CN110263083A - Processing method, device, equipment and the medium of knowledge mapping - Google Patents
Processing method, device, equipment and the medium of knowledge mapping Download PDFInfo
- Publication number
- CN110263083A CN110263083A CN201910537133.5A CN201910537133A CN110263083A CN 110263083 A CN110263083 A CN 110263083A CN 201910537133 A CN201910537133 A CN 201910537133A CN 110263083 A CN110263083 A CN 110263083A
- Authority
- CN
- China
- Prior art keywords
- unit
- knowledge mapping
- evidence unit
- candidate
- entity evidence
- 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
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/26—Visual data mining; Browsing structured data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
Abstract
The embodiment of the invention discloses a kind of processing method of knowledge mapping, device, equipment and media, are related to knowledge mapping technical field.This method comprises: selecting at least two candidate entity evidence units from the entity evidence unit in knowledge mapping said target to be verified field according to knowledge mapping to be verified；Determine the matching degree of the knowledge mapping to be tested and at least two candidate entity evidence unit；According to the matching degree, the target entity evidence unit of the knowledge mapping to be tested is selected from described at least two candidate entity evidence units, for verifying based on the target entity evidence unit to the knowledge mapping to be verified.The embodiment of the present invention from target domain entity evidence unit by selecting at least two candidate entity evidence units, and determine the matching degree of candidate entity evidence unit and knowledge mapping to be tested, according to matching degree selection target entity evidence unit, reduce manpower intervention, improves the accuracy rate and building efficiency of knowledge mapping.
Description
Technical field
The present embodiments relate to Intelligent logistics technical field more particularly to a kind of processing method of knowledge mapping, device,
Equipment and medium.
Background technique
Construct an authority accurately comprehensive medical industry knowledge mapping be many upper layer applications basic data demand, power
Medical knowledge in prestige medical treatment books is the wisdom crystallization that people summarize and proved, and therefrom people can excavate many authoritys
Medical facts.
Existing technology wants the medical facts excavated from medical books to be accurately added in medical map, one
As method be that preliminary screening is carried out by true frequency information, then audited using medical expert.Such method because
It is manpower intervention, there is a problem that human input is big, relatively inefficient.And medical expert audits the process of the medicine fact,
In addition to relying on experience, it is also required for a books query facility in many cases and carries out inquiry confirmation.
Summary of the invention
The embodiment of the present invention provides processing method, device, equipment and the medium of a kind of knowledge mapping, to solve the prior art
Big, the relatively inefficient problem of human input when constructing knowledge mapping.
In a first aspect, the embodiment of the invention provides a kind of processing methods of knowledge mapping, which comprises
According to knowledge mapping to be verified, from the entity evidence unit in knowledge mapping said target to be verified field selection to
Few two candidate entity evidence units；
Determine the matching degree of the knowledge mapping to be tested and at least two candidate entity evidence unit；
According to the matching degree, the knowledge mapping to be tested is selected from described at least two candidate entity evidence units
Target entity evidence unit, for being verified to the knowledge mapping to be verified based on the target entity evidence unit.
Second aspect, the embodiment of the invention provides a kind of processing unit of knowledge mapping, described device includes:
Candidate entity evidence Unit selection module is used for according to knowledge mapping to be verified, belonging to knowledge mapping to be verified
The candidate entity evidence unit of selection at least two in the entity evidence unit of target domain；
Matching degree determining module, for determining the knowledge mapping to be tested and at least two candidate entity evidence list
The matching degree of member；
Target entity evidence Unit selection module, for being demonstrate,proved from described at least two candidate entities according to the matching degree
According to the target entity evidence unit for selecting the knowledge mapping to be tested in unit, for being based on the target entity evidence unit pair
The knowledge mapping to be verified is verified.
The third aspect, the embodiment of the invention provides a kind of equipment, the equipment further include:
One or more processors；
Storage device, for storing one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processing
Device realizes a kind of processing method of knowledge mapping as described in any in the embodiment of the present invention.
Fourth aspect, the embodiment of the invention provides a kind of computer-readable mediums, are stored thereon with computer program, should
A kind of processing method of knowledge mapping as described in any in the embodiment of the present invention is realized when program is executed by processor.
The embodiment of the present invention by selecting at least two candidate entity evidence units from target domain entity evidence unit,
And determine the matching degree of at least two candidate entity evidence units and knowledge mapping to be tested, according to matching degree selection target entity
Evidence unit improves the accurate of knowledge mapping for verifying according to target entity evidence unit to knowledge mapping to be verified
Rate and building efficiency.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be to needed in the embodiment attached
Figure is briefly described, it should be understood that the following drawings illustrates only certain embodiments of the present invention, therefore is not construed as pair
The restriction of range for those of ordinary skill in the art without creative efforts, can also be according to this
A little attached drawings obtain other relevant attached drawings.
Fig. 1 is a kind of flow chart of the processing method for knowledge mapping that the embodiment of the present invention one provides；
Fig. 2 is a kind of flow chart of the processing method of knowledge mapping provided by Embodiment 2 of the present invention；
Fig. 3 is a kind of flow chart of the processing method for knowledge mapping that the embodiment of the present invention three provides；
Fig. 4 is a kind of structural schematic diagram of the processing unit for knowledge mapping that the embodiment of the present invention four provides；
Fig. 5 is a kind of structural schematic diagram for equipment that the embodiment of the present invention five provides.
Specific embodiment
The embodiment of the present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this
Locate described specific embodiment and is used only for explaining the embodiment of the present invention, rather than limitation of the invention.It further needs exist for
Bright, only parts related to embodiments of the present invention are shown for ease of description, in attached drawing rather than entire infrastructure.
Embodiment one
Fig. 1 is a kind of flow chart of the processing method for knowledge mapping that the embodiment of the present invention one provides.The present embodiment is applicable in
In construct medical knowledge map the case where, this method can be held by the processing unit of knowledge mapping provided in an embodiment of the present invention
Row, the device can be realized by the way of software and/or hardware.As shown in Figure 1, this method may include:
S101, according to knowledge mapping to be verified, from the entity evidence unit in knowledge mapping said target to be verified field
The candidate entity evidence unit of selection at least two.
Wherein, knowledge mapping fields to be verified optionally include medical treatment, education, finance and culture etc..Entity evidence
Unit is excavated by the knowledge fact in knowledge mapping said target to be verified field, effect be embodied in as
Lower two aspects: 1, can be used as the evidence of knowledge mapping to be verified, to assist related auditor to knowledge mapping to be verified
It is assessed；2, it can be indexed in knowledge mapping to be verified, to increase the comprehensive of knowledge mapping to be verified.
Optionally, S (physical name), P (attribute-name) and the O (attribute value) in knowledge mapping to be verified are regard as search term
It is scanned in the entity evidence unit of target domain, according to the candidate entity evidence unit of search result selection at least two.
The specification of entity evidence unit is changeable, when the specification of entity evidence unit is larger, such as one complete
Document, a complete paper or a thick and heavy books, since it may be corresponding comprising multiple entities and multiple entities
More attributes, this may occur with the more attributes of single entity or the multiattribute information interference problem of multiple entity.
In order to avoid this information interference problem, optionally before S101, further includes:
A, chapter title information belonging to attribute information and attribute information is extracted from the target domain fact；
Wherein, attribute information includes a kind of verbal description multiattribute to entity comprising Property Name and property content,
Chapter title information includes the verbal description with attribute information associated entity.
Illustratively, with a books html (hypertext markup language, HyperText Markup Language) page
For face, mainly there is two large divisions, first part is the content blocks for including the information such as a description piece, chapter, section, title；Second
Dividing is the content blocks for including description details.Specifically, in first part by including that target keyword searching method extracts
Chapter title information belonging to attribute information, such as " third piece laryngology ", " chapter 4 Specific pharyngitis " and " third section larynx plum
Poison " etc.；Second part by include target keyword searching method extract attribute information, such as " clinical manifestation trachyphonia ", " examine
It is disconnected to put family history " or " therapeutic scheme and principle antisyphilitic treatment " etc..
B, using the chapter title information as the title of entity evidence unit in target domain；
Illustratively, the chapter title information of extraction is " third piece laryngology ", " chapter 4 Specific pharyngitis " and " third
Save syphilis of larynx ", then " third piece laryngology ", " chapter 4 Specific pharyngitis " and " third section syphilis of larynx " is used as entity evidence list
The title of member.
C, using the Property Name in the attribute information as the attribute keyword of entity evidence unit in target domain；
Wherein, the Property Name in attribute information is the upperseat concept of property content.
Illustratively, such as the attribute information of extraction includes " diagnostic point family history ", then Property Name is that " diagnosis is wanted
Point ", and then the attribute keyword by " diagnostic point " as entity evidence unit；In another example the attribute information extracted includes " facing
Bed performance trachyphonia ", then Property Name is " clinical manifestation ", and then " clinical manifestation " is crucial as the attribute of entity evidence unit
Word.
D, using the property content in the attribute information as the attribute key assignments of entity evidence unit in target domain.
Wherein, the property content in attribute information is the Property Name bottom form of expression.
Illustratively, such as the attribute information of extraction includes " diagnostic point family history ", then property content is " family history ",
And then the attribute key assignments by " family history " as entity evidence unit；In another example the attribute information extracted includes " clinical manifestation sound
Neigh ", then property content is " trachyphonia ", and then the attribute key assignments by " trachyphonia " as entity evidence unit.
By extracting chapter title information belonging to attribute information and attribute information from the target domain fact, and it is final true
The title, attribute keyword and attribute key assignments for determining entity evidence unit, avoid information interference problem, to be tested know so that subsequent
Know map to be matched at least two candidate entity evidence units, it is simpler, effective and intuitive.
By selecting at least two candidate entities from the entity evidence unit in knowledge mapping said target to be verified field
Evidence unit reduces the matching times of subsequent knowledge mapping to be tested Yu entity evidence unit, improves efficiency.
S102, the matching degree for determining the knowledge mapping to be tested and at least two candidate entity evidence unit.
Wherein, matching degree embodies the similitude of knowledge mapping to be tested and at least two candidate entity evidence units,
Higher with spending, then the similitude of knowledge mapping to be tested and at least two candidate entity evidence units is higher, and matching degree is lower, then
Knowledge mapping to be tested is lower with the similitude of at least two candidate entity evidence units.
It is determining specifically, determining the matching degree of knowledge mapping to be tested and at least two candidate entity evidence units
Similarity between the information for including in knowledge mapping to be tested and the information for including at least two candidate entity evidence units.
S103, according to the matching degree, select described to be tested to know from described at least two candidate entity evidence units
The target entity evidence unit for knowing map, for being tested based on the target entity evidence unit the knowledge mapping to be verified
Card.
Optionally, it is ranked up according to matching degree, according to ranking results, determines that the target of the knowledge mapping to be tested is real
Body evidence unit.
The embodiment of the present invention by selecting at least two candidate entity evidence units from target domain entity evidence unit,
And determine the matching degree of at least two candidate entity evidence units and knowledge mapping to be tested, according to matching degree selection target entity
Evidence unit reduces manpower intervention, improves the verification efficiency and building efficiency of knowledge mapping.
Embodiment two
Fig. 2 is a kind of flow chart of the processing method of knowledge mapping provided by Embodiment 2 of the present invention.The present embodiment is upper
It states embodiment one and provides a kind of specific implementation, as shown in Fig. 2, this method may include:
S201, using the physical name in knowledge mapping to be verified as search term, in the entity evidence unit of target domain
It is scanned in chapter title information, attribute keyword and attribute key assignments, obtains first instance evidence unit.
Specifically, in order to not miss possible candidate entity evidence unit as far as possible, by chapter title information, attribute keyword
Or there is the entity evidence unit same or similar with the physical name in knowledge mapping to be verified in attribute key assignments, it is real as first
Body evidence unit.
Illustratively, the entity in knowledge mapping to be verified is entitled " Specific pharyngitis ", i.e. search term, entity evidence unit
It include " Specific pharyngitis " that then entity evidence unit 1 is first instance evidence unit in 1 chapter title information；Entity evidence
It include " Specific pharyngitis " that then entity evidence unit 2 is first instance evidence unit in the attribute key assignments of unit 2；Entity evidence
All not comprising same or similar with " Specific pharyngitis " in the chapter title information of unit 3, attribute keyword or attribute key assignments
Word, then entity evidence unit 3 is not first instance evidence unit.
It can also know to be verified optionally in S201 to not miss possible candidate entity evidence unit as far as possible
" alias " and " also known as " for knowing the physical name in map is used as search term.Wherein, " alias " and the collect means of " also known as " include:
1) existing spectrum data is utilized；2) " alias " and " also known as " occurred in books or document is obtained, it is manually being examined
Core.
S202, using the attribute value in knowledge mapping to be verified as search term, in the entity evidence unit of target domain
It is scanned in attribute key assignments, obtains second instance evidence unit.
Specifically, will there is the entity evidence same or similar with the attribute value in knowledge mapping to be verified in attribute key assignments
Unit, as second instance evidence unit.
Illustratively, the attribute value in knowledge mapping to be verified is " cough ", i.e. search term, the category of entity evidence unit 4
Property key assignments in include " cough ", then entity evidence unit 4 be second instance evidence unit；The attribute key assignments of entity evidence unit 5
In do not include " cough ", then entity evidence unit 5 is not second instance evidence unit.
S203, selection at least two is candidate real from the first instance evidence unit and the second instance evidence unit
Body evidence unit.
Specifically, selecting at least two from first instance evidence unit and second instance evidence unit according to preset rules
Candidate entity evidence unit.
Optionally, assign chapter title information, attribute keyword the weighted value different with attribute key assignments, optional weight is big
Small relationship are as follows: chapter title information weight > attribute Keyword Weight > attribute key assignments weight, according to above-mentioned power after the completion of search
Weight size relation is ranked up obtained first instance evidence unit and second instance evidence unit, from big to small according to weight
The first instance evidence unit and second instance evidence unit for choosing preset quantity are as candidate entity evidence unit.
Illustratively, the chapter title information of first instance evidence unit A includes the physical name in knowledge mapping to be verified,
The attribute keyword of first instance evidence unit B includes the physical name in knowledge mapping to be verified, first instance evidence unit C's
Attribute key assignments includes the physical name in knowledge mapping to be verified, then according to chapter title information, attribute keyword and attribute key assignments
Weight size relation, the sequence of first instance evidence unit A, first instance evidence unit B and first instance evidence unit C
Are as follows: first instance evidence unit A, first instance evidence unit B and first instance evidence unit C.
Optionally, on the basis of the above embodiments, assign physical name as search term and attribute value as search term not
Same weight, optional weight size relation are as follows: physical name is as search term weight > attribute value as search term weight, search
Obtained first instance evidence unit and second instance evidence unit are ranked up according to above-mentioned weight size relation after the completion,
The first instance evidence unit and second instance evidence unit for choosing preset quantity from big to small according to weight are as candidate entity
Evidence unit.
Illustratively, the chapter title information of first instance evidence unit A includes the physical name in knowledge mapping to be verified,
The attribute keyword of first instance evidence unit B includes the physical name in knowledge mapping to be verified, second instance evidence unit C's
Attribute key assignments includes the attribute value in knowledge mapping to be verified, then according to chapter title information, attribute keyword and attribute key assignments
Weight size relation and physical name as search term and weight size relation of the attribute value as search term, first instance card
According to the sequence of unit A, first instance evidence unit B and second instance evidence unit C are as follows:: first instance evidence unit A,
One entity evidence unit B and second instance evidence unit C.
S204, according to the mapping relations in attribute-name in knowledge mapping and entity evidence unit between attribute keyword, with
And knowledge mapping to be verified, selection is matched with the knowledge mapping to be verified from described at least two candidate entity evidence units
, and filter out other candidate entity evidence units.
Wherein, the mapping relations embody attribute-name and entity evidence unit in knowledge graph candidate's entity evidence unit spectrum
The incidence relation of middle attribute keyword.
Optionally, by entity evidence unit attribute keyword root according to include hot topic degree be ranked up, selection hot topic degree compared with
The attribute keyword of high preset quantity, by determining attribute-name in knowledge mapping including semantic analysis algorithm and choosing default
Mapping relations between the attribute keyword of quantity, illustratively, attribute-name " clinical manifestation " and entity evidence in knowledge mapping
Attribute keyword " sign performance " in unit, " physical manifestations " there are mapping relations；Attribute-name " diagnostic point " in knowledge mapping
With attribute keyword in entity evidence unit " diagnosis emphasis ", " diagnosis key point " there are mapping relations, finally by what is obtained
There are mapping relations with knowledge mapping to be verified in the candidate entity evidence unit of mapping relations selection at least two, as candidate
Entity evidence unit, and filter out the candidate entity evidence unit there is no mapping relations.
S205, the matching degree for determining the knowledge mapping to be tested and at least two candidate entity evidence unit.
Candidate entity evidence unit is screened since S204 has passed through this dimension of attribute-name.Determining matching degree
When mainly consider physical name and attribute value the two dimensions.Optionally, S205 includes the following: in determining knowledge mapping to be tested
The title similarity between title in physical name, with candidate entity evidence unit；Determine the attribute in knowledge mapping to be tested
Value, with the key assignments similarity between the attribute key assignments in candidate entity evidence unit；According to the candidate entity evidence unit
Title similarity and key assignments similarity determine the matching degree of candidate's entity evidence unit.
Wherein, similarity can be by including determining or true according to text coincidence relation according to the method for prediction model
It is fixed.Specifically, being determined candidate real according to the title similarity of candidate entity evidence unit and key assignments similarity and preset rules
The matching degree of body evidence unit.Optionally, according to the sum of the title similarity of candidate entity evidence unit and key assignments similarity, really
The matching degree of fixed candidate's entity evidence unit.It is demonstrate,proved by the determination knowledge mapping to be tested and at least two candidate entity
According to the matching degree of unit, data basis has been established for the target entity evidence unit of the subsequently selected knowledge mapping to be tested.
S206, according to the matching degree, select described to be tested to know from described at least two candidate entity evidence units
The target entity evidence unit for knowing map, for being tested based on the target entity evidence unit the knowledge mapping to be verified
Card.
The embodiment of the present invention by using in knowledge mapping to be verified physical name and attribute value as search term, in target
It is scanned in the entity evidence unit in field, respectively obtains first instance evidence unit and second instance evidence unit, then
According to the mapping relations in attribute-name in knowledge mapping and entity evidence unit between attribute keyword, select final candidate real
Body evidence unit realizes to obtain the effect of candidate entity evidence unit, for the determination knowledge mapping to be tested and it is described extremely
The matching degree of few two candidate entity evidence units is laid a good foundation.
Embodiment three
Fig. 3 is a kind of flow chart of the processing method for knowledge mapping that the embodiment of the present invention three provides.The present embodiment is upper
It states embodiment one and provides a kind of specific implementation, as shown in figure 3, this method may include:
S301, according to knowledge mapping to be verified, from the entity evidence unit in knowledge mapping said target to be verified field
The candidate entity evidence unit of selection at least two.
S302, physical name in knowledge mapping to be tested is determined, with the mark between the title in candidate entity evidence unit
Inscribe similarity.
Optionally, respectively by according to the method for prediction model and the method for determining text coincidence relation, come determine to
The physical name in knowledge mapping is examined, with the first title similarity and second between the title in candidate entity evidence unit
Title similarity.
Optionally, S302 includes:
A, by the title in the physical name and candidate entity evidence unit in the knowledge mapping to be tested, as prediction mould
The input of type obtains the first title similarity.
Wherein, prediction model is a kind of deep learning model, and optional prediction model includes pairwise model.
Optionally, the training process of prediction model includes: S, P and the O and S that will be obtained when being excavated based on books map
With the place sentence of O, as positive example；It recycles some heuristic rules to construct some counter-examples, then positive example will be obtained and counter-example is defeated
Enter to include the two-way shot and long term memory network model of attention mechanism or depth structure shot and long term memory network model, in turn
Training obtains prediction model.
B, according to the text weight between the physical name of the knowledge mapping to be tested and the title of candidate entity evidence unit
Conjunction relationship determines the second title similarity.
Wherein, text coincidence relation embodies the similarity degree between text.Optionally, the second title similarity can lead to
Following manner is crossed to determine:
1) zero, such as F1=0, F2=0, F3=0 and F4=0 are set by the numerical value of four parameters.
2) if the physical name of knowledge mapping to be tested is identical with the title of candidate entity evidence unit, F1=is set
1, F2=0, F3=0 and F4=0.
3) if the physical name of knowledge mapping to be tested and the title of candidate entity evidence unit are not exactly the same and to be tested
The physical name of knowledge mapping is the suffix of the title of candidate entity evidence unit, then F1=0, F2=1, F3=0 and F4 is arranged
=0.
4) if the physical name of knowledge mapping to be tested and the title of candidate entity evidence unit are not exactly the same, to be tested know
The physical name of knowledge map is not the suffix of the title of candidate entity evidence unit, and the title of candidate entity evidence unit is to be checked
The suffix for testing the physical name of knowledge mapping, then be arranged F1=0, F2=0, F3=1 and F4=0.
5) if the physical name of knowledge mapping to be tested and the title of candidate entity evidence unit are not exactly the same, to be tested know
Know map physical name be not candidate entity evidence unit title suffix, and the title of candidate entity evidence unit be not to
The suffix for examining the physical name of knowledge mapping, then be arranged F1=0, F2=0, F3=0 and F4=1.
6) the second title similarity=F1 × a1+F2 × a2+F3 × a3+F4 × jaccard × a4
Wherein, a1, a2, a3 and a4 are fixed constants, can adjust according to the actual situation, be preferable to provide a1=1, a2=
0.95, a3=0.92 and a4=0.9.Jaccard is Jie Kade similarity, calculating process are as follows: 1) knows respectively by be tested
The title of the physical name and candidate entity evidence unit of knowing map segments；2) number of the two same words after segmenting is calculated；3) will
Ratio both after participle between the number of same words and min (physical name segments number, and title segments number), as
The value of jaccard.
S303, attribute value in knowledge mapping to be tested is determined, between the attribute key assignments in candidate entity evidence unit
Key assignments similarity.
Optionally, respectively by according to the method for prediction model and the method for determining text coincidence relation, come determine to
Examine the attribute value in knowledge mapping, the first key assignments similarity between the attribute key assignments in candidate entity evidence unit and
Second key assignments similarity.
Optionally, S303 includes:
A, by the attribute key assignments in the attribute value and candidate entity evidence unit in the knowledge mapping to be tested, as pre-
The input for surveying model, obtains the first key assignments similarity；
B, according to the text between the attribute value of the knowledge mapping to be tested and the attribute key assignments of candidate entity evidence unit
This coincidence relation determines the second key assignments similarity.
If the title similarity of S304, any candidate entity evidence unit is less than the first title similarity threshold, or is somebody's turn to do
The key assignments similarity of candidate entity evidence unit then filters out candidate's entity evidence unit less than the first key assignments similarity threshold.
Specifically, the first title similarity and the second title similarity that include by title similarity, respectively with corresponding threshold
Value compares；The the first key assignments similarity and the second key assignments similarity for including by key assignments similarity, compare with corresponding threshold value respectively, filter
Except ineligible candidate entity evidence unit.
Optionally, if the first title similarity of any candidate's entity evidence unit is less than first title the first similarity
Threshold value and the second title similarity are less than the sub- threshold value of first the second similarity of title or the first key assignments similarity less than the first key
It is real then to filter out the candidate less than the first sub- threshold value of the second similarity of key assignments for the sub- threshold value of the first similarity of value and the second key assignments similarity
Body evidence unit.
Illustratively, for example, be arranged the first sub- threshold value of the first similarity of title be 0.9, the first sub- threshold of the second similarity of title
Value is 0.85, and the first title similarity of candidate entity evidence unit A is 0.95, and the second title similarity is 0.8, then candidate real
Body evidence unit A should be filtered out；In another example the first sub- threshold value 0.9 of the first similarity of key assignments of setting, the first key assignments second are similar
Sub- threshold value 0.85 is spent, the first key assignments similarity of candidate entity evidence unit B is 0.9, and the second key assignments similarity is 0.9, then waits
Entity evidence unit B is selected not to be filtered out.
If the title similarity of S305, the candidate entity evidence unit are greater than the second title similarity threshold, and key assignments
Similarity is greater than the second key assignments similarity threshold, then adds the first flag bit for the candidate entity evidence unit；It otherwise, is institute
It states candidate entity evidence unit and adds the second flag bit.
Optionally, maximum similar in the first title similarity and the second title similarity for including by title similarity
Degree, compares with the second title similarity threshold；The the first key assignments similarity and the second key assignments similarity for including by key assignments similarity
In maximum similarity, compared with the second key assignments similarity threshold；If in the first title similarity and the second title similarity most
Big similarity is greater than the second title similarity threshold, and maximum similar in the first key assignments similarity and the second key assignments similarity
Degree is greater than the second key assignments similarity threshold, then adds the first flag bit for the candidate entity evidence unit；It otherwise, is the time
Entity evidence unit is selected to add the second flag bit.
Illustratively, the second title similarity threshold is set and the second key assignments similarity threshold is all 0.95, candidate entity
First title similarity of evidence unit B is 0.97, and the second title similarity is 0.9, and the first key assignments similarity is 0.96, second
Key assignments similarity is maximum 0.97 the second title of > of similarity in 0.93, then the first title similarity and the second title similarity
Maximum 0.96 the second key assignments of > of similarity is similar in similarity threshold 0.95, the first key assignments similarity and the second key assignments similarity
Threshold value 0.95 is spent, then adds the first flag bit for candidate entity evidence unit B, preferred flag bit is " 1 ".
Illustratively, the second title similarity threshold is set and the second key assignments similarity threshold is all 0.95, candidate entity
The first title similarity of evidence unit C is 0.94, and the second title similarity is 0.9, and the first key assignments similarity is 0.93, second
Key assignments similarity is maximum 0.94 the second title of < of similarity in 0.93, then the first title similarity and the second title similarity
Maximum 0.93 the second key assignments of < of similarity is similar in similarity threshold 0.95, the first key assignments similarity and the second key assignments similarity
Threshold value 0.95 is spent, then adds the second flag bit for candidate entity evidence unit C, preferred flag bit is " 0 ".
S306, according to the matching degree and zone bit information of the candidate entity evidence unit, at least two candidate entities
Evidence unit is ranked up.
Optionally, the first title similarity, the second title similarity, the first key assignments similarity and the second key assignments is similar
Degree summation, obtains the matching degree of candidate entity evidence unit.
Specifically, the matching degree of candidate entity evidence unit is by the first title similarity, the second title similarity, first
What key assignments similarity and the second key assignments similarity linear superposition obtained.The matching degree the high, sorts more forward, can be to difference
The candidate entity evidence unit of class zone bit information is ranked up respectively by matching degree.For example, the candidate that flag bit is " 1 " is real
Body evidence unit is assigned in the first candidate entity evidence group, and the candidate entity evidence unit that flag bit is " 0 " is assigned to the
Two candidate entity evidence groups.Also, it is real according to the candidate in the first candidate entity evidence group and/or the second candidate entity evidence group
The matching degree of body evidence unit is ranked up.
S307, according to ranking results, determine the target entity evidence unit of the knowledge mapping to be tested.
Specifically, the candidate entity evidence unit in preset quantity threshold value is determined as knowledge mapping to be tested by sequence
Target entity evidence unit.Wherein, the first amount threshold of the first candidate entity evidence group and the second candidate entity evidence group
The second amount threshold it is different, general first amount threshold is greater than the second amount threshold.By in the first candidate entity evidence group
And second candidate entity evidence group determine the target entity evidence unit of knowledge mapping to be tested respectively, avoid omission with to
Examine knowledge mapping similar in entity evidence unit so that it is subsequent include target entity evidence unit after, knowledge graph to be tested
Spectrum is more comprehensively and reliable.
The embodiment of the present invention is by calculating separately the physical name in knowledge mapping to be tested, in candidate entity evidence unit
Title between title similarity and the attribute in the attribute value in knowledge mapping to be tested, with candidate entity evidence unit
Key assignments similarity between key assignments, and target entity evidence list is determined according to the title similarity of acquisition and key assignments similarity
Member improves the building efficiency of knowledge mapping, reduces manpower intervention.
Example IV
Fig. 4 is a kind of structural schematic diagram of the processing unit for knowledge mapping that the embodiment of the present invention four provides, which can
The processing method for executing a kind of knowledge mapping provided by any embodiment of the invention has the corresponding functional module of execution method
And beneficial effect.As shown in figure 4, the apparatus may include:
Candidate entity evidence Unit selection module 41 is used for according to knowledge mapping to be verified, from knowledge mapping institute to be verified
Belong to the candidate entity evidence unit of selection at least two in the entity evidence unit of target domain；
Matching degree determining module 42, for determining the knowledge mapping to be tested and at least two candidate entity evidence
The matching degree of unit；
Target entity evidence Unit selection module 43 is used for according to the matching degree, from described at least two candidate entities
The target entity evidence unit that the knowledge mapping to be tested is selected in evidence unit, for being based on the target entity evidence unit
The knowledge mapping to be verified is verified.
On the basis of the above embodiments, described device further includes that entity evidence unit obtains module, is specifically used for:
Chapter title information belonging to attribute information and attribute information is extracted from the target domain fact；
Using the chapter title information as the title of entity evidence unit in target domain；
Using the Property Name in the attribute information as the attribute keyword of entity evidence unit in target domain；
Using the property content in the attribute information as the attribute key assignments of entity evidence unit in target domain.
On the basis of the above embodiments, the candidate entity evidence Unit selection module 41, is specifically used for:
Using the physical name in knowledge mapping to be verified as search term, in the chapters and sections mark of the entity evidence unit of target domain
It is scanned in topic information, attribute keyword and attribute key assignments, obtains first instance evidence unit；
Using the attribute value in knowledge mapping to be verified as search term, in the property key of the entity evidence unit of target domain
It is scanned in value, obtains second instance evidence unit；
The candidate entity card of selection at least two from the first instance evidence unit and the second instance evidence unit
According to unit.
On the basis of the above embodiments, the candidate entity evidence Unit selection module 41, is specifically also used to:
According to mapping relations in attribute-name in knowledge mapping and entity evidence unit between attribute keyword and to be tested
Knowledge mapping is demonstrate,proved, selection and the matched candidate of knowledge mapping to be verified from described at least two candidate entity evidence units
Entity evidence unit, and filter out other candidate entity evidence units.
On the basis of the above embodiments, the matching degree determining module 42, is specifically used for:
Determine the physical name in knowledge mapping to be tested, the title between the title in candidate entity evidence unit is similar
Degree；
The attribute value in knowledge mapping to be tested is determined, with the key assignments between the attribute key assignments in candidate entity evidence unit
Similarity；
According to the title similarity and key assignments similarity of the candidate entity evidence unit, candidate's entity evidence unit is determined
Matching degree.
On the basis of the above embodiments, the matching degree determining module 42, is specifically also used to:
By the title in the physical name and candidate entity evidence unit in the knowledge mapping to be tested, as prediction model
Input, obtain the first title similarity；
It is overlapped according to the text between the physical name of the knowledge mapping to be tested and the title of candidate entity evidence unit
Relationship determines the second title similarity.
On the basis of the above embodiments, the matching degree determining module 42, is specifically also used to:
By the attribute key assignments in the attribute value and candidate entity evidence unit in the knowledge mapping to be tested, as prediction
The input of model obtains the first key assignments similarity；
According to the text between the attribute value of the knowledge mapping to be tested and the attribute key assignments of candidate entity evidence unit
Coincidence relation determines the second key assignments similarity.
On the basis of the above embodiments, the matching degree determining module 42, is specifically also used to:
If the title similarity of any candidate's entity evidence unit is real less than the first title similarity threshold or the candidate
The key assignments similarity of body evidence unit then filters out candidate's entity evidence unit less than the first key assignments similarity threshold.
On the basis of the above embodiments, the target entity evidence Unit selection module 43, is specifically used for:
If the title similarity of candidate's entity evidence unit is greater than the second title similarity threshold, and key assignments similarity
Greater than the second key assignments similarity threshold, then the first flag bit is added for the candidate entity evidence unit；It otherwise, is the candidate
Entity evidence unit adds the second flag bit；
According to the matching degree and zone bit information of the candidate entity evidence unit, at least two candidate entity evidence lists
Member is ranked up；
According to ranking results, the target entity evidence unit of the knowledge mapping to be tested is determined.
A kind of processing unit of knowledge mapping provided by the embodiment of the present invention, executable any embodiment of that present invention are mentioned
A kind of processing method of the knowledge mapping supplied, has the corresponding functional module of execution method and beneficial effect.Not in the present embodiment
In detailed description technical detail, reference can be made to any embodiment of that present invention provide a kind of knowledge mapping processing method.
Embodiment five
Fig. 5 is a kind of structural schematic diagram for equipment that the embodiment of the present invention five provides.Fig. 5, which is shown, to be suitable for being used to realizing this
The block diagram of the example devices 500 of invention embodiment.The equipment 500 that Fig. 5 is shown is only an example, should not be to the present invention
The function and use scope of embodiment bring any restrictions.
As shown in figure 5, equipment 500 is showed in the form of universal computing device.The component of equipment 500 may include but unlimited
In one or more processor or processing unit 501, system storage 502, different system components (including system is connected
Memory 502 and processing unit 501) bus 503.
Bus 503 indicates one of a few class bus structures or a variety of, including memory bus or Memory Controller,
Peripheral bus, graphics acceleration port, processor or the local bus using any bus structures in a variety of bus structures.It lifts
For example, these architectures include but is not limited to industry standard architecture (ISA) bus, microchannel architecture (MAC)
Bus, enhanced isa bus, Video Electronics Standards Association (VESA) local bus and peripheral component interconnection (PCI) bus.
Equipment 500 typically comprises a variety of computer system readable media.These media can be it is any can be by equipment
The usable medium of 500 access, including volatile and non-volatile media, moveable and immovable medium.
System storage 502 may include the computer system readable media of form of volatile memory, such as deposit at random
Access to memory (RAM) 504 and/or cache memory 505.Equipment 500 may further include other removable/not removable
Dynamic, volatile/non-volatile computer system storage medium.Only as an example, storage system 506 can be used for read and write can not
Mobile, non-volatile magnetic media (Fig. 5 do not show, commonly referred to as " hard disk drive ").Although being not shown in Fig. 5, Ke Yiti
For the disc driver for being read and write to removable non-volatile magnetic disk (such as " floppy disk "), and to moving non-volatile light
The CD drive of disk (such as CD-ROM, DVD-ROM or other optical mediums) read-write.In these cases, each driver
It can be connected by one or more data media interfaces with bus 503.Memory 502 may include that at least one program produces
Product, the program product have one group of (for example, at least one) program module, these program modules are configured to perform of the invention each
The function of embodiment.
Program/utility 508 with one group of (at least one) program module 507, can store in such as memory
In 502, such program module 507 includes but is not limited to operating system, one or more application program, other program modules
And program data, it may include the realization of network environment in each of these examples or certain combination.Program module 507
Usually execute the function and/or method in embodiment described in the invention.
Equipment 500 can also be logical with one or more external equipments 509 (such as keyboard, sensing equipment, display 510 etc.)
Letter, can also be enabled a user to one or more equipment interact with the equipment 500 communicate, and/or with make the equipment 500
Any equipment (such as network interface card, modem etc.) that can be communicated with one or more of the other calculating equipment communicates.This
Kind communication can be carried out by input/output (I/O) interface 511.Also, equipment 500 can also by network adapter 512 with
One or more network (such as local area network (LAN), wide area network (WAN) and/or public network, such as internet) communication.Such as
Shown in figure, network adapter 512 is communicated by bus 503 with other modules of equipment 500.It should be understood that although not showing in figure
Out, other hardware and/or software module can be used with bonding apparatus 500, including but not limited to: microcode, device driver, superfluous
Remaining processing unit, external disk drive array, RAID system, tape drive and data backup storage system etc..
Processing unit 501 by the program that is stored in system storage 502 of operation, thereby executing various function application with
And data processing, such as realize a kind of processing method of knowledge mapping provided by the embodiment of the present invention, comprising:
According to knowledge mapping to be verified, from the entity evidence unit in knowledge mapping said target to be verified field selection to
Few two candidate entity evidence units；
Determine the matching degree of the knowledge mapping to be tested and at least two candidate entity evidence unit；
According to the matching degree, the knowledge mapping to be tested is selected from described at least two candidate entity evidence units
Target entity evidence unit, for being verified to the knowledge mapping to be verified based on the target entity evidence unit.
Embodiment six
The embodiment of the present invention six additionally provides a kind of computer readable storage medium, the computer executable instructions by
For executing a kind of processing method of knowledge mapping when computer processor executes, this method comprises:
According to knowledge mapping to be verified, from the entity evidence unit in knowledge mapping said target to be verified field selection to
Few two candidate entity evidence units；
Determine the matching degree of the knowledge mapping to be tested and at least two candidate entity evidence unit；
According to the matching degree, the knowledge mapping to be tested is selected from described at least two candidate entity evidence units
Target entity evidence unit, for being verified to the knowledge mapping to be verified based on the target entity evidence unit.
Certainly, a kind of storage medium comprising computer executable instructions, computer provided by the embodiment of the present invention
The method operation that executable instruction is not limited to the described above, can also be performed a kind of knowledge provided by any embodiment of the invention
Relevant operation in the processing method of map.The computer readable storage medium of the embodiment of the present invention can use one or more
Any combination of a computer-readable medium.Computer-readable medium can be computer-readable signal media or computer
Readable storage medium storing program for executing.Computer readable storage medium for example may be-but not limited to-electricity, magnetic, optical, electromagnetic, infrared
The system of line or semiconductor, device or device, or any above combination.Computer readable storage medium it is more specific
Example (non exhaustive list) includes: electrical connection with one or more conducting wires, portable computer diskette, hard disk, random
It accesses memory (RAM), read-only memory (ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, portable
Formula compact disc read-only memory (CD-ROM), light storage device, magnetic memory device or above-mentioned any appropriate combination.
In this document, it includes or the tangible medium of storage program that the program can be by that computer readable storage medium, which can be any,
Instruction execution system, device or device use or in connection.
Computer-readable signal media may include in a base band or as carrier wave a part propagate data-signal,
Wherein carry computer-readable program code.The data-signal of this propagation can take various forms, including but unlimited
In electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be that computer can
Any computer-readable medium other than storage medium is read, which can send, propagates or transmit and be used for
By the use of instruction execution system, device or device or program in connection.
The program code for including on computer-readable medium can transmit with any suitable medium, including --- but it is unlimited
In wireless, electric wire, optical cable, RF etc. or above-mentioned any appropriate combination.
The computer for executing operation of the present invention can be write with one or more programming languages or combinations thereof
Program code, described program design language include object oriented program language-such as Java, Smalltalk, C++,
It further include conventional procedural programming language-such as " C " language or similar programming language.Program code can be with
It fully executes, partly execute on the user computer on the user computer, being executed as an independent software package, portion
Divide and partially executes or executed on a remote computer or server completely on the remote computer on the user computer.?
Be related in the situation of remote computer, remote computer can pass through the network of any kind --- including local area network (LAN) or
Wide area network (WAN)-be connected to subscriber computer, or, it may be connected to outer computer (such as mentioned using Internet service
It is connected for quotient by internet).
Note that the above is only a better embodiment of the present invention and the applied technical principle.It will be appreciated by those skilled in the art that
The invention is not limited to the specific embodiments described herein, be able to carry out for a person skilled in the art it is various it is apparent variation,
It readjusts and substitutes without departing from protection scope of the present invention.Therefore, although being carried out by above embodiments to the present invention
It is described in further detail, but the present invention is not limited to the above embodiments only, without departing from the inventive concept, also
It may include more other equivalent embodiments, and the scope of the invention is determined by the scope of the appended claims.
Claims (15)
1. a kind of processing method of knowledge mapping, which is characterized in that the described method includes:
According to knowledge mapping to be verified, at least two are selected from the entity evidence unit in knowledge mapping said target to be verified field
A candidate's entity evidence unit；
Determine the matching degree of the knowledge mapping to be tested and at least two candidate entity evidence unit；
According to the matching degree, the mesh of the knowledge mapping to be tested is selected from described at least two candidate entity evidence units
Entity evidence unit is marked, for verifying based on the target entity evidence unit to the knowledge mapping to be verified.
2. the method according to claim 1, wherein according to knowledge mapping to be verified, from knowledge mapping to be verified
In the entity evidence unit in said target field before the candidate entity evidence unit of selection at least two, further includes:
Chapter title information belonging to attribute information and attribute information is extracted from the target domain fact；
Using the chapter title information as the title of entity evidence unit in target domain；
Using the Property Name in the attribute information as the attribute keyword of entity evidence unit in target domain；
Using the property content in the attribute information as the attribute key assignments of entity evidence unit in target domain.
3. the method according to claim 1, wherein according to knowledge mapping to be verified, from knowledge mapping to be verified
The candidate entity evidence unit of selection at least two in the entity evidence unit in said target field, comprising:
Using the physical name in knowledge mapping to be verified as search term, in the chapter title letter of the entity evidence unit of target domain
It is scanned in breath, attribute keyword and attribute key assignments, obtains first instance evidence unit；
Using the attribute value in knowledge mapping to be verified as search term, in the attribute key assignments of the entity evidence unit of target domain
It scans for, obtains second instance evidence unit；
The candidate entity evidence list of selection at least two from the first instance evidence unit and the second instance evidence unit
Member.
4. according to the method described in claim 3, it is characterized in that, from the first instance evidence unit and the second instance
In evidence unit after the candidate entity evidence unit of selection at least two, further includes:
According to the mapping relations in attribute-name in knowledge mapping and entity evidence unit between attribute keyword and to be verified know
Know map, selection and the matched candidate entity of the knowledge mapping to be verified from described at least two candidate entity evidence units
Evidence unit, and filter out other candidate entity evidence units.
5. the method according to claim 1, wherein determining the knowledge mapping to be tested and described at least two
The matching degree of candidate entity evidence unit, comprising:
The physical name in knowledge mapping to be tested is determined, with the title similarity between the title in candidate entity evidence unit；
Determine the attribute value in knowledge mapping to be tested, the key assignments between the attribute key assignments in candidate entity evidence unit is similar
Degree；
According to the title similarity and key assignments similarity of the candidate entity evidence unit, of candidate's entity evidence unit is determined
With degree.
6. according to the method described in claim 5, it is characterized in that, the physical name in knowledge mapping to be tested is determined, with candidate
The title similarity between title in entity evidence unit, comprising:
By the title in the physical name and candidate entity evidence unit in the knowledge mapping to be tested, as the defeated of prediction model
Enter, obtains the first title similarity；
According to the text coincidence relation between the physical name of the knowledge mapping to be tested and the title of candidate entity evidence unit,
Determine the second title similarity.
7. according to the method described in claim 5, it is characterized in that, the attribute value in knowledge mapping to be tested is determined, with candidate
The key assignments similarity between attribute key assignments in entity evidence unit, comprising:
By the attribute key assignments in the attribute value and candidate entity evidence unit in the knowledge mapping to be tested, as prediction model
Input, obtain the first key assignments similarity；
It is overlapped according to the text between the attribute value of the knowledge mapping to be tested and the attribute key assignments of candidate entity evidence unit
Relationship determines the second key assignments similarity.
8. according to the method described in claim 5, it is characterized in that, according to the title similarity of the candidate entity evidence unit
With key assignments similarity, before the matching degree for determining candidate's entity evidence unit, further includes:
If the title similarity of any candidate's entity evidence unit is demonstrate,proved less than the first title similarity threshold or candidate's entity
According to the key assignments similarity of unit less than the first key assignments similarity threshold, then candidate's entity evidence unit is filtered out.
9. according to the method described in claim 5, it is characterized in that, being selected from described at least two candidate entity evidence units
The target entity evidence unit of the knowledge mapping to be tested, comprising:
If the title similarity of candidate's entity evidence unit is greater than the second title similarity threshold, and key assignments similarity is greater than
Second key assignments similarity threshold then adds the first flag bit for the candidate entity evidence unit；It otherwise, is the candidate entity
Evidence unit adds the second flag bit；
According to the matching degree and zone bit information of the candidate entity evidence unit, at least two candidate entity evidence units into
Row sequence；
According to ranking results, the target entity evidence unit of the knowledge mapping to be tested is determined.
10. a kind of processing unit of knowledge mapping, which is characterized in that described device includes:
Candidate entity evidence Unit selection module is used for according to knowledge mapping to be verified, from knowledge mapping said target to be verified
The candidate entity evidence unit of selection at least two in the entity evidence unit in field；
Matching degree determining module, for determining the knowledge mapping to be tested and described at least two candidate entity evidence units
Matching degree；
Target entity evidence Unit selection module is used for according to the matching degree, from described at least two candidate entity evidence lists
The target entity evidence unit that the knowledge mapping to be tested is selected in member, for being based on the target entity evidence unit to described
Knowledge mapping to be verified is verified.
11. device according to claim 10, which is characterized in that described device further includes that entity evidence unit obtains mould
Block is specifically used for:
Chapter title information belonging to attribute information and attribute information is extracted from the target domain fact；
Using the chapter title information as the title of entity evidence unit in target domain；
Using the Property Name in the attribute information as the attribute keyword of entity evidence unit in target domain；
Using the property content in the attribute information as the attribute key assignments of entity evidence unit in target domain.
12. device according to claim 10, which is characterized in that candidate's entity evidence Unit selection module, specifically
For:
Using the physical name in knowledge mapping to be verified as search term, in the chapter title letter of the entity evidence unit of target domain
It is scanned in breath, attribute keyword and attribute key assignments, obtains first instance evidence unit；
Using the attribute value in knowledge mapping to be verified as search term, in the attribute key assignments of the entity evidence unit of target domain
It scans for, obtains second instance evidence unit；
The candidate entity evidence list of selection at least two from the first instance evidence unit and the second instance evidence unit
Member.
13. device according to claim 12, which is characterized in that candidate's entity evidence Unit selection module, specifically
It is also used to:
According to the mapping relations in attribute-name in knowledge mapping and entity evidence unit between attribute keyword and to be verified know
Know map, selection and the matched candidate entity of the knowledge mapping to be verified from described at least two candidate entity evidence units
Evidence unit, and filter out other candidate entity evidence units.
14. a kind of equipment, which is characterized in that the equipment further include:
One or more processors；
Storage device, for storing one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processors are real
A kind of now processing method of knowledge mapping as described in any in claim 1-9.
15. a kind of computer-readable medium, is stored thereon with computer program, which is characterized in that the program is executed by processor
A kind of processing method of knowledge mapping of the Shi Shixian as described in any in claim 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910537133.5A CN110263083B (en) | 2019-06-20 | 2019-06-20 | Knowledge graph processing method, device, equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910537133.5A CN110263083B (en) | 2019-06-20 | 2019-06-20 | Knowledge graph processing method, device, equipment and medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110263083A true CN110263083A (en) | 2019-09-20 |
| CN110263083B CN110263083B (en) | 2022-04-05 |
Family
ID=67919850
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910537133.5A Active CN110263083B (en) | 2019-06-20 | 2019-06-20 | Knowledge graph processing method, device, equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110263083B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110750654A (en) * | 2019-10-28 | 2020-02-04 | 中国建设银行股份有限公司 | Knowledge graph acquisition method, device, equipment and medium |
| CN111640511A (en) * | 2020-05-29 | 2020-09-08 | 北京百度网讯科技有限公司 | Medical fact verification method and device, electronic equipment and storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106021281A (en) * | 2016-04-29 | 2016-10-12 | 京东方科技集团股份有限公司 | Method for establishing medical knowledge graph, device for same and query method for same |
| CN108268581A (en) * | 2017-07-14 | 2018-07-10 | 广东神马搜索科技有限公司 | The construction method and device of knowledge mapping |
| CN108376160A (en) * | 2018-02-12 | 2018-08-07 | 北京大学 | A kind of Chinese knowledge mapping construction method and system |
| CN108388580A (en) * | 2018-01-24 | 2018-08-10 | 平安医疗健康管理股份有限公司 | Merge the dynamic knowledge collection of illustrative plates update method of medical knowledge and application case |
| CN109086356A (en) * | 2018-07-18 | 2018-12-25 | 哈尔滨工业大学 | The incorrect link relationship diagnosis of extensive knowledge mapping and modification method |
| CN109299285A (en) * | 2018-09-11 | 2019-02-01 | 中国医学科学院医学信息研究所 | A kind of pharmacogenomics knowledge mapping construction method and system |
| CN109829056A (en) * | 2018-11-09 | 2019-05-31 | 广东外语外贸大学 | Predicate explains the fact that template-driven Abductive reasoning method |
| US20190171944A1 (en) * | 2017-12-06 | 2019-06-06 | Accenture Global Solutions Limited | Integrity evaluation of unstructured processes using artificial intelligence (ai) techniques |
-
2019
- 2019-06-20 CN CN201910537133.5A patent/CN110263083B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106021281A (en) * | 2016-04-29 | 2016-10-12 | 京东方科技集团股份有限公司 | Method for establishing medical knowledge graph, device for same and query method for same |
| CN108268581A (en) * | 2017-07-14 | 2018-07-10 | 广东神马搜索科技有限公司 | The construction method and device of knowledge mapping |
| US20190171944A1 (en) * | 2017-12-06 | 2019-06-06 | Accenture Global Solutions Limited | Integrity evaluation of unstructured processes using artificial intelligence (ai) techniques |
| CN108388580A (en) * | 2018-01-24 | 2018-08-10 | 平安医疗健康管理股份有限公司 | Merge the dynamic knowledge collection of illustrative plates update method of medical knowledge and application case |
| CN108376160A (en) * | 2018-02-12 | 2018-08-07 | 北京大学 | A kind of Chinese knowledge mapping construction method and system |
| CN109086356A (en) * | 2018-07-18 | 2018-12-25 | 哈尔滨工业大学 | The incorrect link relationship diagnosis of extensive knowledge mapping and modification method |
| CN109299285A (en) * | 2018-09-11 | 2019-02-01 | 中国医学科学院医学信息研究所 | A kind of pharmacogenomics knowledge mapping construction method and system |
| CN109829056A (en) * | 2018-11-09 | 2019-05-31 | 广东外语外贸大学 | Predicate explains the fact that template-driven Abductive reasoning method |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110750654A (en) * | 2019-10-28 | 2020-02-04 | 中国建设银行股份有限公司 | Knowledge graph acquisition method, device, equipment and medium |
| CN111640511A (en) * | 2020-05-29 | 2020-09-08 | 北京百度网讯科技有限公司 | Medical fact verification method and device, electronic equipment and storage medium |
| CN111640511B (en) * | 2020-05-29 | 2023-08-04 | 北京百度网讯科技有限公司 | Medical fact verification method, device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110263083B (en) | 2022-04-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9558264B2 (en) | Identifying and displaying relationships between candidate answers | |
| CN106940788B (en) | Intelligent scoring method and device, computer equipment and computer readable medium | |
| CN108897867A (en) | For the data processing method of knowledge question, device, server and medium | |
| CN111797214A (en) | FAQ database-based problem screening method and device, computer equipment and medium | |
| EP3729231A1 (en) | Domain-specific natural language understanding of customer intent in self-help | |
| CN110647614A (en) | Intelligent question and answer method, device, medium and electronic equipment | |
| CN111143547B (en) | Big data display method based on knowledge graph | |
| CN110276023A (en) | POI changes event discovery method, apparatus, calculates equipment and medium | |
| CN109783631A (en) | Method of calibration, device, computer equipment and the storage medium of community's question and answer data | |
| CN110388933A (en) | Interest point search method, device, server and storage medium | |
| CN110263083A (en) | Processing method, device, equipment and the medium of knowledge mapping | |
| CN111126422B (en) | Method, device, equipment and medium for establishing industry model and determining industry | |
| CN116402166B (en) | Training method and device of prediction model, electronic equipment and storage medium | |
| CN112925914B (en) | Data security grading method, system, equipment and storage medium | |
| CN110020974A (en) | Lawyer's recommended method, device, medium and electronic equipment | |
| Kalfarisi et al. | Detecting and geolocating city-scale soft-story buildings by deep machine learning for urban seismic resilience | |
| US20210271637A1 (en) | Creating descriptors for business analytics applications | |
| JP2023517518A (en) | Vector embedding model for relational tables with null or equivalent values | |
| CN116805012A (en) | Quality assessment method and device for multi-mode knowledge graph, storage medium and equipment | |
| CN110362688A (en) | Examination question mask method, device, equipment and computer readable storage medium | |
| CN103377381A (en) | Method and device for identifying content attribute of image | |
| CN110941662A (en) | Graphical method, system, storage medium and terminal for scientific research cooperative relationship | |
| CN115809313A (en) | Text similarity determination method and equipment | |
| CN109933788B (en) | Type determining method, device, equipment and medium | |
| Gao et al. | Detecting geometric conflicts for generalisation of polygonal maps |
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 |