CN109241538A - Based on the interdependent Chinese entity relation extraction method of keyword and verb - Google Patents
Based on the interdependent Chinese entity relation extraction method of keyword and verb Download PDFInfo
- Publication number
- CN109241538A CN109241538A CN201811124153.1A CN201811124153A CN109241538A CN 109241538 A CN109241538 A CN 109241538A CN 201811124153 A CN201811124153 A CN 201811124153A CN 109241538 A CN109241538 A CN 109241538A
- Authority
- CN
- China
- Prior art keywords
- verb
- entity
- relationship
- subject
- sentence
- 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
- G06F40/279—Recognition of textual entities
Abstract
The invention discloses a kind of Chinese entity relation extraction methods interdependent based on keyword and verb first to segment text, extracting keywords using extensive unstructured free text as target text, forms text key word dictionary;Then subordinate sentence, participle, part-of-speech tagging, name Entity recognition, interdependent syntactic analysis processing are carried out to text, constructs entity corpus in conjunction with name entity dictionary and keyword thesaurus;The dependence between feature, syntactic structure and word is constituted according to Chinese sentence and constructs entity relationship syntactic rule from verb, then the matching of relationship syntactic rule is carried out to sentence each in text;Last output relation triple, obtains text relationship triplet sets.The present invention can make the entity relation extraction of extensive Chinese text more efficient, more accurate.
Description
Technical field
The present invention relates to Chinese entity relation extraction methods more particularly to a kind of based on keyword and verb dependency analysis
Extensive free text abstracting method.
Background technique
With the rapid development of Internet information technique, explosive growth is presented in the text information on internet.How from
The information that people's needs are quickly and accurately extracted in large-scale text information has become a hot topic of research.Therefore, information extraction
Technology is come into being.Important component of the entity relation extraction as information extraction, its object is to from natural language text
Excavate the semantic association between entity, not only facilitate the foundation of domain body, promote the building of knowledge mapping, and by pair
Semantic information between entity deeply excavates the search intention that user can be further understood with analysis, to provide more for user
Accurately search service improves user's search experience.
Traditional Chinese entity relation extraction towards be specific area text, qualified relation classification, limit entity class
Not Deng extraction, common method is based on the machine learning method for having supervision.This entity relation extraction method for having supervision
It needs manually to mark relationship corpus, predefined relationship type, but unstructured, the cross-cutting Chinese for middle magnanimity of facing the reality
Text is many times difficult to predefine entity relationship type, so there is the machine learning method of supervision in entity relation extraction
In encounter severe challenge.Therefore, in face of extensive free text, open unsupervised Relation extraction method is by increasingly
More concerns.
Existing open unsupervised Relation extraction method mainly has based on clustering algorithm, based on heuristic rule and based on sentence
The entity abstracting method of method analysis.Wherein, the entity relation extraction method based on cluster is obtained by limitations such as distance, positions
Entity pair, then with certain clustering algorithm by semantic similar entity to gathering for same class cluster, then select representative word to make
It is stated for the relationship of each class cluster.No need to reserve adopted relationship classification, mark corpus, the artificial rule of formulation, very great Cheng for this method
Artificial participation is reduced on degree, but such method still has clusters number intrinsic in some shortcomings, such as clustering algorithm, gathers
The determination problem at class center, and need a large amount of related entities to the effect that can have just trained, for low frequency entity to calling together
The rate of returning is lower, is furthermore difficult to accurately conclude the relationship description word of each class cluster.Entity abstracting method one based on heuristic rule
As be by analyzing a large amount of Chinese entity drawing-out structure features, the entity for summing up common triple is adjusted the distance and relationship
Then deictic words position distribution between entity generates candidate triple by the limitation between distance entity and relative position,
Finally candidate triple is filtered.Research is that relationship instruction is excavated using the method for global sequence and byte orderings
Then word is filtered relationship triple using relationship deictic words and clause rule, the entity that this method is extracted from text
Relationship triple accuracy rate is higher, but may filter out some related entities pair by the limitation of distance between entity pair
And relationship deictic words can not will have the triple of semantic association to filter out comprehensively, so that the recall rate extracted is lower.Base
Pass through the interdependent pass relation recognition entity pair between parsing sentence syntactic structure and word in the entity relation extraction of syntactic analysis
Between relationship.Commonly the method based on syntactic analysis is to obtain most short dependency tree between two entities, then between two entities
Vocabulary in most short independent path constructs triple as relative, such method is easy to operate, but obtained relative can
It can be the combination of multiple words, and the portmanteau word obtained in some complicated Chinese sentences may not necessarily also form correct phrase,
So relationship statement may be very abstract, meaning is relatively fuzzyyer.In addition, being to go deep into digging there are also a kind of method based on syntactic analysis
The interdependent semanteme that is contained of most short interdependent path between pick entity is using features such as dependence, part-of-speech information and positional relationships
It limits, obtains interdependent semantic normal form, if the interdependent path in input text between entity pair is matched with the interdependent semantic normal form mentioned,
Triple relationship can then be extracted.This method is relatively accurate to the relative of extraction, and computation complexity is low, and extraction efficiency is high, and
For complex text, especially there is the sentence of multiple entities, it is not satisfactory to extract effect.But on the whole, it is based on syntactic analysis
Entity relationship computation complexity it is low, the entity relation extraction of extensive, cross-cutting Chinese text can be coped with.
No matter at home and abroad the present age many grammarians emphasize that verb is the core of sentence structure, advocate " verb center
Say " because the syntactic-semantic association of objective reality is mainly manifested in verb lexical item to other lexical item (predominantly nouns between lexical item
Lexical item) restriction on, so verb can be used as investigate semantic structure of sentences starting point, verb and front and back noun can be excavated
The semantic combination relationship of property ingredient.By statisticalling analyze to a large amount of relationship examples, the verb of two entities of connection can generally be indicated
Semantic relation between entity, therefore the ingredient for depending on verb can be analyzed, excavate with verb constitute subject-predicate, dynamic guest,
The lexical item of the dependences such as dynamic benefit can construct accurate entity relationship triple if these lexical items are entities.For example,
For the verb of sentence, if finding the subject and object dependent on the verb, and subject and object are all entities, then the master
Language, object and verb can be configured such that triple.
The study found that the entity relation extraction in nonspecific field is taken out mainly for the relationship between people, mechanism, place at present
It takes, and such as one, some articles are about the article for introducing big data or a product description, wherein people, mechanism, place etc.
The frequency for naming entity to occur is less, if only extracting the semantic relation of these name entities, is difficult to excavate the Deep Semantics of text
Relationship.And the vocabulary frequently occurred in a text document, especially noun, it is able to reflect the master of document description to a certain extent
Content is wanted, therefore key nouns can be increased and expand document entity collection, excavates in more texts and is contacted between main information, thus
Construct semantic network more abundant.
Summary of the invention
It is an object of the invention to the entity relation extractions in face of existing extensive unstructured free text, there is supervision
Method due to be difficult to predefined relationship classification, mark relationship corpus and be restricted, existing open unsupervised entity relationship is taken out
Although taking some effects of method, overall accuracy is not high, to propose that one kind can make cross-cutting free text Relation extraction
More accurate unsupervised Relation extraction method, that is, provide a kind of Chinese entity relation extraction interdependent based on keyword and verb
Method keeps the entity relation extraction of extensive Chinese text more efficient, more accurate.
Realizing the technical solution of above-mentioned purpose is:
A kind of Chinese entity relation extraction method interdependent based on keyword and verb, comprising:
Text is segmented, extracting keywords, generates text key word dictionary;
To text subordinate sentence, each simple sentence is segmented respectively, part-of-speech tagging, name Entity recognition and interdependent syntax are divided
Analysis obtains participle, part of speech, name entity and the interdependent syntactic analysis information of each simple sentence;
Obtain verb set and the entity sets in each simple sentence;
When verb and entity number are both greater than 0 in simple sentence, analysis depend on verb lexical item whether matching relationship syntax advise
Then, if matching, obtains initial entity relationship triple, then expand entity relationship triple;Otherwise, the simple sentence is carried out
Next verb matching;
After all simple sentences execute Relation extraction in text, text triples set is obtained.
Preferably, subordinate sentence processing is carried out according to the text of fullstop, exclamation mark and question mark to input, obtains simple sentence set.
Preferably, when extracting keywords, first word segmentation result is filtered according to part of speech feature, only retains nominal lexical item
As candidate keywords, then the TF-IDF weight of candidate keywords is calculated, the word that threshold value is greater than given threshold is finally inputted into text
This keyword set;Wherein, TF refers to that the number that word occurs in the text, IDF refer to inverse file frequency.
Preferably, the entity sets are made of text overall situation keyword set and name entity.
Preferably, the relationship syntactic rule be according to the interdependent syntactic structure of sentence, using verb as candidate relationship word,
Whether other lexical items and the dependence of verb are subject-predicate, dynamic guest, guest Jie, dynamic benefit relationship in parsing sentence, if there are two in sentence
Dependence between a lexical item and verb is two kinds in these relationships, such as subject-predicate and dynamic guest, subject-predicate and guest Jie etc., and this two
A lexical item is all entity, then can determine initial entity relationship triple.
Preferably, the relationship syntactic rule includes that the non-isA for isA rule and other verbs for judging class verb is advised
Then.
Preferably, in isA rule, the sentence structure of entity rule related with relative is expressed as " Entity1+ Noun
+ is+Entity2 " or " Entity2+is+Entity1+Noun ", entity relationship triple be tentatively expressed as (Entity1,
Noun,Entity2);Wherein, Entity1 and Entity2 is the entity pair in sentence, an entity and is judged based on class verb
Meaning relationship or dynamic guest's relationship, another entity and judge class verb without direct relation;Noun indicates the noun in sentence, with judgement
There are subject-predicate relationship or dynamic guest's relationships for class verb, and there are another entity and judge class verb without direct relation, but with this
Noun is surely middle dependence, for modifying the noun;
In isA rule, the entity rule unrelated with relative refers in sentence that there are an entities with class verb is judged and is
Subject-predicate relationship there are a noun and judge that class verb for dynamic guest's relationship, and is coordination between entity and entity, sentence
Structure is represented by " Entity1+Conj+Entity2 (++)+is+Noun ", and relationship triple can be tentatively expressed as
(Entity1,Noun,Entity2);Wherein, Entity1 and Entity2 is the entity pair in sentence, and Entity2 (++) is indicated
There may be one or more entities and Entity1 are arranged side by side, Noun is the noun in sentence.
Preferably, non-isA rule includes that verb has subject rule and verb without subject rule;
Verb have subject rule include Subject, Predicate and Object structure, subject-predicate guest Jie structure, subject-predicate mend guest's structure, preceding guest guest Jie structure and
Other structures specifically include:
The Subject, Predicate and Object structure refers to from a certain verb, and according to interdependent syntax, the subject and object of the verb exist,
And all be entity, initial entity relationship triple can be built up;
The subject-predicate guest Jie structure refers to from a certain verb, and according to interdependent syntax, the subject of the verb exists and is
Entity, the preposition for depending on the verb have object and are entity, can extract initial entity relationship triple out;
The subject-predicate is mended guest's structure and is referred to from a certain verb, which is intransitive verb, should according to interdependent syntax
Verb is there are subject and is entity, there is the complement for depending on verb, and complement has object and is entity, may make up initial reality
Body relationship triple;
The preceding guest guest Jie structure refers to from a certain verb, according to interdependent syntax, exist before depending on the verb
It sets object and is entity, there is the preposition for depending on the verb, and preposition has object and is entity, can form initial entity and close
It is triple;
The other structures refer to from a certain verb, and according to interdependent syntax, the subject of the verb exists and is entity,
In the presence of the other structures for depending on verb, the structure is there are object and is entity, can construct triple relationship;
Verb includes verb parallel construction and sentence without subject structure without subject rule, is specifically included:
The verb parallel construction indicates that there are a lexical items, and dynamic guest's relationship can be directly established with some verb in sentence
Or establish guest's Jie relationship indirectly, mend guest's relationship, and the lexical item is entity, and the lexical item of subject-predicate relationship can not be established with the verb,
But there are other verbs arranged side by side with the verb, the two subject is consistent, so using the subject of its verb arranged side by side as subject, it can structure
Build entity relationship triple;
Subject is not present without subject representation sentence in the sentence, but there are a lexical items directly to build with some verb
Vertical dynamic guest's relationship establishes guest Jie indirectly, mends guest's relationship, and the lexical item is entity, and the sentence can be traced according to Chinese heuristic rule
The previous sentence of son, subject of the subject of former statement kernel word aroused in interest as this;In interdependent syntax theory, advocate that core is dynamic
Word is the center compositions of sentence, dominates other compositions, and a sentence, there may be multiple verbs, each verb may be deposited
In subject, so the rule only takes subject of the subject of the core verb of preceding sentence as this.
Preferably, described that entity relationship triple is expanded, including entity word expands, relative expands and arranged side by side
Triple expands, specifically:
The entity word expansion is to merge keyword entity with its attribute qualifier;
It includes the addition negative adverbial modifier, addition non-physical object that the relative, which expands,;
The triple arranged side by side expansion is incited somebody to action when having obtained the entity in entity relationship triple there are when entity arranged side by side
Entity arranged side by side and the relative form new triple.
The beneficial effects of the present invention are: the present invention be different from general Chinese entity extraction only focus on name entity (name,
Name, mechanism name) between relationship, with the traditional name entity set of text overall situation keyword expansion, increase description text mainly in
The semantic relation of the keyword of appearance keeps the text semantic network of building more abundant.The present invention, which is different from, is generally basede on syntax point
The unsupervised entity abstracting method of analysis is mostly from entity to its relative of set off in search, and inefficient, this method is to wait with verb
Relative is selected, semantic relation between the nominal ingredient in front and back can be connected according to verb in Chinese grammer, with the interdependent syntax of verb
For foothold, without limiting the distance between entity, without considering relatival position, it is relatively generally basede on the entity of syntactic analysis
The triple more precise and high efficiency that abstracting method obtains.
Detailed description of the invention
Fig. 1 is entity relation extraction flow chart of the present invention;
Fig. 2 is relationship syntactic rule classification chart of the present invention;
Fig. 3 is an example schematic of the present invention according to syntactic analysis;
Fig. 4 is two example schematics of the present invention according to syntactic analysis;
Fig. 5 is three example schematics of the present invention according to syntactic analysis;
Fig. 6 is a schematic diagram of relationship syntactic rule proposed by the present invention;
Fig. 7 is two schematic diagrames of relationship syntactic rule proposed by the present invention;
Fig. 8 is three schematic diagrames of relationship syntactic rule proposed by the present invention;
Fig. 9 is four schematic diagrames of relationship syntactic rule proposed by the present invention;
Figure 10 is the example schematic in the present invention according to syntactic analysis;
Figure 11 is two example schematics in the present invention according to syntactic analysis.
Specific embodiment
The present invention will be further described with reference to the accompanying drawings.
The Chinese entity relation extraction method interdependent based on keyword and verb of the invention, analyzes the interdependent pass of verb
System realizes the entity relation extraction of extensive free text, supports for building text semantic network provided data, referring to Fig. 1, tool
Body the following steps are included:
Step 1, the text of input is segmented, extracting keywords, the keyword of extraction is generated into keyword thesaurus.
The purpose for extracting text overall situation keyword is to expand conventional entity collection, and conventional entity collection is just for name, place name, machine
Structure name etc. name entity, and the present invention be towards the extensive no free text in field, if a text document is almost without people
Name, place name and mechanism name, then will lead to not take out entity relationship, so the present invention is using keyword as one of entity set
Point, entity relation extraction is carried out, semantic relation in document is excavated.Keyword is the word or phrase that can indicate text subject,
And keyword is mostly noun, the keyword in a corpus of text may frequently occur in the document, and in other documents
Middle frequency of occurrence is less, so first segmenting to text when the present invention extracts keyword, filtering out one further according to part of speech feature
The word of a little non-name parts of speech, only retains nominal lexical item as candidate keywords, calculates the TF-IDF (word frequency-of candidate keywords
Inverse document frequency) weight, the word that threshold value is greater than given threshold and inputs text key word set as text key word.Its
In, TF refers to that the number that word occurs in the text, IDF refer to inverse file frequency, extracts TF-IDF feature and tends to identification at certain
Often occur in piece document, but the uncommon word in other documents.
Step 2, according to fullstop, exclamation mark, text progress subordinate sentence processing of the question mark to input, output simple sentence set.
Step 3, each simple sentence segmented respectively, part-of-speech tagging, name Entity recognition and interdependent syntactic analysis, obtained
Participle, part of speech, name entity and the interdependent syntactic analysis information of each simple sentence.
Wherein, name entity mainly obtains the entities such as name in sentence, place name, mechanism name.Each language list in sentence structure
First (lexical item) is constantly present certain relationship, and this relationship is being mainly reflected between linguistic unit (lexical item) in sentence
Syntactic-semantic relationship on, interdependent syntactic analysis is exactly the relation of interdependence between each linguistic unit of parsing sentence (lexical item),
To identify in sentence the grammatical items such as " Subject, Predicate and Object ", " determining shape benefit ".
Step 4, verb collection and the entity set in each simple sentence are obtained, if verb and entity number are both greater than 0 in simple sentence,
Step 5 is carried out, the processing to the sentence is otherwise terminated.
Wherein, entity not only includes name entity, further includes the global keyword set for describing text subject content, makes
It obtains text entities library and more enriches and enrich.
Step 5, analysis depend on verb lexical item whether matching relationship syntactic rule, if matching, obtain initial entity
Relationship triple, then entity relationship triple is expanded;Otherwise, the next verb for carrying out the sentence is matched;
Relationship syntactic rule is formulated before entity extracts, and is according to interdependent syntactic analysis and to combine some Chinese grammers
Heuristic rule building.In Modern Chinese grammer, verb has other semantic components such as agent, word denoting the receiver of an action etc. in sentence
Certain restrictive function, it is possible to which verb is excavated into verb and its front and back name as the starting point of parsing sentence semantic structure
The semantic combination relationship of part of speech ingredient.Therefore, relationship syntactic rule of the invention is from verb, and analysis depends on verb
Ingredient, for example when the subject of verb and object exist and be all entity, which just constitutes an entity relationship
Triple.The relationship syntactic rule of satisfaction only obtains initial practice relationship triple, it is also necessary to further to entity relationship ternary
Group carries out modification supplement.
Step 6, after all sentences are carried out Relation extraction in text, text triples set can be obtained.
In conjunction with described above, core of the invention is that relationship syntactic rule and entity relationship triple expand, below
Dense medium continues related content:
One, relationship syntactic rule
In Modern Chinese grammer, verb is the foothold of sentence semantics analysis, it is to other semantic components in sentence
Such as agent, word denoting the receiver of an action have certain restrictive function.By finding to a large amount of entity relationship instance analysis, entity relationship triple is total
Occur with the syntactic structure of certain fixations, and verb often plays connection function in these structures, so the verb of two entities of connection
It generally can be with the semantic relation between presentation-entity.In interdependent syntactic structure, main mark relationship is as shown in table 1, wherein with
The related structure of verb has: subject-predicate relationship, dynamic guest's relationship, preposition object, structure of complementation, verbal endocentric phrase, guest's Jie relationship etc..By this
Several structure combinatorial mappings in a little structures are relationship syntactic rule, can be applied to entity relation extraction.Having a kind of verb is to sentence
Disconnected class verb (such as verb "Yes") not as relative in Relation extraction, and noun associated with it just plays relationship and connects
The effect connect.Therefore, the present invention, which is individually handled, judges class verb, and relation rule is divided into isA rule and non-isA rule two is big
Class, each class are divided into each group again, and referring to fig. 2, isA rule includes entity rule related with relative, entity and pass
The unrelated rule of copula.Non- isA rule includes that verb has subject rule and verb without subject rule, and wherein verb has subject regular
Guest's structure, preceding guest guest Jie structure and other structures are mended including Subject, Predicate and Object structure, subject-predicate guest Jie structure, subject-predicate;Verb is without subject
Structure includes verb parallel construction and sentence without subject structure.
Relationship type | Interdependent label |
Subject-predicate relationship | SBV |
Dynamic guest's relationship | VOB |
Preposition object | FOB |
Verbal endocentric phrase | ADV |
Structure of complementation | CMP |
Guest's Jie relationship | POB |
Nominal endocentric phrase | ATT |
Coordination | COO |
And language | DBL |
Between guest's relationship | IOB |
Left additional relationships | LAD |
Right additional relationships | RAD |
Absolute construction | IS |
Punctuate | WP |
Key Relationships | HAD |
Table 1
Table 1 is that interdependent syntax marks relationship.
1, isA rule:
Judge the effect of verb (such as verb "Yes") mainly with the presence of table judgement, table explanation, table in Modern Chinese.Wherein,
Table judgement is to indicate that things belongs to or what is equal to, if Beijing is the capital of China;Table illustrate be the feature for indicating things,
Situation or situation, if this vehicle is red;It is the existence for indicating things that table, which exists, is such as cattle and sheep everywhere.Present invention research
When entity extracts relationship, table judgement effect therein is only considered.According to relationship example, verb "Yes" is generally not as relationship description
IsA rule is divided into entity rule related with relative, the entity rule unrelated with relative by word.
1.1, entity and the related rule of relative:
Entity rule related with relative refers to that the subject of verb "Yes" is entity, and object is common noun or verb
The subject of "Yes" is common noun, and object is entity, and is surely middle relationship there are another entity and the noun.Its sentence structure
It is expressed as " Entity1+Noun+is+Entity2 " or " Entity2+is+Entity1+Noun ", at the beginning of entity relationship triple
Step is expressed as (Entity1, Noun, Entity2);Wherein, Entity1 and Entity2 is the entity pair in sentence, a reality
Body and verb "Yes" are subject-predicate relationship or dynamic guest's relationship, another entity and verb "Yes" are without direct relation;Noun indicates sentence
In noun, there are subject-predicate relationships or dynamic guest's relationship with verb "Yes", and there are another entity with verb "Yes" without direct
Relationship, but be surely middle dependence with the noun, for modifying the noun.As " Chinese capital is Beijing.", interdependent syntax
Referring to Fig. 3, verb "Yes" is " capital " there are subject " capital ", object " Beijing ", " China " as sentence core word for analysis
Attribute can construct triple (China, capital, Beijing);For another example " Zhang is the son of famous movie star Zhang San.", interdependent syntax
Referring to fig. 4, verb "Yes" is as sentence core word, and there are subject " Zhang ", object " son ", " Zhang San " is " son " for analysis
Attribute may make up triple (Zhang San, son, Zhang).
1.2, the entity rule unrelated with relative:
It is entity that the entity rule unrelated with relative, which is verb "Yes" there are subject, is common noun there are object, deposits
It is coordination in another or multiple entities and subject entity.Its sentence structure is expressed as " Entity1+Noun+is+
Entity2 " or " Entity2+is+Entity1+Noun ", entity relationship triple be tentatively expressed as (Entity1, Noun,
Entity2);Wherein, Entity1 and Entity2 is the entity pair in sentence, an entity and verb "Yes" be subject-predicate relationship or
Dynamic guest's relationship, another entity and verb "Yes" are without direct relation;Noun indicates the noun in sentence, exists with verb "Yes" and leads
Meaning relationship or dynamic guest's relationship, and there are another entity and verb "Yes" without direct relation, but with the noun be in fixed it is interdependent
Relationship, for modifying the noun.As " Xiao Li and Xiao Wang are men and wives.", interdependent syntactic analysis is referring to Fig. 5, verb "Yes" conduct
Sentence core word is coordination there are subject " Xiao Li ", object " man and wife ", " Xiao Li " and " Xiao Wang ", it is (small can to form triple
Lee, man and wife, Xiao Wang).
2, non-isA rule:
Non- isA rule refers to judge that verb as core, constructs entity relationship sentence as the non-of Entity Semantics relationship description
Method rule.Core of the verb as syntactic structure, semantic structure, from syntax, verb determines the basic side of sentence structure
Looks, from semantically, semantic structure be set up by core of verb, so referred to as predication structure or meaning state structure,
Kinetonucleus structure.The present invention constructs the relationship syntax rule of non-isA according to the mapping between basic predication structure and typical syntactic structure
Then, the semantic connection due to verb between lexical item, need to obtain two lexical items dominated by it, whether there is or not by its domination according to verb
Subject, which is divided into verb, has subject rule and verb without subject rule, can be divided into several groups again under every rule-like, (referring to fig. 2)
It is specifically described as follows:
2.1, verb has subject regular:
(1) Subject, Predicate and Object structure, Subject, Predicate and Object structure are the most common structures in Verb Predicate Sentence, and the present invention constructs Subject, Predicate and Object knot
Structure rule refers to from a certain verb, and according to interdependent syntax, the subject and object of the verb exist, and is all entity, building
Initial triple out, dependence is referring to Fig. 6 between verb and entity, wherein E1, E2 presentation-entity, V indicates verb, interdependent
The specific paraphrase of arc is referring to table 1.For example, " king five accesses the U.S.." in verb " access " subject be " king five ", the object of verb
For " U.S. ", exportable triple (king five, access, the U.S.).
(2) subject-predicate guest Jie structure, which refers to that verb and entity are expressed as subject-predicate guest's Jie structure on syntax, i.e., from certain
One verb sets out, and according to interdependent syntax, the subject of the verb exists and is entity, and the preposition for depending on the verb has object and is
Entity can extract initial triple out, and dependence is referring to Fig. 7 between verb and entity, wherein E1, E2 presentation-entity, V table
Show that verb, P indicate preposition, the specific paraphrase of interdependent arc is referring to table 1.For example, verb in " Li Si gives a lecture in Tongji University "
The subject of " delivering " is " Li Si ", and the preposition " " for depending on verb " delivering " has object " Tongji University ", exportable initial
Triple (Li Si delivers, Tongji University).
(3) subject-predicate mends guest's structure, which refers to that verb and entity are expressed as subject-predicate on syntax and mend guest's structure, i.e., from certain
One verb sets out, which is intransitive verb, and according to interdependent syntax, the subject of the verb exists and is entity, and there are interdependent
In the complement of verb, which has object and is entity, may make up preliminary triple, and specific dependence indicates to participate in Fig. 8,
In, E1, E2 presentation-entity, V indicate that verb, P indicate preposition, and the specific paraphrase of interdependent arc is referring to table 1.For example, " king five graduates from
Harvard University." in intransitive verb " graduation " there are subject " king five ", there are complement prepositions " in ", and there are objects for preposition " in "
" Harvard University " can export triple (king five, graduation, Harvard University).
(4) guest guest Jie structure before, which refers to guest Jie's guest's structure before verb and entity are expressed as on syntax, i.e., from certain
One verb sets out, and according to interdependent syntax, which has preposition object and be entity, there is the preposition dominated by it, and preposition is deposited
In object and it is entity, initial triple can be formed, specific dependence is indicated referring to Fig. 9, E1, E2 presentation-entity, V table
Show that verb, P indicate preposition, the specific paraphrase of interdependent arc is referring to table 1.Substantially there is no subjects for the rule, but verb still props up
With two lexical items, so can be regarded as the similar structure for having subject, this structure is common in passive sentence pattern, such as " Zhang San's quilt
Fudan University enrolls." in verb " admission " preposition object be " Zhang San ", there is the preposition " quilt " for depending on " admission ", Jie
Word " quilt " has object " Fudan University ", exportable triple (Fudan University, admission, Zhang San).
(5) other structures, the rule refer to from a certain verb, according to interdependent syntax, the subject of the verb exist and
For entity, there are the other structures for depending on verb, the structure is there are object and is entity, can construct triple relationship.
For example, " Li Si emphasizes to promote financial reform in enterprise's forum.", for the dependence of the sentence referring to Figure 10, verb is " strong
Adjust " there are subject " Li Si " and object " propulsion ", verb " propulsion " can construct initial triple there are object " reform "
(Li Si emphasizes, reforms).
2.2, verb is without subject rule:
(1) verb parallel construction, verb parallel construction indicate that subject is not present in verb, and there are some lexical items can be with the verb
It directly establishes dynamic guest's relationship or establishes dynamic guest's relationship indirectly by another word, and this lexical item is entity, but is existed and the verb
Other verbs arranged side by side, the two subject is consistent, so entity relationship three can be constructed using the subject of its verb arranged side by side as subject
Tuple, specific interdependent syntax are indicated referring to Figure 11, wherein E1, E2, E3 presentation-entity, V1 and V2 indicate verb, interdependent arc
LikeVOB refers to lexical item and verb may be directly dynamic guest's relationship, it is also possible to by some word obtain it is indirect move guest's relationship (such as certain
Lexical item is the object of certain preposition, which depends on some verb).For example, " Li Si accesses China, and delivers in Tongji University
Speech." in previous verb can be extracted triple (Li Si, access, China) according to " Subject, Predicate and Object structure " this rule, second
A verb " delivering " obtains object " Tongji University " by guest's Jie structure, and but without subject, but verb " delivering " and " access " are simultaneously
Column, so according to the subject consistency of two verbs, it may be determined that another preliminary triple (Li Si delivers, Tongji University).This
Embodiment but is not limited to two verbs arranged side by side only with two verb examples arranged side by side, the rule is equally applicable include it is multiple simultaneously
The sentence of column verb.
(2) subject is not present without subject structure, the Rule Expression sentence in sentence, and there are a lexical items can be straight with some verb
It connects the dynamic guest's relationship of foundation or establishes the relationships such as guest Jie, dynamic benefit guest indirectly, and the lexical item is entity, it can according to Chinese heuristic rule
Trace the previous sentence of the sentence, subject of the subject of former statement kernel word aroused in interest as this;It is main in interdependent syntax theory
The center compositions that core verb is sentence are opened, other compositions are dominate, and there may be multiple verbs, each verbs for a sentence
May all there be subject, so the rule only takes subject of the subject of preceding statement kernel word aroused in interest as this.
Two, entity relationship triple expands
Meet relationship syntactic rule and only obtain initial triple, the present invention also needs further to expand triple, packet
It includes entity expansion, relative expansion and triple expands side by side, specifically:
(1) entity expands, the purpose that entity expands be the participle stage is split noun phrase for multiple lexical items into
Row merges, and the present invention merges keyword entity (non-name, place name, mechanism name) with its attribute qualifier;If keyword entity
Previous word and the keyword entity are surely middle relationships, and do not constitute and connect for quantifier or keyword and preceding several non-quantifier lexical items
Continuous nominal endocentric phrase then merges it with keyword entity.As " Li Si emphasizes to promote in enterprise's forum in above-described embodiment
Financial reform." initial triple (Li Si emphasizes, reforms) is extracted, " finance " is to modify the attribute of " reform " in the sentence,
So the two is merged, triple is updated to (Li Si emphasize, financial reform).
(2) relative expands, and in order to keep relationship description between entity more accurate specific, the present invention gives relative addition negative
The adverbial modifier, non-physical object expand;Wherein, if there is negative adverb modification before verb as candidate relationship word, expression be with
The opposite meaning of verb, therefore need to will be added in relative with the negative adverb before candidate relationship word, for example, " Zhang San does not like
Joyous sea ", according to relationship syntactic rule, the initial triple of extraction is (Zhang San likes, sea), is expanded more by relative
Newly it is (Zhang San does not like, sea).If not can make between two entities in addition, entity word acts as the object of verb in sentence
Relationship is more clear.In above-described embodiment in " Li Si gives a lecture in Tongji University " triple by it is initial (Li Si delivers,
Tongji University) it is extended to (Li Si gives a lecture, Tongji University).
(3) triple expands side by side, is commonly present subject arranged side by side or object arranged side by side in sentence, when these subjects or object are real
When entity in body relationship triple, its entity arranged side by side need to be formed new triple.For example, for sentence " Xiao Fang, it is small red and
Small China is the daughter of Xiao Li and Xiao Wang.", the relationship syntactic rule proposed according to the present invention, the initial triple of extraction is (small
Lee, daughter, Xiao Fang), expanded by entity arranged side by side, increase other 5 triples, respectively (Xiao Li, daughter are small red),
(Xiao Li, daughter, small China), (Xiao Wang, daughter, Xiao Fang), (Xiao Wang, daughter are small red), (Xiao Wang, daughter, small China).
Above embodiments are used for illustrative purposes only, rather than limitation of the present invention, the technology people in relation to technical field
Member, without departing from the spirit and scope of the present invention, can also make various transformation or modification, therefore all equivalent
Technical solution also should belong to scope of the invention, should be limited by each claim.
Claims (9)
1. a kind of Chinese entity relation extraction method interdependent based on keyword and verb characterized by comprising
Text is segmented, extracting keywords, generates text key word dictionary;
To text subordinate sentence, each simple sentence is segmented respectively, part-of-speech tagging, names Entity recognition and interdependent syntactic analysis, is obtained
Obtain participle, part of speech, name entity and the interdependent syntactic analysis information of each simple sentence;
Obtain verb set and the entity sets in each simple sentence;
When verb and entity number are both greater than 0 in simple sentence, analysis depend on verb lexical item whether matching relationship syntactic rule, if
Matching, obtains initial entity relationship triple, then expand entity relationship triple;Otherwise, it carries out under the simple sentence
One verb matching;
After all simple sentences execute Relation extraction in text, text triples set is obtained.
2. the Chinese entity relation extraction method interdependent based on keyword and verb according to claim 1, feature exist
In carrying out subordinate sentence processing to the text of input according to fullstop, exclamation mark and question mark, obtain simple sentence set.
3. the Chinese entity relation extraction method interdependent based on keyword and verb according to claim 1, feature exist
When, extracting keywords, first word segmentation result is filtered according to part of speech feature, only retains nominal lexical item as candidate key
Word, then the TF-IDF weight of candidate keywords is calculated, the word that threshold value is greater than given threshold is finally inputted into text key word set;
Wherein, TF refers to that the number that word occurs in the text, IDF refer to inverse file frequency.
4. the Chinese entity relation extraction method interdependent based on keyword and verb according to claim 1, feature exist
In the entity sets are made of text overall situation keyword set and name entity.
5. the Chinese entity relation extraction method interdependent based on keyword and verb according to claim 1, feature exist
In, the relationship syntactic rule be according to the interdependent syntactic structure of sentence, using verb as candidate relationship word, its in parsing sentence
Whether his lexical item and the dependence of verb are subject-predicate, dynamic guest, guest Jie, dynamic benefit relationship, if there are two lexical items and verbs in sentence
Between dependence be two kinds in these relationships, for example subject-predicate and dynamic guest, subject-predicate and guest Jie etc., and the two lexical items are all real
Body then can determine initial entity relationship triple.
6. the Chinese entity relation extraction method interdependent based on keyword and verb according to claim 5, feature exist
In the relationship syntactic rule includes the non-isA rule for isA rule and other verbs for judging class verb.
7. the Chinese entity relation extraction method interdependent based on keyword and verb according to claim 6, feature exist
In,
In isA rule, the sentence structure of entity rule related with relative is expressed as " Entity1+Noun+is+Entity2 "
Or " Entity2+is+Entity1+Noun ", entity relationship triple are tentatively expressed as (Entity1, Noun, Entity2);Its
In, Entity1 and Entity2 are the entities pair in sentence, an entity and judge class verb for subject-predicate relationship or dynamic guest's relationship,
Another entity and judge class verb without direct relation;Noun indicates the noun in sentence, and there are subject-predicate passes with class verb is judged
System or dynamic guest's relationship, and exist another entity and judge class verb without direct relation, but with the noun be calmly in interdependent pass
System, for modifying the noun;
IsA rule in, the entity rule unrelated with relative refer in sentence there are an entity with judge class verb for subject-predicate
Relationship there are a noun and judge that class verb for dynamic guest's relationship, and is coordination between entity and entity, sentence structure
Be represented by " Entity1+Conj+Entity2 (++)+is+Noun ", relationship triple can tentatively be expressed as (Entity1,
Noun,Entity2);Wherein, Entity1 and Entity2 is the entity pair in sentence, and Entity2 (++) indicates that there may be one
A or multiple entities and Entity1 are arranged side by side, and Noun is the noun in sentence.
8. the Chinese entity relation extraction method interdependent based on keyword and verb according to claim 6, feature exist
In non-isA rule includes that verb has subject rule and verb without subject rule;
Verb have subject rule include Subject, Predicate and Object structure, subject-predicate guest Jie structure, subject-predicate mend guest's structure, preceding guest guest Jie structure and other
Structure specifically includes:
The Subject, Predicate and Object structure refers to from a certain verb, and according to interdependent syntax, the subject and object of the verb exist, and all
For entity, initial entity relationship triple can be built up;
The subject-predicate guest Jie structure refers to from a certain verb, and according to interdependent syntax, the subject of the verb exists and is entity,
The preposition for depending on the verb has object and is entity, can extract initial entity relationship triple out;
The subject-predicate is mended guest's structure and is referred to from a certain verb, which is intransitive verb, according to interdependent syntax, the verb
There are subject and it is entity, there is the complement for depending on verb, and complement has object and is entity, may make up initial entity and closes
It is triple;
The preceding guest guest Jie structure refers to from a certain verb, according to interdependent syntax, there is the preposition guest for depending on the verb
, there is the preposition for depending on the verb in language and be entity, and preposition has object and for entity, can form initial entity relationship three
Tuple;
The other structures refer to from a certain verb, and according to interdependent syntax, the subject of the verb exists and is entity, exist
The other structures of verb are depended on, the structure is there are object and is entity, can construct triple relationship;
Verb includes verb parallel construction and sentence without subject structure without subject rule, is specifically included:
The verb parallel construction indicate to exist in sentence a lexical item can directly be established with some verb move guest's relationship or
It connects guest's Jie relationship of establishing, mend guest's relationship, and the lexical item is entity, can not establish the lexical item of subject-predicate relationship with the verb, but deposit
In other verbs arranged side by side with the verb, the two subject is consistent, so can construct reality using the subject of its verb arranged side by side as subject
Body relationship triple;
Subject is not present without subject representation sentence in the sentence, but can directly establish and move with some verb there are a lexical item
Guest's relationship establishes guest Jie indirectly, mends guest's relationship, and the lexical item is entity, and the sentence can be traced according to Chinese heuristic rule
Previous sentence, subject of the subject of former statement kernel word aroused in interest as this;In interdependent syntax theory, advocate that core verb is
The center compositions of sentence, dominate other compositions, and a sentence is there may be multiple verbs, and each verb may have master
Language, so the rule only takes subject of the subject of the core verb of preceding sentence as this.
9. the Chinese entity relation extraction method interdependent based on keyword and verb according to claim 1, feature exist
In, it is described that entity relationship triple is expanded, including entity word expands, relative expands and triple expands side by side,
Specifically:
The entity word expansion is to merge keyword entity with its attribute qualifier;
It includes the addition negative adverbial modifier, addition non-physical object that the relative, which expands,;
The described triple arranged side by side expansion be when having obtained the entity in entity relationship triple there are when entity arranged side by side, will be arranged side by side
Entity and the relative form new triple.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811124153.1A CN109241538B (en) | 2018-09-26 | 2018-09-26 | Chinese entity relation extraction method based on dependency of keywords and verbs |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811124153.1A CN109241538B (en) | 2018-09-26 | 2018-09-26 | Chinese entity relation extraction method based on dependency of keywords and verbs |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109241538A true CN109241538A (en) | 2019-01-18 |
CN109241538B CN109241538B (en) | 2022-12-20 |
Family
ID=65056162
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811124153.1A Active CN109241538B (en) | 2018-09-26 | 2018-09-26 | Chinese entity relation extraction method based on dependency of keywords and verbs |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109241538B (en) |
Cited By (78)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109582975A (en) * | 2019-01-31 | 2019-04-05 | 北京嘉和美康信息技术有限公司 | It is a kind of name entity recognition methods and device |
CN109918672A (en) * | 2019-03-13 | 2019-06-21 | 东华大学 | A kind of structuring processing method of the Thyroid ultrasound report based on tree construction |
CN109977235A (en) * | 2019-04-04 | 2019-07-05 | 吉林大学 | A kind of determination method and apparatus of trigger word |
CN109992651A (en) * | 2019-03-14 | 2019-07-09 | 广州智语信息科技有限公司 | A kind of problem target signature automatic identification and abstracting method |
CN109992777A (en) * | 2019-03-26 | 2019-07-09 | 浙江大学 | A kind of crucial semantic information extracting method of Chinese medicine state of an illness text based on keyword |
CN110032649A (en) * | 2019-04-12 | 2019-07-19 | 北京科技大学 | Relation extraction method and device between a kind of entity of TCM Document |
CN110046351A (en) * | 2019-04-19 | 2019-07-23 | 福州大学 | Text Relation extraction method under regular drive based on feature |
CN110083822A (en) * | 2019-03-06 | 2019-08-02 | 杭州电子科技大学 | It is a kind of from demand text conversion to the conversion method of SysML demand figure |
CN110110329A (en) * | 2019-04-30 | 2019-08-09 | 湖南星汉数智科技有限公司 | A kind of entity behavior derivation method, apparatus, computer installation and computer readable storage medium |
CN110119510A (en) * | 2019-05-17 | 2019-08-13 | 浪潮软件集团有限公司 | A kind of Relation extraction method and device based on transmitting dependence and structural auxiliary word |
CN110162788A (en) * | 2019-05-06 | 2019-08-23 | 三角兽(北京)科技有限公司 | The determination method and device of entity dependence |
CN110196913A (en) * | 2019-05-23 | 2019-09-03 | 北京邮电大学 | Multiple entity relationship joint abstracting method and device based on text generation formula |
CN110222332A (en) * | 2019-04-29 | 2019-09-10 | 闽江学院 | The method for realizing name of the dish Entity recognition based on dependency analysis |
CN110263336A (en) * | 2019-06-12 | 2019-09-20 | 东华大学 | A method of building breast ultrasound domain body |
CN110263341A (en) * | 2019-06-20 | 2019-09-20 | 贵州电网有限责任公司 | A method of profile is excavated and positioned from text |
CN110263120A (en) * | 2019-04-26 | 2019-09-20 | 北京零秒科技有限公司 | Corpus labeling method and device |
CN110309393A (en) * | 2019-03-28 | 2019-10-08 | 平安科技(深圳)有限公司 | Data processing method, device, equipment and readable storage medium storing program for executing |
CN110362673A (en) * | 2019-07-17 | 2019-10-22 | 福州大学 | Computer vision class papers contents method of discrimination and system based on abstract semantic analysis |
CN110377901A (en) * | 2019-06-20 | 2019-10-25 | 湖南大学 | A kind of text mining method for making a report on case for distribution line tripping |
CN110413732A (en) * | 2019-07-16 | 2019-11-05 | 扬州大学 | The knowledge searching method of software-oriented defect knowledge |
CN110502642A (en) * | 2019-08-21 | 2019-11-26 | 武汉工程大学 | A kind of entity relation extraction method based on interdependent syntactic analysis and rule |
CN110532553A (en) * | 2019-08-21 | 2019-12-03 | 河海大学 | A kind of method water conservancy spatial relationship word identification and extracted |
CN110543630A (en) * | 2019-08-21 | 2019-12-06 | 北京仝睿科技有限公司 | Method and device for generating text structured representation and computer storage medium |
CN110555083A (en) * | 2019-08-26 | 2019-12-10 | 北京工业大学 | non-supervision entity relationship extraction method based on zero-shot |
EP3579120A1 (en) * | 2018-06-04 | 2019-12-11 | Infosys Limited | Extraction of tokens and relationship between tokens from documents to form an entity relationship map |
CN110569510A (en) * | 2019-09-17 | 2019-12-13 | 四川长虹电器股份有限公司 | method for identifying named entity of user request data |
CN110569504A (en) * | 2019-09-04 | 2019-12-13 | 北京明略软件系统有限公司 | relation word determining method and device |
CN110597998A (en) * | 2019-07-19 | 2019-12-20 | 中国人民解放军国防科技大学 | Military scenario entity relationship extraction method and device combined with syntactic analysis |
CN110705310A (en) * | 2019-09-20 | 2020-01-17 | 北京金山数字娱乐科技有限公司 | Article generation method and device |
CN110765759A (en) * | 2019-10-21 | 2020-02-07 | 普信恒业科技发展(北京)有限公司 | Intention identification method and device |
CN110874396A (en) * | 2019-11-07 | 2020-03-10 | 腾讯科技(深圳)有限公司 | Keyword extraction method and device and computer storage medium |
CN110909537A (en) * | 2019-11-19 | 2020-03-24 | 曲英洲 | Artificial intelligence method for modern Chinese component analysis |
CN110991180A (en) * | 2019-11-28 | 2020-04-10 | 同济人工智能研究院(苏州)有限公司 | Command identification method based on keywords and Word2Vec |
CN111126052A (en) * | 2019-12-26 | 2020-05-08 | 中科鼎富(北京)科技发展有限公司 | Function point generation method and device, electronic equipment and computer readable storage medium |
CN111178079A (en) * | 2019-12-31 | 2020-05-19 | 北京明略软件系统有限公司 | Triple extraction method and device |
CN111177215A (en) * | 2019-12-20 | 2020-05-19 | 京东数字科技控股有限公司 | Method and device for generating financial data |
CN111198932A (en) * | 2019-12-30 | 2020-05-26 | 北京明略软件系统有限公司 | Triple acquiring method and device, electronic equipment and readable storage medium |
CN111241827A (en) * | 2020-01-10 | 2020-06-05 | 同方知网(北京)技术有限公司 | Attribute extraction method based on sentence retrieval mode |
CN111382571A (en) * | 2019-11-08 | 2020-07-07 | 南方科技大学 | Information extraction method, system, server and storage medium |
CN111597794A (en) * | 2020-05-11 | 2020-08-28 | 浪潮软件集团有限公司 | Dependency relationship-based 'yes' word and sentence relationship extraction method and device |
CN111597812A (en) * | 2020-05-09 | 2020-08-28 | 北京合众鼎成科技有限公司 | Financial field multiple relation extraction method based on mask language model |
CN111597349A (en) * | 2020-04-30 | 2020-08-28 | 西安理工大学 | Rail transit standard entity relation automatic completion method based on artificial intelligence |
CN111597351A (en) * | 2020-05-14 | 2020-08-28 | 上海德拓信息技术股份有限公司 | Visual document map construction method |
CN111611399A (en) * | 2020-04-15 | 2020-09-01 | 广发证券股份有限公司 | Information event mapping system and method based on natural language processing |
CN111639499A (en) * | 2020-06-01 | 2020-09-08 | 北京中科汇联科技股份有限公司 | Composite entity extraction method and system |
CN111666738A (en) * | 2020-06-09 | 2020-09-15 | 南京师范大学 | Formalized coding method for motion description natural text |
CN111666767A (en) * | 2020-06-10 | 2020-09-15 | 创新奇智(上海)科技有限公司 | Data identification method and device, electronic equipment and storage medium |
CN111680492A (en) * | 2020-06-10 | 2020-09-18 | 创新奇智(青岛)科技有限公司 | New word mining method and device and electronic equipment |
CN111798847A (en) * | 2020-06-22 | 2020-10-20 | 广州小鹏车联网科技有限公司 | Voice interaction method, server and computer-readable storage medium |
CN111984778A (en) * | 2020-09-08 | 2020-11-24 | 四川长虹电器股份有限公司 | Dependency syntax analysis and Chinese grammar-based multi-round semantic analysis method |
CN112052340A (en) * | 2020-08-10 | 2020-12-08 | 深圳数联天下智能科技有限公司 | Data model construction method and device and electronic equipment |
CN112084793A (en) * | 2020-09-14 | 2020-12-15 | 深圳前海微众银行股份有限公司 | Semantic recognition method, device and readable storage medium based on dependency syntax |
CN112131343A (en) * | 2020-09-14 | 2020-12-25 | 杭州东信北邮信息技术有限公司 | Chinese novel dialect dialogue character recognition method |
CN112148838A (en) * | 2020-09-23 | 2020-12-29 | 北京中电普华信息技术有限公司 | Business source object extraction method and device |
CN112183059A (en) * | 2020-09-24 | 2021-01-05 | 万齐智 | Chinese structured event extraction method |
CN112364648A (en) * | 2020-12-02 | 2021-02-12 | 中金智汇科技有限责任公司 | Keyword extraction method and device, electronic equipment and storage medium |
CN112380868A (en) * | 2020-12-10 | 2021-02-19 | 广东泰迪智能科技股份有限公司 | Petition-purpose multi-classification device based on event triples and method thereof |
CN112560488A (en) * | 2020-12-07 | 2021-03-26 | 北京明略软件系统有限公司 | Noun phrase extraction method, system, storage medium and electronic equipment |
CN112580349A (en) * | 2020-12-24 | 2021-03-30 | 竹间智能科技(上海)有限公司 | Phrase extraction method and device and electronic equipment |
CN112651226A (en) * | 2020-09-21 | 2021-04-13 | 深圳前海黑顿科技有限公司 | Knowledge analysis system and method based on dependency syntax tree |
CN112699677A (en) * | 2020-12-31 | 2021-04-23 | 竹间智能科技(上海)有限公司 | Event extraction method and device, electronic equipment and storage medium |
CN112749549A (en) * | 2021-01-22 | 2021-05-04 | 中国科学院电子学研究所苏州研究院 | Chinese entity relation extraction method based on incremental learning and multi-model fusion |
CN112749548A (en) * | 2020-11-02 | 2021-05-04 | 万齐智 | Rule-based Chinese structured financial event default completion extraction method |
CN112784574A (en) * | 2021-02-02 | 2021-05-11 | 网易(杭州)网络有限公司 | Text segmentation method and device, electronic equipment and medium |
CN113361272A (en) * | 2021-06-22 | 2021-09-07 | 海信视像科技股份有限公司 | Method and device for extracting concept words of media asset title |
CN113515630A (en) * | 2021-06-10 | 2021-10-19 | 深圳数联天下智能科技有限公司 | Triple generating and checking method and device, electronic equipment and storage medium |
CN113609838A (en) * | 2021-07-14 | 2021-11-05 | 华东计算技术研究所(中国电子科技集团公司第三十二研究所) | Document information extraction and mapping method and system |
CN113705198A (en) * | 2021-10-21 | 2021-11-26 | 北京达佳互联信息技术有限公司 | Scene graph generation method and device, electronic equipment and storage medium |
CN113743090A (en) * | 2021-09-08 | 2021-12-03 | 度小满科技(北京)有限公司 | Keyword extraction method and device |
CN113761919A (en) * | 2020-06-04 | 2021-12-07 | 国家计算机网络与信息安全管理中心 | Entity attribute extraction method of spoken short text and electronic device |
CN113821605A (en) * | 2021-10-12 | 2021-12-21 | 广州汇智通信技术有限公司 | Event extraction method |
CN113971216A (en) * | 2021-10-22 | 2022-01-25 | 北京百度网讯科技有限公司 | Data processing method and device, electronic equipment and memory |
CN114186552A (en) * | 2021-12-13 | 2022-03-15 | 北京百度网讯科技有限公司 | Text analysis method, device and equipment and computer storage medium |
CN114997398A (en) * | 2022-03-09 | 2022-09-02 | 哈尔滨工业大学 | Knowledge base fusion method based on relation extraction |
CN115062609A (en) * | 2022-08-19 | 2022-09-16 | 北京语言大学 | Method and device for enhancing syntax dependence of Chinese language |
CN115238217A (en) * | 2022-09-23 | 2022-10-25 | 山东省齐鲁大数据研究院 | Method for extracting numerical information from bulletin text and terminal |
CN116361422A (en) * | 2023-06-02 | 2023-06-30 | 深圳得理科技有限公司 | Keyword extraction method, text retrieval method and related equipment |
CN117609518A (en) * | 2024-01-17 | 2024-02-27 | 江西科技师范大学 | Hierarchical Chinese entity relation extraction method and system for centering structure |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2002097662A1 (en) * | 2001-06-01 | 2002-12-05 | Synomia | Method and large syntactical analysis system of a corpus, a specialised corpus in particular |
WO2010050675A2 (en) * | 2008-10-29 | 2010-05-06 | 한국과학기술원 | Method for automatically extracting relation triplets through a dependency grammar parse tree |
CN104933027A (en) * | 2015-06-12 | 2015-09-23 | 华东师范大学 | Open Chinese entity relation extraction method using dependency analysis |
CN107180045A (en) * | 2016-03-10 | 2017-09-19 | 中国科学院地理科学与资源研究所 | A kind of internet text contains the abstracting method of geographical entity relation |
CN107291687A (en) * | 2017-04-27 | 2017-10-24 | 同济大学 | It is a kind of based on interdependent semantic Chinese unsupervised open entity relation extraction method |
-
2018
- 2018-09-26 CN CN201811124153.1A patent/CN109241538B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2002097662A1 (en) * | 2001-06-01 | 2002-12-05 | Synomia | Method and large syntactical analysis system of a corpus, a specialised corpus in particular |
WO2010050675A2 (en) * | 2008-10-29 | 2010-05-06 | 한국과학기술원 | Method for automatically extracting relation triplets through a dependency grammar parse tree |
CN104933027A (en) * | 2015-06-12 | 2015-09-23 | 华东师范大学 | Open Chinese entity relation extraction method using dependency analysis |
CN107180045A (en) * | 2016-03-10 | 2017-09-19 | 中国科学院地理科学与资源研究所 | A kind of internet text contains the abstracting method of geographical entity relation |
CN107291687A (en) * | 2017-04-27 | 2017-10-24 | 同济大学 | It is a kind of based on interdependent semantic Chinese unsupervised open entity relation extraction method |
Non-Patent Citations (2)
Title |
---|
李明耀等: "基于依存分析的开放式中文实体关系抽取方法", 《计算机工程》 * |
翟羽佳等: "基于文本挖掘的中文领域本体构建方法研究", 《情报科学》 * |
Cited By (128)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11568142B2 (en) | 2018-06-04 | 2023-01-31 | Infosys Limited | Extraction of tokens and relationship between tokens from documents to form an entity relationship map |
EP3579120A1 (en) * | 2018-06-04 | 2019-12-11 | Infosys Limited | Extraction of tokens and relationship between tokens from documents to form an entity relationship map |
CN109582975B (en) * | 2019-01-31 | 2023-05-23 | 北京嘉和海森健康科技有限公司 | Named entity identification method and device |
CN109582975A (en) * | 2019-01-31 | 2019-04-05 | 北京嘉和美康信息技术有限公司 | It is a kind of name entity recognition methods and device |
CN110083822B (en) * | 2019-03-06 | 2022-11-15 | 杭州电子科技大学 | Conversion method for converting requirement text into SysML requirement diagram |
CN110083822A (en) * | 2019-03-06 | 2019-08-02 | 杭州电子科技大学 | It is a kind of from demand text conversion to the conversion method of SysML demand figure |
CN109918672A (en) * | 2019-03-13 | 2019-06-21 | 东华大学 | A kind of structuring processing method of the Thyroid ultrasound report based on tree construction |
CN109918672B (en) * | 2019-03-13 | 2023-06-02 | 东华大学 | Structural processing method of thyroid ultrasound report based on tree structure |
CN109992651B (en) * | 2019-03-14 | 2024-01-02 | 广州智语信息科技有限公司 | Automatic identification and extraction method for problem target features |
CN109992651A (en) * | 2019-03-14 | 2019-07-09 | 广州智语信息科技有限公司 | A kind of problem target signature automatic identification and abstracting method |
CN109992777A (en) * | 2019-03-26 | 2019-07-09 | 浙江大学 | A kind of crucial semantic information extracting method of Chinese medicine state of an illness text based on keyword |
CN110309393A (en) * | 2019-03-28 | 2019-10-08 | 平安科技(深圳)有限公司 | Data processing method, device, equipment and readable storage medium storing program for executing |
CN110309393B (en) * | 2019-03-28 | 2023-06-20 | 平安科技(深圳)有限公司 | Data processing method, device, equipment and readable storage medium |
CN109977235A (en) * | 2019-04-04 | 2019-07-05 | 吉林大学 | A kind of determination method and apparatus of trigger word |
CN109977235B (en) * | 2019-04-04 | 2022-10-25 | 吉林大学 | Method and device for determining trigger word |
CN110032649B (en) * | 2019-04-12 | 2021-10-01 | 北京科技大学 | Method and device for extracting relationships between entities in traditional Chinese medicine literature |
CN110032649A (en) * | 2019-04-12 | 2019-07-19 | 北京科技大学 | Relation extraction method and device between a kind of entity of TCM Document |
CN110046351B (en) * | 2019-04-19 | 2022-06-14 | 福州大学 | Text relation extraction method based on features under rule driving |
CN110046351A (en) * | 2019-04-19 | 2019-07-23 | 福州大学 | Text Relation extraction method under regular drive based on feature |
CN110263120A (en) * | 2019-04-26 | 2019-09-20 | 北京零秒科技有限公司 | Corpus labeling method and device |
CN110222332A (en) * | 2019-04-29 | 2019-09-10 | 闽江学院 | The method for realizing name of the dish Entity recognition based on dependency analysis |
CN110222332B (en) * | 2019-04-29 | 2023-06-16 | 闽江学院 | Method for realizing identification of famous entity based on dependency analysis |
CN110110329B (en) * | 2019-04-30 | 2022-05-17 | 湖南星汉数智科技有限公司 | Entity behavior extraction method and device, computer device and computer readable storage medium |
CN110110329A (en) * | 2019-04-30 | 2019-08-09 | 湖南星汉数智科技有限公司 | A kind of entity behavior derivation method, apparatus, computer installation and computer readable storage medium |
CN110162788A (en) * | 2019-05-06 | 2019-08-23 | 三角兽(北京)科技有限公司 | The determination method and device of entity dependence |
CN110162788B (en) * | 2019-05-06 | 2021-02-09 | 腾讯科技(深圳)有限公司 | Entity dependency relationship determination method and device |
CN110119510B (en) * | 2019-05-17 | 2023-02-14 | 浪潮软件集团有限公司 | Relationship extraction method and device based on transfer dependency relationship and structure auxiliary word |
CN110119510A (en) * | 2019-05-17 | 2019-08-13 | 浪潮软件集团有限公司 | A kind of Relation extraction method and device based on transmitting dependence and structural auxiliary word |
CN110196913A (en) * | 2019-05-23 | 2019-09-03 | 北京邮电大学 | Multiple entity relationship joint abstracting method and device based on text generation formula |
CN110263336A (en) * | 2019-06-12 | 2019-09-20 | 东华大学 | A method of building breast ultrasound domain body |
CN110263336B (en) * | 2019-06-12 | 2023-06-23 | 东华大学 | Method for constructing breast ultrasound field ontology |
CN110377901A (en) * | 2019-06-20 | 2019-10-25 | 湖南大学 | A kind of text mining method for making a report on case for distribution line tripping |
CN110263341A (en) * | 2019-06-20 | 2019-09-20 | 贵州电网有限责任公司 | A method of profile is excavated and positioned from text |
CN110263341B (en) * | 2019-06-20 | 2023-06-20 | 贵州电网有限责任公司 | Method for mining and locating personal ability from text |
CN110413732B (en) * | 2019-07-16 | 2023-11-24 | 扬州大学 | Knowledge searching method for software defect knowledge |
CN110413732A (en) * | 2019-07-16 | 2019-11-05 | 扬州大学 | The knowledge searching method of software-oriented defect knowledge |
CN110362673A (en) * | 2019-07-17 | 2019-10-22 | 福州大学 | Computer vision class papers contents method of discrimination and system based on abstract semantic analysis |
CN110362673B (en) * | 2019-07-17 | 2022-07-08 | 福州大学 | Computer vision type thesis content distinguishing method and system based on abstract semantic analysis |
CN110597998A (en) * | 2019-07-19 | 2019-12-20 | 中国人民解放军国防科技大学 | Military scenario entity relationship extraction method and device combined with syntactic analysis |
CN110532553B (en) * | 2019-08-21 | 2023-08-22 | 河海大学 | Water conservancy space relation word recognition and extraction method |
CN110543630A (en) * | 2019-08-21 | 2019-12-06 | 北京仝睿科技有限公司 | Method and device for generating text structured representation and computer storage medium |
CN110502642A (en) * | 2019-08-21 | 2019-11-26 | 武汉工程大学 | A kind of entity relation extraction method based on interdependent syntactic analysis and rule |
CN110502642B (en) * | 2019-08-21 | 2024-01-23 | 武汉工程大学 | Entity relation extraction method based on dependency syntactic analysis and rules |
CN110532553A (en) * | 2019-08-21 | 2019-12-03 | 河海大学 | A kind of method water conservancy spatial relationship word identification and extracted |
CN110555083A (en) * | 2019-08-26 | 2019-12-10 | 北京工业大学 | non-supervision entity relationship extraction method based on zero-shot |
CN110555083B (en) * | 2019-08-26 | 2021-06-25 | 北京工业大学 | Non-supervision entity relationship extraction method based on zero-shot |
CN110569504A (en) * | 2019-09-04 | 2019-12-13 | 北京明略软件系统有限公司 | relation word determining method and device |
CN110569504B (en) * | 2019-09-04 | 2022-11-15 | 北京明略软件系统有限公司 | Relation word determining method and device |
CN110569510A (en) * | 2019-09-17 | 2019-12-13 | 四川长虹电器股份有限公司 | method for identifying named entity of user request data |
CN110705310B (en) * | 2019-09-20 | 2023-07-18 | 北京金山数字娱乐科技有限公司 | Article generation method and device |
CN110705310A (en) * | 2019-09-20 | 2020-01-17 | 北京金山数字娱乐科技有限公司 | Article generation method and device |
CN110765759A (en) * | 2019-10-21 | 2020-02-07 | 普信恒业科技发展(北京)有限公司 | Intention identification method and device |
CN110874396A (en) * | 2019-11-07 | 2020-03-10 | 腾讯科技(深圳)有限公司 | Keyword extraction method and device and computer storage medium |
CN110874396B (en) * | 2019-11-07 | 2024-02-09 | 腾讯科技(深圳)有限公司 | Keyword extraction method and device and computer storage medium |
CN111382571B (en) * | 2019-11-08 | 2023-06-06 | 南方科技大学 | Information extraction method, system, server and storage medium |
CN111382571A (en) * | 2019-11-08 | 2020-07-07 | 南方科技大学 | Information extraction method, system, server and storage medium |
CN110909537A (en) * | 2019-11-19 | 2020-03-24 | 曲英洲 | Artificial intelligence method for modern Chinese component analysis |
CN110991180A (en) * | 2019-11-28 | 2020-04-10 | 同济人工智能研究院(苏州)有限公司 | Command identification method based on keywords and Word2Vec |
CN111177215A (en) * | 2019-12-20 | 2020-05-19 | 京东数字科技控股有限公司 | Method and device for generating financial data |
CN111126052B (en) * | 2019-12-26 | 2023-11-03 | 鼎富智能科技有限公司 | Function point generation method, device, electronic equipment and computer readable storage medium |
CN111126052A (en) * | 2019-12-26 | 2020-05-08 | 中科鼎富(北京)科技发展有限公司 | Function point generation method and device, electronic equipment and computer readable storage medium |
CN111198932A (en) * | 2019-12-30 | 2020-05-26 | 北京明略软件系统有限公司 | Triple acquiring method and device, electronic equipment and readable storage medium |
CN111198932B (en) * | 2019-12-30 | 2023-03-21 | 北京明略软件系统有限公司 | Triple acquiring method and device, electronic equipment and readable storage medium |
CN111178079A (en) * | 2019-12-31 | 2020-05-19 | 北京明略软件系统有限公司 | Triple extraction method and device |
CN111178079B (en) * | 2019-12-31 | 2023-05-26 | 北京明略软件系统有限公司 | Triplet extraction method and device |
CN111241827A (en) * | 2020-01-10 | 2020-06-05 | 同方知网(北京)技术有限公司 | Attribute extraction method based on sentence retrieval mode |
CN111241827B (en) * | 2020-01-10 | 2022-05-20 | 同方知网(北京)技术有限公司 | Attribute extraction method based on sentence retrieval mode |
CN111611399A (en) * | 2020-04-15 | 2020-09-01 | 广发证券股份有限公司 | Information event mapping system and method based on natural language processing |
CN111597349A (en) * | 2020-04-30 | 2020-08-28 | 西安理工大学 | Rail transit standard entity relation automatic completion method based on artificial intelligence |
CN111597349B (en) * | 2020-04-30 | 2022-10-11 | 西安理工大学 | Rail transit standard entity relation automatic completion method based on artificial intelligence |
CN111597812A (en) * | 2020-05-09 | 2020-08-28 | 北京合众鼎成科技有限公司 | Financial field multiple relation extraction method based on mask language model |
CN111597794A (en) * | 2020-05-11 | 2020-08-28 | 浪潮软件集团有限公司 | Dependency relationship-based 'yes' word and sentence relationship extraction method and device |
CN111597794B (en) * | 2020-05-11 | 2023-06-06 | 浪潮软件集团有限公司 | Dependency relationship-based 'Yes' word and sentence relationship extraction method and device |
CN111597351A (en) * | 2020-05-14 | 2020-08-28 | 上海德拓信息技术股份有限公司 | Visual document map construction method |
CN111639499B (en) * | 2020-06-01 | 2023-06-16 | 北京中科汇联科技股份有限公司 | Composite entity extraction method and system |
CN111639499A (en) * | 2020-06-01 | 2020-09-08 | 北京中科汇联科技股份有限公司 | Composite entity extraction method and system |
CN113761919A (en) * | 2020-06-04 | 2021-12-07 | 国家计算机网络与信息安全管理中心 | Entity attribute extraction method of spoken short text and electronic device |
CN111666738A (en) * | 2020-06-09 | 2020-09-15 | 南京师范大学 | Formalized coding method for motion description natural text |
CN111680492A (en) * | 2020-06-10 | 2020-09-18 | 创新奇智(青岛)科技有限公司 | New word mining method and device and electronic equipment |
CN111666767A (en) * | 2020-06-10 | 2020-09-15 | 创新奇智(上海)科技有限公司 | Data identification method and device, electronic equipment and storage medium |
CN111666767B (en) * | 2020-06-10 | 2023-07-18 | 创新奇智(上海)科技有限公司 | Data identification method and device, electronic equipment and storage medium |
CN111798847A (en) * | 2020-06-22 | 2020-10-20 | 广州小鹏车联网科技有限公司 | Voice interaction method, server and computer-readable storage medium |
CN112052340A (en) * | 2020-08-10 | 2020-12-08 | 深圳数联天下智能科技有限公司 | Data model construction method and device and electronic equipment |
CN111984778B (en) * | 2020-09-08 | 2022-06-03 | 四川长虹电器股份有限公司 | Dependency syntax analysis and Chinese grammar-based multi-round semantic analysis method |
CN111984778A (en) * | 2020-09-08 | 2020-11-24 | 四川长虹电器股份有限公司 | Dependency syntax analysis and Chinese grammar-based multi-round semantic analysis method |
CN112131343B (en) * | 2020-09-14 | 2023-07-07 | 新讯数字科技(杭州)有限公司 | Method for identifying characters in Chinese novel dialogue |
CN112131343A (en) * | 2020-09-14 | 2020-12-25 | 杭州东信北邮信息技术有限公司 | Chinese novel dialect dialogue character recognition method |
CN112084793A (en) * | 2020-09-14 | 2020-12-15 | 深圳前海微众银行股份有限公司 | Semantic recognition method, device and readable storage medium based on dependency syntax |
CN112651226B (en) * | 2020-09-21 | 2022-03-29 | 深圳前海黑顿科技有限公司 | Knowledge analysis system and method based on dependency syntax tree |
CN112651226A (en) * | 2020-09-21 | 2021-04-13 | 深圳前海黑顿科技有限公司 | Knowledge analysis system and method based on dependency syntax tree |
CN112148838B (en) * | 2020-09-23 | 2024-04-19 | 北京中电普华信息技术有限公司 | Service source object extraction method and device |
CN112148838A (en) * | 2020-09-23 | 2020-12-29 | 北京中电普华信息技术有限公司 | Business source object extraction method and device |
CN112183059A (en) * | 2020-09-24 | 2021-01-05 | 万齐智 | Chinese structured event extraction method |
CN112749548A (en) * | 2020-11-02 | 2021-05-04 | 万齐智 | Rule-based Chinese structured financial event default completion extraction method |
CN112749548B (en) * | 2020-11-02 | 2024-04-26 | 万齐智 | Rule-based default completion extraction method for Chinese structured financial events |
CN112364648A (en) * | 2020-12-02 | 2021-02-12 | 中金智汇科技有限责任公司 | Keyword extraction method and device, electronic equipment and storage medium |
CN112560488A (en) * | 2020-12-07 | 2021-03-26 | 北京明略软件系统有限公司 | Noun phrase extraction method, system, storage medium and electronic equipment |
CN112380868A (en) * | 2020-12-10 | 2021-02-19 | 广东泰迪智能科技股份有限公司 | Petition-purpose multi-classification device based on event triples and method thereof |
CN112380868B (en) * | 2020-12-10 | 2024-02-13 | 广东泰迪智能科技股份有限公司 | Multi-classification device and method for interview destination based on event triplets |
CN112580349A (en) * | 2020-12-24 | 2021-03-30 | 竹间智能科技(上海)有限公司 | Phrase extraction method and device and electronic equipment |
CN112580349B (en) * | 2020-12-24 | 2023-09-29 | 竹间智能科技(上海)有限公司 | Phrase extraction method and device and electronic equipment |
CN112699677B (en) * | 2020-12-31 | 2023-05-02 | 竹间智能科技(上海)有限公司 | Event extraction method and device, electronic equipment and storage medium |
CN112699677A (en) * | 2020-12-31 | 2021-04-23 | 竹间智能科技(上海)有限公司 | Event extraction method and device, electronic equipment and storage medium |
CN112749549A (en) * | 2021-01-22 | 2021-05-04 | 中国科学院电子学研究所苏州研究院 | Chinese entity relation extraction method based on incremental learning and multi-model fusion |
CN112749549B (en) * | 2021-01-22 | 2023-10-13 | 中国科学院电子学研究所苏州研究院 | Chinese entity relation extraction method based on incremental learning and multi-model fusion |
CN112784574B (en) * | 2021-02-02 | 2023-09-15 | 网易(杭州)网络有限公司 | Text segmentation method and device, electronic equipment and medium |
CN112784574A (en) * | 2021-02-02 | 2021-05-11 | 网易(杭州)网络有限公司 | Text segmentation method and device, electronic equipment and medium |
CN113515630B (en) * | 2021-06-10 | 2024-04-09 | 深圳数联天下智能科技有限公司 | Triplet generation and verification method and device, electronic equipment and storage medium |
CN113515630A (en) * | 2021-06-10 | 2021-10-19 | 深圳数联天下智能科技有限公司 | Triple generating and checking method and device, electronic equipment and storage medium |
CN113361272A (en) * | 2021-06-22 | 2021-09-07 | 海信视像科技股份有限公司 | Method and device for extracting concept words of media asset title |
CN113361272B (en) * | 2021-06-22 | 2023-03-21 | 海信视像科技股份有限公司 | Method and device for extracting concept words of media asset title |
CN113609838A (en) * | 2021-07-14 | 2021-11-05 | 华东计算技术研究所(中国电子科技集团公司第三十二研究所) | Document information extraction and mapping method and system |
CN113743090B (en) * | 2021-09-08 | 2024-04-12 | 度小满科技(北京)有限公司 | Keyword extraction method and device |
CN113743090A (en) * | 2021-09-08 | 2021-12-03 | 度小满科技(北京)有限公司 | Keyword extraction method and device |
CN113821605A (en) * | 2021-10-12 | 2021-12-21 | 广州汇智通信技术有限公司 | Event extraction method |
CN113705198A (en) * | 2021-10-21 | 2021-11-26 | 北京达佳互联信息技术有限公司 | Scene graph generation method and device, electronic equipment and storage medium |
CN113971216A (en) * | 2021-10-22 | 2022-01-25 | 北京百度网讯科技有限公司 | Data processing method and device, electronic equipment and memory |
CN114186552A (en) * | 2021-12-13 | 2022-03-15 | 北京百度网讯科技有限公司 | Text analysis method, device and equipment and computer storage medium |
CN114186552B (en) * | 2021-12-13 | 2023-04-07 | 北京百度网讯科技有限公司 | Text analysis method, device and equipment and computer storage medium |
CN114997398A (en) * | 2022-03-09 | 2022-09-02 | 哈尔滨工业大学 | Knowledge base fusion method based on relation extraction |
CN114997398B (en) * | 2022-03-09 | 2023-05-26 | 哈尔滨工业大学 | Knowledge base fusion method based on relation extraction |
CN115062609A (en) * | 2022-08-19 | 2022-09-16 | 北京语言大学 | Method and device for enhancing syntax dependence of Chinese language |
CN115062609B (en) * | 2022-08-19 | 2022-12-09 | 北京语言大学 | Method and device for enhancing syntax dependence of Chinese language |
CN115238217A (en) * | 2022-09-23 | 2022-10-25 | 山东省齐鲁大数据研究院 | Method for extracting numerical information from bulletin text and terminal |
CN116361422B (en) * | 2023-06-02 | 2023-09-19 | 深圳得理科技有限公司 | Keyword extraction method, text retrieval method and related equipment |
CN116361422A (en) * | 2023-06-02 | 2023-06-30 | 深圳得理科技有限公司 | Keyword extraction method, text retrieval method and related equipment |
CN117609518A (en) * | 2024-01-17 | 2024-02-27 | 江西科技师范大学 | Hierarchical Chinese entity relation extraction method and system for centering structure |
CN117609518B (en) * | 2024-01-17 | 2024-04-26 | 江西科技师范大学 | Hierarchical Chinese entity relation extraction method and system for centering structure |
Also Published As
Publication number | Publication date |
---|---|
CN109241538B (en) | 2022-12-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109241538A (en) | Based on the interdependent Chinese entity relation extraction method of keyword and verb | |
US10496756B2 (en) | Sentence creation system | |
CN110502642B (en) | Entity relation extraction method based on dependency syntactic analysis and rules | |
CN102662936B (en) | Chinese-English unknown words translating method blending Web excavation, multi-feature and supervised learning | |
Pourvali et al. | Automated text summarization base on lexicales chain and graph using of wordnet and wikipedia knowledge base | |
CN109960756A (en) | Media event information inductive method | |
CN109522418A (en) | A kind of automanual knowledge mapping construction method | |
Sahu et al. | Prashnottar: a Hindi question answering system | |
CN110851714A (en) | Text recommendation method and system based on heterogeneous topic model and word embedding model | |
Lynn et al. | An improved method of automatic text summarization for web contents using lexical chain with semantic-related terms | |
Verma et al. | A new approach for idiom identification using meanings and the web | |
CN103678287A (en) | Method for unifying keyword translation | |
Bella et al. | Domain-based sense disambiguation in multilingual structured data | |
Adhitama et al. | Topic labeling towards news document collection based on Latent Dirichlet Allocation and ontology | |
Abend et al. | Fully unsupervised core-adjunct argument classification | |
Ogrodniczuk et al. | Rule-based coreference resolution module for Polish | |
Ung et al. | Combination of features for vietnamese news multi-document summarization | |
Tran et al. | A named entity recognition approach for tweet streams using active learning | |
CN107480142B (en) | Method for extracting evaluation object based on dependency relationship | |
Ibrahiem et al. | FEATURE EXTRACTION ENCHANCEMENT IN USERS’ATTITUDE DETECTION | |
Sahin | Classification of turkish semantic relation pairs using different sources | |
Ji et al. | Measurement of sentence similarity based on constituency parsing and dilated convolution | |
Wu et al. | Fine-grained product feature extraction in chinese reviews | |
Ding et al. | Building chinese event type paradigm based on trigger clustering | |
AU2021105953A4 (en) | Method for fine-grained domain terminology self-learning based on contextual semantics |
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 |