is limited using computer with the syntax rule of object language
As reference, clause fixed in the requirement description after constraint is automatically formalized for logic and step.Formalization is usual
It is carried out according to target developing language (for example, to automate business process composition language BPEL), i.e., requirement description is converted into energy
A kind of enough language by computer understanding.The original intention of the present invention is for path combination language, and by that will constrain, treated will have
There is the requirement description of certain step, the corresponding sentence of path combination language is converted by formalization.So as to pass through formalization
Realize the bridge between the requirement description and object language of flow.
Therefore, in the step, from the angle of semantic matches, using synonym dictionary, based on root and synonym pair
Natural language requirement description carries out matching algorithm, to fixed clause in the natural language requirement description after constraint, including
The candidate set of words of root in natural language requirement description.
Below with reference to Fig. 2, final wordset is being obtained to the statement of requirements A of natural language descriptionAProcess carry out
It is described in detail.Specifically comprise the following steps：
Step S21)：According to the determiner of setting, the statement of requirements of natural language description is obtained, after statement of requirements is segmented
Form primary set of words.
Wherein, the statement of requirements A of natural language description obtains constraint sentence A' after constraint and formalization.If constraint
Sentence A' includes the determiner of whose setting, then extracts the natural language requirement description sentence of constraint sentence A', and is divided
Word obtains primary set of words as wordset'A.Under normal conditions, primary word is the target language that requirement description will convert
It says defined, therefore determiner can be set in advance, to be obtained from the dictionary of object language.
It is a kind of it is preferable that, the determiner that object language setting is switched to for statement of requirements A can be using prefix as marking
Know, for example, to automate business process composition language BPEL as an example, the prefix of constraint sentence A' as [RECEIVE] or
[INVOKE], then the natural language requirement description sentence for extracting constraint sentence A' are A ", primary word collection of the A " after participle
It is combined into wordset'A.Here, [RECEIVE] represents to receive a message, and [INVOKE] represents to call a service.
Step S22)：The stop words in primary set of words is removed, is formed and is applicable in set of words.
Under normal conditions, in statement of requirements other than the notional words such as name, adjective, verb class, it is also possible to auxiliary word,
Preposition, conjunction etc. are found with statement of requirements from all file destinations in semanteme without the function word of practical significance based on realizing
On most matched message and the purpose of operation, these will generate interference with semantic incoherent word to semantic matching, because
It is necessary to weed out them during matching degree is calculated for this.Therefore, it is further preferred that, in order to ensure the pure of dictionary
Net property is stored using auxiliary word, preposition, conjunction etc. as stop words in advance as stop words dictionary D.According to stop words dictionary D, to word
Language collection is combined into wordset'AStop words is removed, i.e., from wordset'AMiddle removal stop words obtains being applicable in set of wordsFor wordset'AIn any one word w, if w ∈ D,
Step S23)：Synonym extension is carried out to each word being applicable in set of words.
In this step, according to synonym dictionary C (can be understood as total english dictionary) to being applicable in set of wordsIn each word carry out synonym extension.ForIn any one word w, in synonym dictionary C
The synonym collection synonyms (w) of w is inquired, all synonyms of w are added toIn,The set of words that is expanded wordset "A。
Step S24)：Root reduction is carried out to extension set of words.
To extending set of words wordset "AIn the step of carrying out root reduction, for wordset "AIn any one
Word w, the root w' of w is calculated with root retrieving algorithm, and wordset " is replaced with w'AW, obtain obtain include nature language
Say the candidate set of words wordset of the root in requirement descriptionA, i.e. wordsetA=wordset "A-w+w'.Here, w' remembers
For Porter (w).Specific root retrieving algorithm can be Porter algorithms or Lucene algorithms, not limit here.
By above steps, carry out gradation successively to the statement of requirements of natural language description, remove stop words, synonym
Extension and the processing of root reduction, you can the root in the sentence of natural language description, the extension with root same level are obtained, and
It is not interfered again by stop words, therefore can realize user and developer extension packets semantic in communication process to the maximum extent
Hold, to provide the candidate matches basis of more horn of plenty to computer language conversion.
Step S3)：Message name in object language/operation name is segmented, acquisition is included in message name/operation name
The spare set of words of root.
In this step, message name/operation name is segmented, the set of words after formation is the requirement description after constraint
In fixed clause.Message name/spare set of words of the operation name B after participle is wordsetB。
Here it will be understood that since each computer language all has particularity, to the limit of message name/operation name B
Concrete syntax is needed to make a concrete analysis of surely.
Step S4)：Calculate the matching degree of candidate set of words and spare set of words.
The present invention in line with by the flow functional requirement of natural language description be automatically converted to development language description application, be
Promote the matching algorithm that accuracy increases root and synonym in terms of semantic processes.Therefore, to step S2) in obtain
Candidate set of words and step S3) in obtained spare set of words carry out matching degree calculating, to ensure in spare set of words
With the word in maximum similarity matching candidate set of words.
At present, matching degree computational methods include Dice-Euclidean similarity algorithms.In the present embodiment, in order to more
Accurately search natural language corresponding to flow, it is contemplated that root and synonym, to similarity calculation algorithm Dice algorithms into
It has gone improvement, wordset is calculated with DicePlus algorithmsAAnd wordsetBMatching degree.
The similarity calculation algorithm DicePlus of improved extension includes step：
Step S41)：The word of spare set of words is traversed, there are the words of intersection with candidate set of words for screening.
In this step, spare set of words wordset is traversedBIn each word, if wordsetBIn word w
In wordsetAThe synonym of middle presence or word w are in wordsetAMiddle presence, so as to judge spare set of words
wordsetBIn word and candidate set of words wordsetAIn word whether there is intersection.
Step S42)：To meeting the word of intersection, matching degree is calculated.
It is in order to find the program statement for meeting matching degree instead of corresponding demand descriptive statement, if looked for calculate matching degree
Less than developer oneself is then needed to write corresponding sentence.In this step, candidate set of words is calculated using the following formula
wordsetAWith spare set of words wordsetBMatching degree
Wherein, count is the word number of the semantic similarity found, | wordsetA| to be segmented in requirement description sentence
Number, | wordsetB| for the participle number in message name/operation name.
Based on above-mentioned matching degree algorithm and similarity algorithm, the demand of natural language description can be converted into computer
The description language that can be identified, you can realize the computer automatic programming of the sentence according to natural language description.At this point, it uses
Notional word used in the requirement description of family might not just the same with word used in developer (such as receive and get be
Represent to receive message), still can precisely it be matched.
Natural language and computer language are in lasting development and update, and semantic matching method of the invention can not
Can have exhaustive, self study addition in use that can be afterwards, accumulate gradually dictionary, enriches and improve matching constantly.
The present invention the semantic matching method that can recognize that natural language description based on computer, participle, remove stop words,
Root restore and similar calculating on the basis of, increase synonym extension and change similar calculating, with suitable for requirement description with
The matching of message name/operation name, energy coordinates user and developer, using upper disagreement, realize machine language for natural language
Automated programming.
Correspondingly, the present embodiment also provides the semantic matches system that can recognize that natural language description based on computer, can assist
Family and developer is called, using upper disagreement, to realize the automated programming of machine talk for natural language.
As shown in figure 3, should based on computer can recognize that natural language description semantic matches system include constraints module 1,
Candidate set of words forms module 2, spare set of words forms module 3 and matching module 4, wherein：
Constraints module 1, for the logic and step that are limited with the syntax rule of object language as reference, by natural language
Requirement description is constrained to the structure for having the step of logic；
Candidate set of words forms module 2, for fixed clause in the natural language requirement description after constraint, obtaining
Candidate set of words including the root in natural language requirement description；
Spare set of words forms module 3, for being segmented to the message name in object language/operation name, is wrapped
Include the spare set of words of the root in message name/operation name；
Matching module 4, for calculating the matching degree of candidate set of words and spare set of words.
Wherein, candidate set of words forms module 2 and includes primary set of words unit, be applicable in set of words unit, be synonymous
Word expanding element and root reduction unit, wherein：
Primary set of words unit for the determiner according to setting, obtains the statement of requirements of natural language description, need to
Sentence is asked to form primary set of words after segmenting.In primary set of words unit, object language is switched to for statement of requirements and is set
Fixed determiner is using prefix as mark；
Set of words unit is applicable in, for removing the stop words in primary set of words, is formed and is applicable in set of words.Suitable
With in word aggregation units, auxiliary word, preposition, conjunction class are prestored as stop words as stop words dictionary；
Synonym expanding element, for carrying out synonym extension to each word being applicable in set of words.Expand in synonym
It opens up in unit, synonym extension is carried out to each word being applicable in set of words according to synonym dictionary；
Root reduction unit, for carrying out root reduction to extension set of words, acquisition includes natural language requirement description
In root candidate set of words.In root reduction unit, root retrieving algorithm is calculated for Porter algorithms or Lucene
Method.
Matching module 4 includes asking presentate member, matching unit, wherein：
Ask presentate first, for traversing the word of spare set of words, there are the words of intersection with candidate set of words for screening；
Matching unit to meeting the word of intersection, calculates matching degree.
In matching unit, the formula of the matching degree of candidate set of words and spare set of words is：
Wherein, count is the word number of the semantic similarity found, | wordsetA| to be segmented in requirement description sentence
Number, | wordsetB| for the participle number in message name/operation name.
The present invention the semantic matches system that can recognize that natural language description based on computer, participle, remove stop words,
Root restore and similar calculating on the basis of, increase synonym extension and change similar calculating, with suitable for requirement description with
The matching of message name/operation name, energy coordinates user and developer, using upper disagreement, realize machine language for natural language
Automated programming.
It is understood that the principle that embodiment of above is intended to be merely illustrative of the present and the exemplary implementation that uses
Mode, however the present invention is not limited thereto.For those skilled in the art, in the essence for not departing from the present invention
In the case of refreshing and essence, various changes and modifications can be made therein, these variations and modifications are also considered as protection scope of the present invention.
Claims (10)
1. a kind of semantic matching method that can recognize that natural language description based on computer, which is characterized in that including step：
Step S1)：With the logic of the syntax rule restriction of object language and step as reference, by natural language requirement description about
Beam is the structure for having the step of logic；
Step S2)：To fixed clause in the natural language requirement description after constraint, acquisition is included in natural language requirement description
Root candidate set of words；
Step S3)：Message name in object language/operation name is segmented, obtains the root included in message name/operation name
Spare set of words；
Step S4)：Calculate the matching degree of candidate set of words and spare set of words.
2. the semantic matching method according to claim 1 that can recognize that natural language description based on computer, feature are existed
In step S2) include：
Step S21)：According to the determiner of setting, the statement of requirements of natural language description is obtained, is formed after statement of requirements is segmented
Primary set of words；
Step S22)：The stop words in primary set of words is removed, is formed and is applicable in set of words；
Step S23)：Synonym extension is carried out to each word being applicable in set of words；
Step S24)：Root reduction is carried out to extension set of words, obtains the time for including the root in natural language requirement description
Select set of words.
3. the semantic matching method according to claim 2 that can recognize that natural language description based on computer, feature are existed
In,
Step S21) in, the determiner of object language setting is switched to using prefix as mark for statement of requirements；
Step S22) in, auxiliary word, preposition, conjunction class are prestored as stop words as stop words dictionary；
Step S23) in, synonym extension is carried out to each word being applicable in set of words according to synonym dictionary；
Step S24) in, root retrieving algorithm is Porter algorithms or Lucene algorithms.
4. the semantic matching method according to claim 1 that can recognize that natural language description based on computer, feature are existed
In step S4) including step：
Step S41)：The word of spare set of words is traversed, there are the words of intersection with candidate set of words for screening；
Step S42)：To meeting the word of intersection, matching degree is calculated.
5. the semantic matching method according to claim 4 that can recognize that natural language description based on computer, feature are existed
In step S4) in, the formula of the matching degree of candidate set of words and spare set of words is：
Wherein, count is the word number of the semantic similarity found, | wordsetA| to segment number in requirement description sentence,
|wordsetB| for the participle number in message name/operation name.
6. a kind of semantic matches system that can recognize that natural language description based on computer, which is characterized in that including constraints module,
Candidate set of words forms module, spare set of words forms module and matching module, wherein：
The constraints module, for the logic and step that are limited with the syntax rule of object language as reference, by natural language
Requirement description is constrained to the structure for having the step of logic；
Candidate's set of words forms module, for fixed clause in the natural language requirement description after constraint, obtaining
Candidate set of words including the root in natural language requirement description；
The spare set of words forms module, for being segmented to the message name in object language/operation name, including
The spare set of words of root in message name/operation name；
The matching module, for calculating the matching degree of candidate set of words and spare set of words.
7. the semantic matches system according to claim 6 that can recognize that natural language description based on computer, feature are existed
In it is single including primary set of words unit, applicable set of words unit, synonym extension that candidate's set of words forms module
Member and root reduction unit, wherein：
The primary set of words unit, for the determiner according to setting, obtains the statement of requirements of natural language description, need to
Sentence is asked to form primary set of words after segmenting；
The applicable set of words unit for removing the stop words in primary set of words, forms and is applicable in set of words；
The synonym expanding element, for carrying out synonym extension to each word being applicable in set of words；
The root reduction unit, for carrying out root reduction to extension set of words, acquisition includes natural language requirement description
In root candidate set of words.
8. the semantic matches system according to claim 7 that can recognize that natural language description based on computer, feature are existed
In,
In the primary set of words unit, the determiner of object language setting is switched to using prefix as mark for statement of requirements
Know；
In the applicable set of words unit, auxiliary word, preposition, conjunction class are prestored as stop words as stop words dictionary；
In the synonym expanding element, synonym expansion is carried out to each word being applicable in set of words according to synonym dictionary
Exhibition；
In the root reduction unit, root retrieving algorithm is Porter algorithms or Lucene algorithms.
9. the semantic matches system according to claim 6 that can recognize that natural language description based on computer, feature are existed
In, the matching module includes asking presentate member, matching unit, wherein：
Described to ask presentate first, for traversing the word of spare set of words, there are the words of intersection with candidate set of words for screening；
The matching unit to meeting the word of intersection, calculates matching degree.
10. the semantic matches system according to claim 9 that can recognize that natural language description based on computer, feature are existed
In in the matching unit, the formula of the matching degree of candidate set of words and spare set of words is：
Wherein, count is the word number of the semantic similarity found, | wordsetA| to segment number in requirement description sentence,
|wordsetB| for the participle number in message name/operation name.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711460123.3A CN108170679B (en) | 2017-12-28 | 2017-12-28 | Semantic matching method and system based on computer recognizable natural language description |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711460123.3A CN108170679B (en) | 2017-12-28 | 2017-12-28 | Semantic matching method and system based on computer recognizable natural language description |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108170679A true CN108170679A (en) | 2018-06-15 |
| CN108170679B CN108170679B (en) | 2021-09-03 |
Family
ID=62519156
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711460123.3A Active CN108170679B (en) | 2017-12-28 | 2017-12-28 | Semantic matching method and system based on computer recognizable natural language description |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108170679B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110413267A (en) * | 2019-08-08 | 2019-11-05 | 四川爱创科技有限公司 | Self-adapted service flow modeling method based on business rule |
| CN114238619A (en) * | 2022-02-23 | 2022-03-25 | 成都数联云算科技有限公司 | Method, system, device and medium for screening Chinese nouns based on edit distance |
Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1417707A (en) * | 2002-12-02 | 2003-05-14 | 刘莎 | Natural language semantic information united-coding method |
| US20070130561A1 (en) * | 2005-12-01 | 2007-06-07 | Siddaramappa Nagaraja N | Automated relationship traceability between software design artifacts |
| CN101595474A (en) * | 2007-01-04 | 2009-12-02 | 思解私人有限公司 | Language analysis |
| CN103186574A (en) * | 2011-12-29 | 2013-07-03 | 北京百度网讯科技有限公司 | Method and device for generating searching result |
| CN103309852A (en) * | 2013-06-14 | 2013-09-18 | 瑞达信息安全产业股份有限公司 | Method for discovering compound words in specific field based on statistics and rules |
| CN103699667A (en) * | 2013-12-24 | 2014-04-02 | 天津大学 | Web service multi-dimensional semantic model building 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 |
| CN104133812A (en) * | 2014-07-17 | 2014-11-05 | 北京信息科技大学 | User-query-intention-oriented Chinese sentence similarity hierarchical calculation method and user-query-intention-oriented Chinese sentence similarity hierarchical calculation device |
| CN105930452A (en) * | 2016-04-21 | 2016-09-07 | 北京紫平方信息技术股份有限公司 | Smart answering method capable of identifying natural language |
| CN106055537A (en) * | 2016-05-23 | 2016-10-26 | 王立山 | Natural language machine recognition method and system |
| CN106339366A (en) * | 2016-08-08 | 2017-01-18 | 北京百度网讯科技有限公司 | Method and device for requirement identification based on artificial intelligence (AI) |
| CN106372117A (en) * | 2016-08-23 | 2017-02-01 | 电子科技大学 | Word co-occurrence-based text classification method and apparatus |
| CN106407196A (en) * | 2015-07-29 | 2017-02-15 | 成都诺铱科技有限公司 | Semantic analysis intelligent instruction robot applied to logistics management software |
| CN106776532A (en) * | 2015-11-25 | 2017-05-31 | 中国移动通信集团公司 | A kind of knowledge question answering method and device |
| US20170270107A1 (en) * | 2005-10-26 | 2017-09-21 | Cortica, Ltd. | System and method for signature-enhanced multimedia content searching |
| CN107391111A (en) * | 2017-06-22 | 2017-11-24 | 刘武丰 | Artificial intelligence co-development framework and implementation method |
| CN107463683A (en) * | 2017-08-09 | 2017-12-12 | 上海壹账通金融科技有限公司 | The naming method and terminal device of code element |
-
2017
- 2017-12-28 CN CN201711460123.3A patent/CN108170679B/en active Active
Patent Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1417707A (en) * | 2002-12-02 | 2003-05-14 | 刘莎 | Natural language semantic information united-coding method |
| US20170270107A1 (en) * | 2005-10-26 | 2017-09-21 | Cortica, Ltd. | System and method for signature-enhanced multimedia content searching |
| US20070130561A1 (en) * | 2005-12-01 | 2007-06-07 | Siddaramappa Nagaraja N | Automated relationship traceability between software design artifacts |
| CN101595474A (en) * | 2007-01-04 | 2009-12-02 | 思解私人有限公司 | Language analysis |
| CN103186574A (en) * | 2011-12-29 | 2013-07-03 | 北京百度网讯科技有限公司 | Method and device for generating searching result |
| CN103309852A (en) * | 2013-06-14 | 2013-09-18 | 瑞达信息安全产业股份有限公司 | Method for discovering compound words in specific field based on statistics and rules |
| CN103699667A (en) * | 2013-12-24 | 2014-04-02 | 天津大学 | Web service multi-dimensional semantic model building 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 |
| CN104133812A (en) * | 2014-07-17 | 2014-11-05 | 北京信息科技大学 | User-query-intention-oriented Chinese sentence similarity hierarchical calculation method and user-query-intention-oriented Chinese sentence similarity hierarchical calculation device |
| CN106407196A (en) * | 2015-07-29 | 2017-02-15 | 成都诺铱科技有限公司 | Semantic analysis intelligent instruction robot applied to logistics management software |
| CN106776532A (en) * | 2015-11-25 | 2017-05-31 | 中国移动通信集团公司 | A kind of knowledge question answering method and device |
| CN105930452A (en) * | 2016-04-21 | 2016-09-07 | 北京紫平方信息技术股份有限公司 | Smart answering method capable of identifying natural language |
| CN106055537A (en) * | 2016-05-23 | 2016-10-26 | 王立山 | Natural language machine recognition method and system |
| CN106339366A (en) * | 2016-08-08 | 2017-01-18 | 北京百度网讯科技有限公司 | Method and device for requirement identification based on artificial intelligence (AI) |
| CN106372117A (en) * | 2016-08-23 | 2017-02-01 | 电子科技大学 | Word co-occurrence-based text classification method and apparatus |
| CN107391111A (en) * | 2017-06-22 | 2017-11-24 | 刘武丰 | Artificial intelligence co-development framework and implementation method |
| CN107463683A (en) * | 2017-08-09 | 2017-12-12 | 上海壹账通金融科技有限公司 | The naming method and terminal device of code element |
Non-Patent Citations (1)
| Title |
|---|
| 欧阳柳波等: "基于领域需求结构化描述的自动分析建模方法", 《计算机工程与应用》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110413267A (en) * | 2019-08-08 | 2019-11-05 | 四川爱创科技有限公司 | Self-adapted service flow modeling method based on business rule |
| CN110413267B (en) * | 2019-08-08 | 2023-05-26 | 四川爱创科技有限公司 | Self-adaptive business process modeling method based on business rules |
| CN114238619A (en) * | 2022-02-23 | 2022-03-25 | 成都数联云算科技有限公司 | Method, system, device and medium for screening Chinese nouns based on edit distance |
| CN114238619B (en) * | 2022-02-23 | 2022-04-29 | 成都数联云算科技有限公司 | Method, system, device and medium for screening Chinese nouns based on edit distance |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108170679B (en) | 2021-09-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3660733B1 (en) | Method and system for information extraction from document images using conversational interface and database querying | |
| US20200159755A1 (en) | Summary generating apparatus, summary generating method and computer program | |
| EP3153978B1 (en) | Address search method and device | |
| CN107729322A (en) | Segmenting method and device, establish sentence vector generation model method and device | |
| CN104866517A (en) | Method and device for capturing webpage content | |
| CN108170679A (en) | It can recognize that the semantic matching method and system of natural language description based on computer | |
| CN110633360B (en) | Semantic matching method and related device | |
| CN113157959A (en) | Cross-modal retrieval method, device and system based on multi-modal theme supplement | |
| JP2018025956A (en) | Model creation device, estimation device, method, and program | |
| CN105373527B (en) | Omission recovery method and question-answering system | |
| CN115688753A (en) | Knowledge injection method and interaction system of Chinese pre-training language model | |
| KR102182619B1 (en) | Knowledge extraction system using frame based on ontology | |
| CN107180024A (en) | A kind of multi-source heterogeneous data entity recognition methods of center connected subgraph and system | |
| JP6782944B2 (en) | Information processing equipment, information processing methods, and programs | |
| CN114579605B (en) | Table question-answer data processing method, electronic equipment and computer storage medium | |
| CN116011429A (en) | Emotion triplet extraction method and system based on graph neural network | |
| CN114416941B (en) | Knowledge graph-fused dialogue knowledge point determination model generation method and device | |
| CN114970666A (en) | Spoken language processing method and device, electronic equipment and storage medium | |
| JP5911931B2 (en) | Predicate term structure extraction device, method, program, and computer-readable recording medium | |
| CN115269767A (en) | Model training method, device and storage medium | |
| Jing et al. | Developing a research ideas creation system through reusing knowledge bases for ontology construction | |
| Kaur | Khushleen@ IECSIL-FIRE-2018: Indic Language Named Entity Recognition using Bidirectional LSTMs with Subword Information. | |
| CN112597748A (en) | Corpus generation method, apparatus, device and computer readable storage medium | |
| US20200202079A1 (en) | Text generation method, text generation device, and learning-completed model | |
| JP2006323624A (en) | Method, program, and device for calculating similarity in sequence diagram |
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 |