CN108717405A - The complementing method of the default subject of staircase design specification based on mind map - Google Patents
The complementing method of the default subject of staircase design specification based on mind map Download PDFInfo
- Publication number
- CN108717405A CN108717405A CN201810349079.7A CN201810349079A CN108717405A CN 108717405 A CN108717405 A CN 108717405A CN 201810349079 A CN201810349079 A CN 201810349079A CN 108717405 A CN108717405 A CN 108717405A
- Authority
- CN
- China
- Prior art keywords
- sentence
- subject
- mind map
- staircase design
- word
- 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
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/211—Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/221—Parsing markup language streams
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Machine Translation (AREA)
Abstract
The complementing method of the present invention provides a kind of default subject of staircase design specification based on mind map.Pass through the inheritance between ontology in mind map, it can determine the subject of the incomplete subordinate clause missing of syntactic structure in complex sentence and carry out polishing, and subject-predicate-object in pivotal sentence and even meaning sentence can be extracted, can preferably to carry out natural language processing, the knowledge mapping of final structure stair construction standards, realizes automatic check of drawings；The present invention has effectively evaded the uncertain factor being likely to occur during check of drawings, and false drop rate is low, easy to operate, saves manpower, greatly improves the completion efficiency of building trade engineering project.
Description
Technical field
The invention belongs to Computer Natural Language Processing technology technical fields, and in particular to the stair based on mind map are set
Count the complementing method of the default subject of specification.
Background technology
With the development of computer science and technology, each engineering field has introduced the concept of knowledge mapping, knowledge mapping energy
Enough descriptions show the contact between the entity and entity in the world, provide a kind of ability for going problem analysis from the angle of relationship,
It can realize building together, inquire, share and reusing to knowledge.At present using knowledge mapping industry field include medicine, military affairs,
Education etc..But the informatization of building trade is still in infancy, the engineering project of a building trade firstly the need of
Designing institute is planned, is designed, and is then examined designed model and drawing, is provided to unit in charge of construction and is applied
Work finally comes into operation.Accurate drawing and model can be reduced to a great extent the construction stage and do over again caused by design alteration
Work holdup phenomenon, but due to technical shortage at this stage, in China most project be all after design by expert come into
Capable artificially check of drawings, the design specification of designed drawing and building trade is compared, this will produce false drop rate height, missing inspection
Rate is high, examines the problems such as dynamics is low, uncertain factor is more.
Invention content
The complementing method of the object of the present invention is to provide a kind of default subject of staircase design specification based on mind map, solution
The problem of division statement subject is not difficult to entirely in staircase design of having determined specification.
A kind of complementing method of the default subject of staircase design specification based on mind map of the present invention, by specification Chinese
This progress polishing can preferably carry out natural language processing, and the final knowledge mapping for building stair construction standards is realized automatic
Check of drawings, help solve the problems, such as that false drop rate is higher when artificial check of drawings existing in the prior art.
The technical solution adopted in the present invention is the completion side of the default subject of staircase design specification based on mind map
Method includes the following steps：
Step 1：Corpus of the specification in relation to staircase design as processing is obtained from house building design standard, and
Using the Forward Maximum Method algorithm based on dictionary to urtext segment and Statistics-Based Method to participle after
Word carry out part-of-speech tagging, obtain pretreated text.
Step 2：It, will be between the relevant ontology of staircase design and ontology with reference to the descriptor format of stair specification in IFC standards
Relationship is combed into mind map, and builds corresponding index tree.
Step 3：Pretreated text is carried out syntax parsing using context-free grammar, is encountering pivotal sentence or company
The case where calling sentence determines the sentence pattern ingredient in sentence with the relevant ontology of object by being searched in index tree, language is built with this
Method tree, and whether the syntactic structure for analyzing a sentence is complete.The last language material for therefrom filtering out missing subject.
Step 4：By being scanned in index tree, his father's knot is searched to the object ontology with imperfect sentence pattern structure
Point and to root node unique paths, father node is this default subject, in addition to father node, on this path
All nodes be subject modification attribute, after default subject is added to prototype statement, export the complete stair of Subject, Predicate and Object
Design specification.
In step 1：Method is the Forward Maximum Method algorithm based on dictionary used by participle.Part-of-speech tagging using
Part-of-speech tagging method based on hidden Markov model.
In step 3：The position that occurs according to the part of speech of word and in sentence when carrying out syntax parsing determines it
Ingredient in sentence.
The method that syntax tree is built in step 3 is as follows：Define first context-free grammar G=N, ∑, X,
S}.Wherein N indicates the mark of one group of n omicronn-leaf child node；Σ indicates the mark of one group of leafy node, that is, forms the word of sentence；X tables
Show the rule of one group of syntax, the as production of N, can be expressed as X=Y per rules and regulations1Y2...Yn, X ∈ N, Yi∈(N∪Σ)；X
In at least one production α be served as by S.And S indicates the mark that syntax tree starts.
Using bottom-up method, from character start of string to be analyzed, matching context is removed with character string to be analyzed
The right character of Grammars rule X arrows replaces with left character after successful match, and until S occurs, syntax tree has been built
Finish.
In step 4：Using ergodic algorithm when searching father node.
The beneficial effects of the invention are as follows：
A kind of complementing method of the default subject of staircase design specification based on mind map.By between ontology in mind map
Inheritance, it may be determined that the subject of syntactic structure incomplete subordinate clause missing and carry out polishing in complex sentence, and can carry
Subject-predicate-object in pivotal sentence and even meaning sentence is taken, it is final to help to facilitate natural language processing and build knowledge mapping
Automatic check of drawings；The present invention has effectively evaded the uncertain factor being likely to occur during artificial check of drawings, and false drop rate is low, operation
Simply, manpower is saved, the completion efficiency of building trade engineering project is greatly improved.
Description of the drawings
Fig. 1 is the main flow of the complementing method of the default subject of staircase design specification the present invention is based on mind map；
Fig. 2 is the mind map model defined with reference to IFC standards.
Specific implementation mode
The following describes the present invention in detail with reference to the accompanying drawings and specific embodiments.
The present invention is based on the complementing methods of the default subject of staircase design specification of mind map, include the following steps：
Step 1：Corpus of the specification in relation to staircase design as processing is obtained from house building design standard, and
Urtext is segmented using the Forward Maximum Method algorithm based on dictionary, and the method based on hidden Markov model
Part-of-speech tagging is carried out to the word string after participle, obtains pretreated text.
Step 2：It, will be between the relevant ontology of staircase design and ontology with reference to the descriptor format of stair specification in IFC standards
Relationship is combed into mind map, and builds corresponding index tree.
Step 3：Pretreated text is carried out syntax parsing using context-free grammar, is encountering pivotal sentence or company
The case where calling sentence determines the sentence pattern ingredient in sentence with the relevant ontology of object by being searched in index tree, language is built with this
Method tree, and whether the syntactic structure for analyzing a sentence is complete.The last language material for therefrom filtering out missing subject.
Step 4：By being scanned in index tree, his father's knot is searched to the object ontology with imperfect sentence pattern structure
Point and to root node unique paths, father node is this default subject, in addition to father node, on this path
All nodes be subject modification attribute, after default subject is added to prototype statement, export the complete stair of Subject, Predicate and Object
Design specification.
In step 1：Method is the Forward Maximum Method algorithm based on dictionary used by participle.Part-of-speech tagging using
Part-of-speech tagging method based on hidden Markov model.
In step 3：The position that occurs according to the part of speech of word and in sentence when carrying out syntax parsing determines it
Ingredient in sentence.
The method that syntax tree is built in step 3 is as follows：Define first context-free grammar G=N, ∑, X,
S}.Wherein N indicates the mark of one group of n omicronn-leaf child node；Σ indicates the mark of one group of leafy node, that is, forms the word of sentence；X tables
Show the rule of one group of syntax, the as production of N, can be expressed as X=Y per rules and regulations1Y2...Yn, X ∈ N, Yi∈(N∪Σ)；X
In at least one production α be served as by S.And S indicates the mark that syntax tree starts.
Using bottom-up method, from character start of string to be analyzed, matching context is removed with character string to be analyzed
The right character of Grammars rule X arrows replaces with left character after successful match, and until S occurs, syntax tree has been built
Finish.
In step 4：Using ergodic algorithm when searching father node.
Since the text of Chinese is made of word, and without apparent boundary sign between word and word, so processing text
The first step be input language material segmented, this patent use segmenting method be the Forward Maximum Method based on dictionary
Algorithm：
A character string s is taken out first from text to be slit1, and the ontology in stair specification is formed into dictionary, and
Build Hash table.The long maxlen of most major term defined in dictionary, from character string s1The left side start, take out length be not more than
The substring w of maxlen.And：Search for whether w is a word in Hash table, if being a word by w outputs.If not one
Word then subtracts a word from by the tail portion of w, and continuation iteratively judges w whether in Hash table, until w is empty or s1For sky.
In order to determine the syntactic structure of sentence, sentence pattern ingredient must be just determined first, and the sentence pattern ingredient in Chinese is by this word
Part of speech and its ordinal position occurred in sentence determine what ingredient it makees in sentence.Part-of-speech tagging is using based on hidden
The method of Markov model：
The method is divided into three modules：Initialise, Induction, Back tracing the best
tagging。
Each part of speech is counted in Initialise steps first and appears in the probability of language material text beginning of the sentence, and is multiplied by part of speech
The probability for ejecting word obtains the score scores of a word.Then it is calculated with viterbi algorithms in Induction steps every
The score scores of the word of two adjacent appearance, the initial score scores equal to this part of speech are multiplied by the probability converted between part of speech and multiply
The probability of this word is ejected into part of speech.The part of speech value for selecting fractional value big from final score scores is recorded in
In Backpointer.Backtracking from back to front is finally carried out in step Back tracing the best tagging, is obtained
The sequence string str constituted to part of speech2。
It only not can determine that whether the sentence pattern structure of this sentence is complete with part of speech, also need by building syntax tree come into one
Step judges.The method of structure syntax tree is as follows：
Context-free grammar shown in being defined as follows first：
(a) N indicates the mark, such as original sentence, noun phrase, verb phrase etc. of one group of n omicronn-leaf child node.
(b) Σ indicates the mark of one group of leafy node, that is, forms the word of sentence.
(c) X indicates the rule of one group of syntax, the as production of N, can be expressed as X=Y per rules and regulations1Y2...Yn, X ∈
N, Yi∈(N∪Σ)。
(d) S indicates the mark that syntax tree starts.
Its four-tuple is defined by specific sentence of context-free grammar pair, can be obtained by bottom-up derivation
To a syntax tree.If encountering pivotal sentence or even meaning sentence, there is two or more noun before predicate, can not judge
When subject ingredient, need to determine the ontology that there is inheritance with object by mind map, whom to be determined in these nouns
It is subject and other nouns only make attribute, modification defines the range of subject.It can readily judge sentence by syntax tree
The incomplete corpus of method structure, is extracted.
After the complex sentence for extracting missing subject, need to find out object from clause, and by object in mind map
Location determination has the father node of inheritance therewith.To the subject of its missing of polishing, which is input object ontology, output
There is the subject of direct inheritance as all nodes on its subject and subject to the exclusive path of root node with this object
The attribute phrase of composition, for modifying the field for limiting subject.
The attached flow shown in FIG. 1 for entire completion subject.While participle, label part of speech, it will be lived with reference to IFC standards
The relationship of ontology is built into mind map model in the specification of residence stair, language material is built syntax tree by this model, and therefrom
The incomplete corpus of syntactic structure is filtered out, inheritance of the ontology of object component in mind map is made by inquiry,
Its father node is found to decide language, and its to the exclusive path of root node be action scope that modification limits this specification.
According to the algorithm of the Forward Maximum Method based on dictionary, first from《Code for design of dwelling houses》In win and advise 4.1.2
As language material sample：" residence stairs bench span width is no less than 1.1m, when being equipped with railing on one side, no less than 1m.".
A character string to be slit is taken to be segmented from original text, at this time s1=" residence stairs bench span width is not answered small
In 1.1m, when being equipped with railing on one side, no less than 1m." according to the dictionary of construction determine maxlen be=10, s2It is initialized as
It is empty.By dictionary construction at Hash table.
From s1The left side choose length be not more than maxlen substring w=" residence stairs bench span width is not ", judge that w is
It is no for sky, be not sky, judge whether w is a word in Hash table, traverses Hash table, does not find occurrence.It will subtract on the right of w
A few word, w=" residence stairs bench span width " continue iteration.Until w is reduced at " house ", searched in Hash table at
House is added to s by work(2In, s2=" house/".s1=" stair bench span width is no less than 1.1m, is equipped with railing on one side
When, no less than 1m." iteration is to s1For sky when, export s at this time2=" house/stair/bench/span width/do not answer/is less than/
1.1m/ ,/on one side/be equipped with/railing/when/,/not answering/is less than/1m/."
The part of speech of tagged words after participle marks the process of part of speech as shown in algorithm 2.A part of language material of handmarking first
Then the part of speech of collection is trained parameter with viterbi algorithms, marked automatically to remaining corpus by machine learning
Note, wherein unregistered word is smoothed, the unregistered word after mark is smoothed, correct language material is added
It is added to and continues to train relatively reliable parameter in training set.The language material finally exported is { noun:House } { noun:Stair }
{noun:Bench } { noun:Span width } { adv:Do not answer { v:Less than { num:1.1m }, { adv:{ v on one side }:Equipped with { noun:
Railing } { adv:When, { adv:Do not answer { v:Less than { num:1m}.
Relationship between ontology in the residence stairs design specification defined according to IFC standards, the thinking defined with reference to this format
It is as shown in Fig. 2 to lead graph model.
Syntax tree is built according to the language material of pretreatment output, wherein S indicates sentence；NP, VP, PP are noun, verb, preposition
Phrase (phrase rank)；A, V, P are adverbial word, verb, preposition respectively；Noun is noun, and Num is number.Define context-free
Grammer production is as follows：
1)S→NPVP
2)NP→NP|NounNoun|ε
3)VP→AVN|AVNA
4)A→Adv
5)V→Verb
6)N→Noun|Num
7) Noun → house | stair | bench | span width | railing
8) Adv → do not answer
9)Num→1.1m|1m
10) Verb → be less than | on one side | when | it is equipped with
Production X：The definition of α → β needs to meet following condition：
1) α can be the arbitrary mark of leafy node and n omicronn-leaf child node, cannot be ε；
2) β can be the arbitrary mark of leafy node and n omicronn-leaf child node, can be ε；
3) α of at least one production must be served as by S in X.
It is gone from character start of string to be analyzed with character string to be analyzed using methods bottom-up, based on stipulations
The right character of context-free grammar rule arrow is matched, left character is replaced with after successful match, until S.
The syntax tree of structure is denoted as following format：
A) [[[[Adv is or not A by VP by S [NP [NP [Noun houses] [Noun stair]] [NP [Noun bench] [Noun span widths]]]
Answer]] [V [Verb is less than]] [N [Num 1.1m]]]]
B) [S [NP [ε] [VP [A [Adv is on one side]] [V [Verb is equipped with]] [N [Noun railings]] [A [when Adv]]]]
C) [S [NP [ε]] [VP [A [Adv is not answered]] [V [Verb is less than]] [N [Num 1m]]]]
Syntactic structure is analyzed, by taking second clause as an example, centered on predicate, the analysis subject before predicate and the adverbial modifier,
Participle object after predicate and complement can determine whether that adverbial word " one side " makees the adverbial modifier, verb " being equipped with " makees predicate, name by analysis
Word " railing " makees object, last adverbial word " when ", it represents this clause and makees the adverbial modifier in entire sentence, but this clause is not
Completely, if it is not from the context, it can not judge to be what " being equipped with railing ".
It was found that the incomplete sentence of syntactic structure has s1=" when being equipped with railing on one side ", s2=" being no less than 1m ".
The process of subject is searched with s1For, node " railing " is searched first in B index trees, is then searched this node and is arrived
One paths of root node：The father node for finding railing is " bench ", and the father node of " bench " is " stair ", and the father of " stair " ties
Point is " house ", and the father node of " house " is " building ".That is the subject of this clause missing is bench, by s2Supplement is completely " ladder
When being equipped with railing on one side of section ", the path to root node is the attribute for modifying subject, i.e.,：" stair of the house of building "
" when bench is equipped with railing on one side ".Since attribute does not do ingredient in sentence pattern structure, therefore can be without polishing.
In addition to the subject of completion missing can determine subject in the indefinite specification of sentence pattern ingredient using the method.Example
As first clause " residence stairs bench span width is no less than 1.1m " of language material sample has found simultaneously when building syntax tree
Subject is not lacked, but extracts triple from this specification and is not so good as subject-predicate sentence and be so easy, because not doing semantic analysis
Under the premise of, can not judge it is what is no less than 1.1m actually, be not easy to semantic analysis.So analyzing this sentence
Syntactic structure when obtain first determine subject.In mind map determine house, stair, bench which include span width this
Attribute can determine it is that bench does subject ingredient by traversing index tree, and residence stairs make attribute, and modification limits the range of subject.
Finally by clause, all supplement is completely：" span width of the bench of the stair of house is no less than 1.1m, bench
When being equipped with railing on one side, the span width of bench is no less than 1m.".
The successful completion staircase design specification text of absence of subject according to the method for the present invention.The purpose of the present invention is
A kind of complementing method of the default subject of staircase design specification based on mind map is provided, by being mended to text in specification
Together, natural language processing can be preferably carried out, the final knowledge mapping for building stair construction standards is realized automatic check of drawings, had
Effect has evaded the uncertain factor being likely to occur during artificial check of drawings, and false drop rate is low, easy to operate, labor-saving
Meanwhile greatly improving the completion efficiency of building trade engineering project.
Claims (5)
1. the complementing method of the default subject of staircase design specification based on mind map, which is characterized in that include the following steps：
Step 1：Corpus of the specification in relation to staircase design as processing is obtained from house building design standard, and is used
Forward Maximum Method algorithm based on dictionary segments urtext, and based on the method for hidden Markov model to dividing
Word after word carries out part-of-speech tagging, obtains pretreated text；
Step 2：With reference to the descriptor format of stair specification in IFC standards, will between the relevant ontology of staircase design and ontology relationship
It is combed into mind map, and builds corresponding index tree；
Step 3：Pretreated text is carried out syntax parsing using context-free grammar, is encountering pivotal sentence or even meaning sentence
The case where, the sentence pattern ingredient in sentence is determined with the relevant ontology of object, syntax tree is built with this by being searched in index tree,
And whether the syntactic structure for analyzing a sentence is complete.The last language material for therefrom filtering out missing subject；
Step 4：By being scanned in index tree, to the object ontology with imperfect sentence pattern structure search its father node with
And unique paths to root node, father node is this default subject, in addition to father node, the institute on this path
It is the modification attribute of subject to have node, after default subject is added to prototype statement, export the complete staircase design of Subject, Predicate and Object
Specification.
2. the complementing method of the staircase design specification default subject according to claim 1 based on mind map, feature
It is, in the step 1：Method is the Forward Maximum Method algorithm based on dictionary used by participle.What part-of-speech tagging used
Method is the part-of-speech tagging algorithm based on hidden Markov model.
3. the complementing method of the staircase design specification default subject according to claim 1 based on mind map, feature
It is, in the step 3：In use hereafter after Grammars progress syntax parsing according to the part of speech of word and in sentence
The position of appearance determines the ingredient in sentence.
4. the complementing method of the staircase design specification default subject according to claim 1 based on mind map, feature
It is, the method that syntax tree is built in the step 3 is as follows：Define first context-free grammar G=N, ∑,
X, S }, wherein N indicates the mark of one group of n omicronn-leaf child node；Σ indicates the mark of one group of leafy node, that is, forms the word of sentence；X
It indicates the rule of one group of syntax, the as production of N, can be expressed as X=Y per rules and regulations1Y2...Yn, X ∈ N, Yi∈(N∪
Σ)；The α of at least one production is obtained and is served as by S in X.And S indicates the mark that syntax tree starts；
Using bottom-up method, from character start of string to be analyzed, matching context-free is gone with character string to be analyzed
The right character of grammar rule X arrows replaces with left character after successful match, and until S occurs, syntax tree structure finishes.
5. the complementing method of the staircase design specification default subject according to claim 1 based on mind map, feature
It is, in the step 4：Using ergodic algorithm when searching father node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810349079.7A CN108717405B (en) | 2018-04-18 | 2018-04-18 | The complementing method of the default subject of staircase design specification based on mind map |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810349079.7A CN108717405B (en) | 2018-04-18 | 2018-04-18 | The complementing method of the default subject of staircase design specification based on mind map |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108717405A true CN108717405A (en) | 2018-10-30 |
| CN108717405B CN108717405B (en) | 2019-08-16 |
Family
ID=63899153
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810349079.7A Active CN108717405B (en) | 2018-04-18 | 2018-04-18 | The complementing method of the default subject of staircase design specification based on mind map |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108717405B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110287304A (en) * | 2019-06-30 | 2019-09-27 | 联想(北京)有限公司 | Question and answer information processing method, device and computer equipment |
| CN111027287A (en) * | 2019-12-24 | 2020-04-17 | 深圳集智数字科技有限公司 | Method for converting computer executable script and related device |
| CN111091006A (en) * | 2019-12-20 | 2020-05-01 | 北京百度网讯科技有限公司 | Entity intention system establishing method, device, equipment and medium |
| CN111708882A (en) * | 2020-05-29 | 2020-09-25 | 西安理工大学 | Transformer-based Chinese text information missing completion method |
| CN113987199A (en) * | 2021-10-19 | 2022-01-28 | 清华大学 | BIM intelligent image examination method, system and medium with standard automatic interpretation |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140324925A1 (en) * | 2013-04-24 | 2014-10-30 | Dell Products, Lp | Efficient Rename in a Lock-Coupled Traversal of B+Tree |
| CN106021229A (en) * | 2016-05-19 | 2016-10-12 | 苏州大学 | Chinese event co-reference resolution method and system |
-
2018
- 2018-04-18 CN CN201810349079.7A patent/CN108717405B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140324925A1 (en) * | 2013-04-24 | 2014-10-30 | Dell Products, Lp | Efficient Rename in a Lock-Coupled Traversal of B+Tree |
| CN106021229A (en) * | 2016-05-19 | 2016-10-12 | 苏州大学 | Chinese event co-reference resolution method and system |
Non-Patent Citations (2)
| Title |
|---|
| ENTESAR ALMOSALLAM ET AL.: "ITQAN: a Mobile Based Assistant for Mastering Quran Memorization", 《2015 FIFTH INTERNATIONAL CONFERENCE ON E-LEARNING》 * |
| 杨国基 等: "自然语言处理中基于短语结构的语法分析方法", 《微处理机》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110287304A (en) * | 2019-06-30 | 2019-09-27 | 联想(北京)有限公司 | Question and answer information processing method, device and computer equipment |
| CN111091006A (en) * | 2019-12-20 | 2020-05-01 | 北京百度网讯科技有限公司 | Entity intention system establishing method, device, equipment and medium |
| CN111091006B (en) * | 2019-12-20 | 2023-08-29 | 北京百度网讯科技有限公司 | Method, device, equipment and medium for establishing entity intention system |
| CN111027287A (en) * | 2019-12-24 | 2020-04-17 | 深圳集智数字科技有限公司 | Method for converting computer executable script and related device |
| CN111027287B (en) * | 2019-12-24 | 2023-08-29 | 深圳集智数字科技有限公司 | Method and related device for converting computer executable script |
| CN111708882A (en) * | 2020-05-29 | 2020-09-25 | 西安理工大学 | Transformer-based Chinese text information missing completion method |
| CN111708882B (en) * | 2020-05-29 | 2022-09-30 | 西安理工大学 | Transformer-based Chinese text information missing completion method |
| CN113987199A (en) * | 2021-10-19 | 2022-01-28 | 清华大学 | BIM intelligent image examination method, system and medium with standard automatic interpretation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108717405B (en) | 2019-08-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108717405B (en) | The complementing method of the default subject of staircase design specification based on mind map | |
| CN110413732B (en) | Knowledge searching method for software defect knowledge | |
| CN111931506B (en) | Entity relationship extraction method based on graph information enhancement | |
| CN108681574B (en) | Text abstract-based non-fact question-answer selection method and system | |
| CN105095204B (en) | The acquisition methods and device of synonym | |
| Shindo et al. | Bayesian symbol-refined tree substitution grammars for syntactic parsing | |
| CN110413319B (en) | Code function taste detection method based on deep semantics | |
| Bar-Haim et al. | Semantic inference at the lexical-syntactic level | |
| CN105869634A (en) | Field-based method and system for feeding back text error correction after speech recognition | |
| CN107832289A (en) | A kind of name entity recognition method based on LSTM CNN | |
| CN109614620B (en) | HowNet-based graph model word sense disambiguation method and system | |
| CN111339269A (en) | Knowledge graph question-answer training and application service system with automatically generated template | |
| CN110991180A (en) | Command identification method based on keywords and Word2Vec | |
| CN106446147A (en) | Emotion analysis method based on structuring features | |
| CN113312922B (en) | Improved chapter-level triple information extraction method | |
| CN106610952A (en) | Mixed text feature word extraction method | |
| CN110399496A (en) | A kind of knowledge mapping construction method based on CR decision tree | |
| CN106610949A (en) | Text feature extraction method based on semantic analysis | |
| CN108763355A (en) | A kind of intelligent robot interaction data processing system and method based on user | |
| CN113792542A (en) | Intention understanding method fusing syntactic analysis and semantic role pruning | |
| CN109992777B (en) | Keyword-based traditional Chinese medicine disease condition text key semantic information extraction method | |
| CN112632272A (en) | Microblog emotion classification method and system based on syntactic analysis | |
| CN107423281A (en) | The algorithm of improved Small World Model extraction text feature | |
| Pettersson et al. | Parsing the past-identification of verb constructions in historical text | |
| Bladier et al. | German and French neural supertagging experiments for LTAG parsing |
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 |