CN113792114A - Credible evaluation method and system for urban field knowledge graph - Google Patents
Credible evaluation method and system for urban field knowledge graph Download PDFInfo
- Publication number
- CN113792114A CN113792114A CN202110890875.3A CN202110890875A CN113792114A CN 113792114 A CN113792114 A CN 113792114A CN 202110890875 A CN202110890875 A CN 202110890875A CN 113792114 A CN113792114 A CN 113792114A
- Authority
- CN
- China
- Prior art keywords
- knowledge
- evidence
- entity
- credibility
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000011156 evaluation Methods 0.000 title claims abstract description 32
- 238000000034 method Methods 0.000 claims abstract description 22
- 238000012795 verification Methods 0.000 claims abstract description 21
- 238000004364 calculation method Methods 0.000 claims abstract description 20
- 238000004422 calculation algorithm Methods 0.000 claims description 10
- 238000012790 confirmation Methods 0.000 claims description 7
- 230000015572 biosynthetic process Effects 0.000 claims description 5
- 230000004927 fusion Effects 0.000 claims description 5
- 238000010606 normalization Methods 0.000 claims description 5
- 230000008520 organization Effects 0.000 claims description 4
- 230000036962 time dependent Effects 0.000 claims description 4
- 238000012217 deletion Methods 0.000 claims description 3
- 230000037430 deletion Effects 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 9
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 239000013598 vector Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012358 sourcing Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a credible evaluation method and a credible evaluation system for a knowledge graph in the urban field, wherein the method comprises the following steps: s100, acquiring credible evidences of a knowledge acquisition stage, a knowledge reasoning stage and a knowledge verification stage from a credible evidence model of the domain knowledge; s200, calculating the credible attribute characteristic values of the knowledge entity and the knowledge model based on the obtained credible evidence and a calculation method of the credible attribute characteristic values in the domain knowledge credible attribute model; s300, calculating the credibility of the knowledge entity based on the credibility attribute characteristic value of the knowledge entity, or calculating the credibility of the knowledge model based on the credibility attribute characteristic value of the knowledge model; and S400, dividing the credibility level of the knowledge entity or the knowledge model. The method can quantify the credibility of the knowledge or knowledge model, ensure the quality of the knowledge model in the urban field, and meet the knowledge requirements of different users in different fields.
Description
Technical Field
The invention relates to the field of knowledge graphs, in particular to a credible assessment method and system for knowledge graphs in the urban field.
Background
Knowledge graph describes the fact in the real world through knowledge triples, noise is inevitably introduced in the process of constructing the knowledge graph, and the existing knowledge graph system lacks credibility evaluation and guarantee of knowledge, which can bring potential damage to the knowledge graph application. In order to provide credible knowledge service for diversified application scenes in cities, credibility assessment and guarantee of knowledge in a knowledge map become key problems to be solved. At present, the research on knowledge credibility at home and abroad is still in the starting stage, and no standards or specifications of a knowledge credibility attribute model, a knowledge credibility evaluation method and knowledge credibility grading evaluation exist. In the prior art, some methods utilize representation learning to represent knowledge triples as low-dimensional vectors, and detect noise in the knowledge graph through vector distance.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide a credible evaluation method and a credible evaluation system for an urban domain knowledge map, which can quantify the credibility of knowledge or a knowledge model, ensure the quality of the urban domain knowledge model and meet the knowledge requirements of different users in different fields.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
a credible assessment method for a knowledge graph in the urban field comprises the following steps:
s100, acquiring credible evidences of a knowledge acquisition stage, a knowledge reasoning stage and a knowledge verification stage from a credible evidence model of the domain knowledge;
s200, calculating the credible attribute characteristic values of the knowledge entity and the knowledge model based on the obtained credible evidence and a calculation method of the credible attribute characteristic values in the domain knowledge credible attribute model, wherein the credible attributes comprise reliability, correctness, timeliness, safety and accessibility;
s300, calculating the credibility of the knowledge entity based on the credibility attribute characteristic value of the knowledge entity, or calculating the credibility of the knowledge model based on the credibility attribute characteristic value of the knowledge model;
and S400, based on the credibility of the knowledge entity or the knowledge model and the support degree of the credibility evidence of the life cycle of the knowledge model to the credibility of the knowledge entity, dividing the credibility level of the knowledge entity or the knowledge model.
Further, in the system as described above, in S200, the knowledge entity includes knowledge concepts, knowledge instances, knowledge attributes, and knowledge relationships, and calculating the credible attribute feature value of the knowledge entity includes:
(1) measuring the reliability of the knowledge entity: calculating a reliability characteristic value R of the kth knowledge entity by:
β=max(eds)*max(edt)-min(eds)*min(edf),
wherein e isdsAs evidence of data origin, edfAs evidence of data type, edtFor data time evidence, β is the normalization factor and N is the dataThe number of sources;
(2) measuring the correctness of the knowledge entity: the correctness feature value C of the k-th knowledge entity is calculated by:
C=a1edq+a2(era*err)+a3max(euv,etr),
edq=(1-edd)*(1-edr)*edc+ecp*ech*eem+ete*eec,
wherein e isdqAs evidence of data quality, euvFor the user to visit the evidence, eraFor reasoning about the proof of algorithm performance, errFor reasoning about the evidence of the result, etrVerifying the reported evidence for a third party, a1,a2,a3The weights of the knowledge acquisition stage, the knowledge inference stage and the knowledge verification stage are respectively set as default average values 1/3, eddAs evidence of data loss rate, edrAs evidence of data redundancy rate, edcFor proof of concept matching confidence, ecpExtracting evidence for a concept, echFor proof of concept hierarchy relationship, eemFor element fusion matching probability, eteFor triple formation confidence evidence, eecExtracting algorithm confidence evidence;
(3) measuring the timeliness of the knowledge entity: calculating the timeliness characteristic value T of the k-th knowledge entity by the following formula:
T=max(edt,eee),
wherein e isdtAs evidence of data time, eeeEditing evidence for an expert;
(4) measuring the safety of the knowledge entity: calculating a security feature value S for the kth knowledge entity by:
S=1-erd*(1-euf),
wherein e isrdAs evidence of the degree of risk, eufFeeding back evidence for the user;
(5) measuring accessibility of a knowledge entity: the accessibility feature value a of the k-th knowledge entity is calculated by:
A=max(edm),
wherein e isdmIs field evidence.
Further, in the system as described above, in S200, the knowledge model includes one or more knowledge entities, and calculating the credible attribute feature value of the knowledge model includes:
(1) measuring the reliability of the knowledge model: the reliability characteristic value R' of the knowledge model is calculated by:
wherein k is the number of knowledge entities, RiReliability characteristic value of the ith knowledge entity;
(2) measuring the correctness of the knowledge model: the correctness feature C' of the knowledge model is calculated by:
wherein k is the number of knowledge entities, CiThe correctness characteristic value of the ith knowledge entity is obtained;
(3) measuring the timeliness of the knowledge model: calculating the timeliness characteristic value T' of the knowledge model by the following formula:
T'=min(T1,T2,...,Tk),
wherein k is the number of knowledge entities, CkThe timeliness characteristic value of the k knowledge entity is obtained;
(4) measuring the safety of the knowledge model: the safety feature value S' of the knowledge model is calculated by:
S'=min(S1,S2,...,Sk),
wherein k is the number of knowledge entities, SkThe safety characteristic value of the k-th knowledge entity;
(5) measuring accessibility of the knowledge model: the accessibility feature value H of the knowledge model is calculated by:
H={m|Am≥θ,m=1,2,...,k}
where θ is the accessibility threshold, k is the number of knowledge entities, AmIs the accessibility feature value of the mth knowledge entity.
Further, the system as described above, S300 includes:
calculating the trustworthiness KT of the knowledge entity by:
KT=αRR+αCC+αTT+αSS+αAA,
αR+αC+αT+αS+αA=1,
wherein alpha isRIs the weight, alpha, of the reliability characteristic value R of the knowledge entityCIs the weight, alpha, of the correctness feature C of the knowledge entityTIs the weight, alpha, of the time-dependent characteristic value T of the knowledge entitySIs the weight, alpha, of the security feature value S of the knowledge entityAIs the weight of the accessibility feature value A of the knowledge entity.
Further, in the system as described above, S300 further includes:
and when the credibility of the knowledge entity is calculated, carrying out credible attribute weight configuration and credible evidence configuration according to user requirements, wherein the weights of the five credible attributes are all 0.2 under default configuration, and all credible evidences are used for credible evaluation of knowledge.
Further, the system as described above, S300 includes:
calculating the confidence C of the knowledge model byr:
Cr=α1R'+α2C'+α3T'+α4S'+α5A',
α1+α2+α3+α4+α5=1,
Wherein alpha is1Is the weight, alpha, of the reliability characteristic value R' of the knowledge model2For the correctness feature C' of the knowledge modelWeight, α3Is the weight, alpha, of the time-dependent characteristic value T' of the knowledge model4Is the weight, alpha, of the security feature value S' of the knowledge model5Is the weight of the accessibility feature value H of the knowledge model.
Further, in the system as described above, S300 further includes:
performing iterative incremental calculation on the credibility of the knowledge model, wherein the credibility attribute characteristic value of newly added and deleted knowledge entities is less than 1,2k,Ck,Tk,Sk,Ak>. includes:
(1) measuring the reliability of the newly added and deleted knowledge entities: calculating the reliability characteristic values of the added and deleted knowledge entities by the following formula:
wherein N is the number of the original knowledge entities, R is the average reliability of the original knowledge entities, R' is the average reliability of the newly added knowledge entities, RiReliability characteristic value of the ith knowledge entity;
(2) measuring the correctness of the addition and deletion of the knowledge entities: calculating the correctness characteristic values of the added and deleted knowledge entities by the following formula:
wherein N is the number of the original knowledge entities, C is the average correctness of the original knowledge entities, C' is the average correctness of the newly added knowledge entities, CiThe correctness characteristic value of the ith knowledge entity is obtained;
(3) measuring timeliness of adding and deleting knowledge entities: calculating the timeliness characteristic values of the added and deleted knowledge entities by the following formula:
T=min(T,T'),T'=min(T1,T2,...,Tk),
wherein T is original knowledgeIdentifying the timeliness of the model, T' being the minimum timeliness of the newly added knowledge entity, TkThe timeliness of the kth newly added knowledge entity;
(4) measuring the safety of adding and deleting knowledge entities: calculating the security characteristic values of the added and deleted knowledge entities by the following formula:
S=min(S,S'),S'=min(S1,S2,...,Sk),
wherein S is the security of the original knowledge model, S' is the minimum security of the newly added knowledge entity, SkThe safety characteristic value of the kth newly added knowledge entity is obtained;
(5) measuring the accessibility of the added and deleted knowledge entities: the accessibility feature value of the newly added knowledge entity is calculated by:
the accessibility feature value of the deleted knowledge entity is calculated by:
where θ is the accessibility threshold, k is the number of knowledge entities, AmThe accessibility feature value for the mth deleted knowledge entity.
Further, the system as described above, S400 includes:
the credibility of knowledge is divided into five grades, which are respectively: the method comprises the following steps of classifying the credibility level of a knowledge entity according to rules, wherein the credibility level comprises an unknown level, a existence level, a confirmation level, a utility level and an evaluation level, and the classification comprises the following steps:
1) if the evidence missing rate of the knowledge entity is more than 80 percent, and the evidence missing rate is not common knowledge and axiom, or the calculated credibility is less than 0.2, defining the credibility level of the knowledge entity as unknown level;
2) if the evidence of knowledge credibility exists, the calculated credibility is less than 0.5, but no credible attribute which can be verified exists, the credibility level of the knowledge entity is defined as unknown level;
3) if the credible attribute of the knowledge entity has verifiable credible evidence and the characteristic value of the credible attribute is greater than 0.5, defining the credible grade of the knowledge entity as a practical grade;
4) if the evidence of the knowledge entity also comprises the evidence of the successful application case on the basis of meeting the condition of the confirmation level, defining the credibility level of the knowledge entity as a practical level;
5) and if the evidence of the knowledge entity also comprises evidence which passes the credibility analysis of the independent authority verification and analysis organization, defining the credibility level of the knowledge entity as an evaluation level.
An urban area knowledge graph credible evaluation system, comprising:
the evidence obtaining module is used for obtaining credible evidences of a knowledge obtaining stage, a knowledge reasoning stage and a knowledge verification stage from the credible evidence model of the domain knowledge;
the first calculation module is used for calculating the credible attribute characteristic values of the knowledge entity and the knowledge model based on the acquired credible evidence and a calculation method of the credible attribute characteristic values in the domain knowledge credible attribute model, wherein the credible attributes comprise reliability, correctness, timeliness, safety and accessibility;
the second calculation module is used for calculating the credibility of the knowledge entity based on the credibility attribute characteristic value of the knowledge entity or calculating the credibility of the knowledge model based on the credibility attribute characteristic value of the knowledge model;
and the grade dividing module is used for dividing the credibility grade of the knowledge entity or the knowledge model based on the credibility of the knowledge entity or the knowledge model and the support degree of the credibility evidence of the life cycle of the knowledge model to the credibility of the knowledge entity.
Further, the system as described above, the knowledge entity includes knowledge concepts, knowledge instances, knowledge attributes and knowledge relationships, and calculating the credible attribute feature value of the knowledge entity includes:
(1) measuring the reliability of the knowledge entity: calculating a reliability characteristic value R of the kth knowledge entity by:
β=max(eds)*max(edt)-min(eds)*min(edf),
wherein e isdsAs evidence of data origin, edfAs evidence of data type, edtFor the data time evidence, beta is a normalization factor, and N is the number of data sources;
(2) measuring the correctness of the knowledge entity: the correctness feature value C of the k-th knowledge entity is calculated by:
C=a1edq+a2(era*err)+a3max(euv,etr),
edq=(1-edd)*(1-edr)*edc+ecp*ech*eem+ete*eec,
wherein e isdqAs evidence of data quality, euvFor the user to visit the evidence, eraFor reasoning about the proof of algorithm performance, errFor reasoning about the evidence of the result, etrVerifying the reported evidence for a third party, a1,a2,a3The weights of the knowledge acquisition stage, the knowledge inference stage and the knowledge verification stage are respectively set as default average values 1/3, eddAs evidence of data loss rate, edrAs evidence of data redundancy rate, edcFor proof of concept matching confidence, ecpExtracting evidence for a concept, echFor proof of concept hierarchy relationship, eemFor element fusion matching probability, eteFor triple formation confidence evidence, eecExtracting algorithm confidence evidence;
(3) measuring the timeliness of the knowledge entity: calculating the timeliness characteristic value T of the k-th knowledge entity by the following formula:
T=max(edt,eee),
wherein e isdtAs evidence of data time, eeeEditing evidence for an expert;
(4) measuring the safety of the knowledge entity: calculating a security feature value S for the kth knowledge entity by:
S=1-erd*(1-euf),
wherein e isrdAs evidence of the degree of risk, eufFeeding back evidence for the user;
(5) measuring accessibility of a knowledge entity: the accessibility feature value a of the k-th knowledge entity is calculated by:
A=max(edm),
wherein e isdmIs field evidence.
The invention has the beneficial effects that: according to the method, the credibility connotation of knowledge is split by using the credibility attribute model of knowledge, credibility-related data and information in the complete life cycle of knowledge are collected by using the credibility attribute model as credibility evidence to evaluate the credibility attribute value of knowledge, and then the credibility attribute value of knowledge is synthesized, so that the credibility of knowledge in the knowledge map can be comprehensively and multi-dimensionally quantitatively evaluated.
Drawings
Fig. 1 is a schematic flowchart of a trusted evaluation method for a multi-view urban domain knowledge graph according to an embodiment of the present invention;
fig. 2 is a block diagram of a multi-view urban domain knowledge graph credible evaluation method provided in an embodiment of the present invention;
FIG. 3 is a diagram of a domain knowledge credible attribute model provided in an embodiment of the present invention;
FIG. 4 is a diagram of a domain knowledge credible evidence model provided in an embodiment of the invention;
FIG. 5 is a schematic diagram of a configuration interface of a trusted evaluation method provided in an embodiment of the present invention;
FIG. 6 is a diagram illustrating a trust hierarchy model provided in an embodiment of the present invention;
fig. 7 is a schematic structural diagram of a multi-view urban domain knowledge-graph credible evaluation system provided in an embodiment of the present invention.
Detailed Description
In order to make the technical problems solved, the technical solutions adopted, and the technical effects achieved by the present invention clearer, the technical solutions of the embodiments of the present invention will be further described in detail with reference to the accompanying drawings.
The embodiment of the invention provides a multi-view-angle credible evaluation method for an urban domain knowledge graph, which comprises the following steps of:
s100, obtaining credible evidences of a knowledge acquisition stage, a knowledge reasoning stage and a knowledge verification stage from the credible evidence model of the domain knowledge.
Fig. 3 is a schematic diagram of a domain knowledge credibility evidence model, which defines an evidence set supporting knowledge credibility assessment and is a data source for knowledge credibility assessment. We divide the knowledge lifecycle into three phases: the method comprises a knowledge acquisition stage, a knowledge reasoning stage and a knowledge verification stage, wherein evidences related to knowledge credibility are collected from each stage. The credible evidence model records the evidence names and the types generated in the knowledge life cycle, and provides a basis for the credibility evaluation and guarantee of the knowledge of the domain knowledge model.
S200, calculating the credible attribute characteristic values of the knowledge entity and the knowledge model based on the obtained credible evidence and a calculation method of the credible attribute characteristic values in the domain knowledge credible attribute model, wherein the credible attributes comprise reliability, correctness, timeliness, safety and accessibility.
Fig. 4 is a schematic diagram of a domain knowledge credibility attribute model, and five credibility attributes of a knowledge graph model are calculated by using evidence collected by an evidence model according to a calculation method of credibility attribute feature values in a credibility attribute model defined in a domain knowledge credibility measurement method, and then calculated by using the credibility attributes of knowledge through a statistical method.
In S200, the knowledge entity includes knowledge concepts, knowledge instances, knowledge attributes, and knowledge relationships, and the calculating of the credible attribute feature value of the knowledge entity includes:
(1) measuring the reliability of the knowledge entity: calculating a reliability characteristic value R of the kth knowledge entity by:
β=max(eds)*max(edt)-min(eds)*min(edf),
wherein e isdsAs evidence of data origin, edfAs evidence of data type, edtFor data time evidence, β is the normalization factor and N is the number of data sources.
(2) Measuring the correctness of the knowledge entity: the correctness feature value C of the k-th knowledge entity is calculated by:
C=a1edq+a2(era*err)+a3max(euv,etr),
edq=(1-edd)*(1-edr)*edc+ecp*ech*eem+ete*eec,
wherein e isdqAs evidence of data quality, euvFor the user to visit the evidence, eraFor reasoning about the proof of algorithm performance, errFor reasoning about the evidence of the result, etrVerifying the reported evidence for a third party, a1,a2,a3The weights of the knowledge acquisition stage, the knowledge inference stage and the knowledge verification stage are respectively set as default average values 1/3, eddAs evidence of data loss rate, edrAs evidence of data redundancy rate, edcFor proof of concept matching confidence, ecpExtracting evidence for a concept, echFor proof of concept hierarchy relationship, eemFor element fusion matching probability, eteFor triple formation confidence evidence, eecTo extract algorithm confidence evidence.
The correctness of the knowledge entity can be analyzed and verified through a third party authority, whether the knowledge entity is correct or not is judged, and a third party verification report is provided. When the third party verifies that the report evidence exists, the correctness is set to 1 through the verification of a third party institution, namely the knowledge entity is correct; correctness is set to 0 without verification by the third party, indicating that the knowledge entity is erroneous. In the wisdom city domain knowledge graph model, there are a large number of knowledge entities that have not been authoritatively verified by the third party authority, and there is no third party verification report, so the evidence value is set to 0.5 by default.
(3) Measuring the timeliness of the knowledge entity: calculating the timeliness characteristic value T of the k-th knowledge entity by the following formula:
T=max(edt,eee),
wherein e isdtAs evidence of data time, eeeEdit the evidence for the expert.
The timeliness T is mainly obtained by jointly evaluating data time evidence and expert operation time evidence. For knowledge such as common sense or axiom, T is 1, common sense or axiom needs to be obtained through human-computer interaction, and is mainly obtained through feedback of experts, crowd-sourcing task participants and users.
(4) Measuring the safety of the knowledge entity: calculating a security feature value S for the kth knowledge entity by:
S=1-erd*(1-euf),
wherein e isrdAs evidence of the degree of risk, eufAnd feeding back evidence for the user.
(5) Measuring accessibility of a knowledge entity: the accessibility feature value a of the k-th knowledge entity is calculated by:
A=max(edm),
wherein e isdmIs field evidence.
In S200, the knowledge model includes one or more knowledge entities, and the calculating of the credible attribute feature value of the knowledge model includes:
(1) measuring the reliability of the knowledge model: the reliability characteristic value R' of the knowledge model is calculated by:
wherein k is the number of knowledge entities, RiReliability characteristic value of the ith knowledge entity;
(2) measuring the correctness of the knowledge model: the correctness feature C' of the knowledge model is calculated by:
wherein k is the number of knowledge entities, CiThe correctness characteristic value of the ith knowledge entity is obtained;
(3) measuring the timeliness of the knowledge model: calculating the timeliness characteristic value T' of the knowledge model by the following formula:
T'=min(T1,T2,...,Tk),
wherein k is the number of knowledge entities, CkThe timeliness characteristic value of the k knowledge entity is obtained;
(4) measuring the safety of the knowledge model: the safety feature value S' of the knowledge model is calculated by:
S'=min(S1,S2,...,Sk),
wherein k is the number of knowledge entities, SkThe safety characteristic value of the k-th knowledge entity;
(5) measuring accessibility of the knowledge model: the accessibility feature value H of the knowledge model is calculated by:
H={m|Am≥θ,m=1,2,...,k}
where θ is the accessibility threshold, k is the number of knowledge entities, AmIs the accessibility feature value of the mth knowledge entity.
S300, calculating the credibility of the knowledge entity based on the credibility attribute characteristic value of the knowledge entity, or calculating the credibility of the knowledge model based on the credibility attribute characteristic value of the knowledge model.
S300 comprises the following steps:
the trustworthiness KT of the knowledge entity is calculated by:
KT=αRR+αCC+αTT+αSS+αAA,
αR+αC+αT+αS+αA=1,
wherein alpha isRWeight for reliability R, αCAs a weight of correctness C, αTFor the weight of the timeliness T, αSAs a weight of security S, αAIs the weight of accessibility a.
When the system is oriented to a specific application scenario, the weights of the five credible attributes can be configured by a user according to the requirements of the user, as shown in fig. 5, the system provides a knowledge credibility calculation configuration function, and the user can set two levels according to the requirements: 1) the credible attribute weight configuration has different attention points to the credible attribute of knowledge in different fields, for example, people pay more attention to the timeliness of knowledge for weather and weather, and people pay more attention to the correctness of knowledge for the medical field. 2) And (4) trusted evidence configuration, a user can select trusted evidence participating in knowledge credibility evaluation according to requirements, and unselected trusted evidence does not participate in credibility evaluation of knowledge. Under the default configuration, the weight of the five credible attributes is 0.2, and all credible evidences are used for credible evaluation of knowledge. The customizable knowledge credibility assessment method enables the calculated knowledge credibility to better meet the requirements of corresponding application scenarios.
S300 comprises the following steps:
calculating the confidence C of the knowledge model byr:
Cr=α1R'+α2C'+α3T'+α4S'+α5A',
α1+α2+α3+α4+α5=1,
Wherein alpha is1Weight for reliability R', α2Weight for correctness C', α3As a weight of the timeliness T', alpha4Weight for security S', α5Is the weight of accessibility H.
The weights of the five credible attributes in the knowledge credible attribute model are determined by a user, and the weights of the five credible attributes are 0.2 under the default condition. Due to the continuous updating and evolution of the domain knowledge model, the local model or the whole model becomes more and more huge and complex, and if the confidence level of the knowledge model is recalculated after the domain knowledge model is changed every time, not only can the calculation resources be wasted, but also the usability of the domain knowledge model is influenced. Therefore, iterative incremental knowledge model confidence calculations need to be performed.
Performing iterative incremental calculation on the credibility of the knowledge model, wherein the credibility attribute characteristic value of newly added and deleted knowledge entities is less than 1,2k,Ck,Tk,Sk,Ak>. includes:
(1) measuring the reliability of the newly added and deleted knowledge entities: calculating the reliability characteristic values of the added and deleted knowledge entities by the following formula:
wherein N is the number of the original knowledge entities, R is the average reliability of the original knowledge entities, R' is the average reliability of the newly added knowledge entities, RiReliability characteristic value of the ith knowledge entity;
(2) measuring the correctness of the addition and deletion of the knowledge entities: calculating the correctness characteristic values of the added and deleted knowledge entities by the following formula:
wherein N is the number of the original knowledge entities, C is the average correctness of the original knowledge entities, C' is the average correctness of the newly added knowledge entities, CiThe correctness characteristic value of the ith knowledge entity is obtained;
(3) measuring timeliness of adding and deleting knowledge entities: calculating the timeliness characteristic values of the added and deleted knowledge entities by the following formula:
T=min(T,T'),T'=min(T1,T2,...,Tk),
wherein T is the timeliness of the original knowledge model, T' is the minimum timeliness of the newly added knowledge entity, TkThe timeliness of the kth newly added knowledge entity;
(4) measuring the safety of adding and deleting knowledge entities: calculating the security characteristic values of the added and deleted knowledge entities by the following formula:
S=min(S,S'),S'=min(S1,S2,...,Sk),
wherein S is the security of the original knowledge model, S' is the minimum security of the newly added knowledge entity, SkThe safety characteristic value of the kth newly added knowledge entity is obtained;
(5) measuring the accessibility of the added and deleted knowledge entities: the accessibility feature value of the newly added knowledge entity is calculated by:
the accessibility feature value of the deleted knowledge entity is calculated by:
where θ is the accessibility threshold, k is the number of knowledge entities, AmThe accessibility feature value for the mth deleted knowledge entity.
S400, based on the satisfaction degree of the knowledge entity or the knowledge model to the credibility attribute expected by the user and the support degree of the credibility evidence of the life cycle of the knowledge model to the credibility of the knowledge entity, the credibility grade is divided for the knowledge entity or the knowledge model.
S400 includes:
the credibility of knowledge is divided into five grades, which are respectively: the method comprises the following steps of classifying the credibility level of a knowledge entity according to rules, wherein the credibility level comprises an unknown level, a existence level, a confirmation level, a utility level and an evaluation level, and the classification comprises the following steps:
1) if the evidence missing rate of the knowledge entity is more than 80 percent, and the evidence missing rate is not common knowledge and axiom, or the calculated credibility is less than 0.2, defining the credibility level of the knowledge entity as unknown level;
2) if the evidence of knowledge credibility exists, the calculated credibility is less than 0.5, but no credible attribute which can be verified exists, the credibility level of the knowledge entity is defined as unknown level;
3) if the credible attribute of the knowledge entity has verifiable credible evidence and the characteristic value of the credible attribute is greater than 0.5, defining the credible grade of the knowledge entity as a practical grade;
4) if the evidence of the knowledge entity also comprises the evidence of the successful application case on the basis of meeting the condition of the confirmation level, defining the credibility level of the knowledge entity as a practical level;
5) and if the evidence of the knowledge entity also comprises evidence which passes the credibility analysis of the independent authority verification and analysis organization, defining the credibility level of the knowledge entity as an evaluation level.
As shown in fig. 6, a diagram of a credibility classification model for classifying the credibility of the knowledge entity or knowledge model. The confidence level of knowledge is a quantitative scale of confidence of knowledge. And according to the satisfaction degree of the knowledge or the knowledge model to the credibility attribute expected by the user and the support degree of credibility evidence of the knowledge model life cycle to the credibility of the knowledge. And (4) by taking the credibility grading standard of the software as reference, grading different grades according to the satisfied degree and the credibility evaluation value. The credibility of knowledge is divided into 5 grades which are respectively named as: unknown level, presence level, attestation level, utility level, and assessment level. And dividing the credibility level of the knowledge according to rules: 1) the evidence about the credibility of the knowledge is few, the lack rate of the evidence is more than 80%, the credibility of the evidence is not common knowledge and axiom, or the credibility of the calculation is less than 0.2, and the credibility level of the knowledge is defined as unknown level; 2) there is evidence of knowledge trustworthiness, the calculated trustworthiness is less than 0.5, but there are no trustworthy attributes that can be verified, and the trustworthiness level of knowledge is defined as unknown. 3) The credibility attribute of the knowledge entity has verifiable credibility evidence, the characteristic value of the credibility attribute is larger than 0.5, and the credibility grade of the knowledge entity is defined as a practical grade. 4) On the basis of meeting the condition of the confirmation level, the evidence of the knowledge entity also comprises the evidence of the successful application case, and the credibility level of the knowledge entity is defined as the utility level. 5) The evidence of the knowledge entity also comprises evidence of credible analysis of an independent authority verification and analysis organization, and the credibility level of the knowledge entity is defined as an evaluation level.
By adopting the method of the embodiment of the invention, the credibility connotation of the knowledge is split by using the credibility attribute model of the knowledge, then the credibility attribute model is used for collecting credibility-related data and information in the complete life cycle of the knowledge as credibility evidence to evaluate the credibility attribute value of the knowledge, and then the credibility attribute value of the knowledge is integrated to obtain the credibility of the knowledge, so that the credibility of the knowledge in the knowledge map can be comprehensively and quantitatively evaluated in a multi-dimensional manner.
It should be noted that, for simplicity of description, the above-mentioned method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the present invention is not limited by the order of acts, as some steps may occur in other orders or concurrently in accordance with the invention. Further, those skilled in the art should also appreciate that the embodiments described in the specification are preferred embodiments and that the acts and modules referred to are not necessarily required by the invention.
According to another aspect of the embodiments of the present invention, there is also provided a system for credible assessment of urban area knowledge base, as shown in fig. 7, including:
the evidence obtaining module 100 is configured to obtain credible evidence of a knowledge obtaining stage, a knowledge reasoning stage and a knowledge verification stage from the credible evidence model of the domain knowledge;
the first computing module 200 is configured to compute the credible attribute characteristic values of the knowledge entity and the knowledge model based on the obtained credible evidence and a computing method of the credible attribute characteristic values in the domain knowledge credible attribute model, where the credible attributes include reliability, correctness, timeliness, security, and accessibility;
the second calculation module 300 is configured to calculate the credibility of the knowledge entity based on the credibility attribute feature value of the knowledge entity, or calculate the credibility of the knowledge model based on the credibility attribute feature value of the knowledge model;
the ranking module 400 is configured to rank the knowledge entity or the knowledge model based on the credibility of the knowledge entity or the knowledge model and the support degree of the credibility evidence of the lifecycle of the knowledge model on the credibility of the knowledge entity.
It should be noted that the urban domain knowledge graph credible evaluation system and the urban domain knowledge graph credible evaluation method belong to the same inventive concept, and the specific implementation mode is not described any more.
By adopting the system provided by the embodiment of the invention, the credibility connotation of knowledge is split by using the credibility attribute model of knowledge, credibility-related data and information in the complete life cycle of knowledge are collected by using the credibility attribute model as credibility evidence to evaluate the credibility attribute value of knowledge, and then the credibility attribute value of knowledge is synthesized, so that the credibility of knowledge in the knowledge map can be comprehensively and quantitatively evaluated in a multi-dimensional manner.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is intended to include such modifications and variations.
Claims (10)
1. A credible assessment method for a knowledge graph in the urban field is characterized by comprising the following steps:
s100, acquiring credible evidences of a knowledge acquisition stage, a knowledge reasoning stage and a knowledge verification stage from a credible evidence model of the domain knowledge;
s200, calculating the credible attribute characteristic values of the knowledge entity and the knowledge model based on the obtained credible evidence and a calculation method of the credible attribute characteristic values in the domain knowledge credible attribute model, wherein the credible attributes comprise reliability, correctness, timeliness, safety and accessibility;
s300, calculating the credibility of the knowledge entity based on the credibility attribute characteristic value of the knowledge entity, or calculating the credibility of the knowledge model based on the credibility attribute characteristic value of the knowledge model;
and S400, based on the credibility of the knowledge entity or the knowledge model and the support degree of the credibility evidence of the life cycle of the knowledge model to the credibility of the knowledge entity, dividing the credibility level of the knowledge entity or the knowledge model.
2. The method of claim 1, wherein the knowledge entity comprises knowledge concepts, knowledge instances, knowledge attributes and knowledge relationships, and the calculating the credible attribute feature value of the knowledge entity comprises:
(1) measuring the reliability of the knowledge entity: calculating a reliability characteristic value R of the kth knowledge entity by:
β=max(eds)*max(edt)-min(eds)*min(edf),
wherein e isdsAs evidence of data origin, edfAs evidence of data type, edtFor the data time evidence, beta is a normalization factor, and N is the number of data sources;
(2) measuring the correctness of the knowledge entity: the correctness feature value C of the k-th knowledge entity is calculated by:
C=a1edq+a2(era*err)+a3max(euv,etr),
edq=(1-edd)*(1-edr)*edc+ecp*ech*eem+ete*eec,
wherein e isdqAs evidence of data quality, euvFor the user to visit the evidence, eraFor reasoning about the proof of algorithm performance, errFor reasoning about the evidence of the result, etrVerifying the reported evidence for a third party, a1,a2,a3The weights of the knowledge acquisition stage, the knowledge inference stage and the knowledge verification stage are respectively set as default average values 1/3, eddAs evidence of data loss rate, edrAs evidence of data redundancy rate, edcFor proof of concept matching confidence, ecpExtracting evidence for a concept, echFor proof of concept hierarchy relationship, eemFor element fusion matching probability, eteFor triple formation confidence evidence, eecExtracting algorithm confidence evidence;
(3) measuring the timeliness of the knowledge entity: calculating the timeliness characteristic value T of the k-th knowledge entity by the following formula:
T=max(edt,eee),
wherein e isdtAs evidence of data time, eeeEditing evidence for an expert;
(4) measuring the safety of the knowledge entity: calculating a security feature value S for the kth knowledge entity by:
S=1-erd*(1-euf),
wherein e isrdAs evidence of the degree of risk, eufFeeding back evidence for the user;
(5) measuring accessibility of a knowledge entity: the accessibility feature value a of the k-th knowledge entity is calculated by:
A=max(edm),
wherein e isdmIs field evidence.
3. The method according to claim 1 or 2, wherein the knowledge model comprises one or more knowledge entities, and the calculating of the credible attribute feature value of the knowledge model comprises:
(1) measuring the reliability of the knowledge model: the reliability characteristic value R' of the knowledge model is calculated by:
wherein k is the number of knowledge entities, RiReliability characteristic value of the ith knowledge entity;
(2) measuring the correctness of the knowledge model: the correctness feature C' of the knowledge model is calculated by:
wherein k is the number of knowledge entities, CiThe correctness characteristic value of the ith knowledge entity is obtained;
(3) measuring the timeliness of the knowledge model: calculating the timeliness characteristic value T' of the knowledge model by the following formula:
T'=min(T1,T2,...,Tk),
wherein k is the number of knowledge entities, CkThe timeliness characteristic value of the k knowledge entity is obtained;
(4) measuring the safety of the knowledge model: the safety feature value S' of the knowledge model is calculated by:
S'=min(S1,S2,...,Sk),
wherein k is the number of knowledge entities, SkThe safety characteristic value of the k-th knowledge entity;
(5) measuring accessibility of the knowledge model: the accessibility feature value H of the knowledge model is calculated by:
H={m|Am≥θ,m=1,2,...,k}
where θ is the accessibility threshold, k is the number of knowledge entities, AmIs the accessibility feature value of the mth knowledge entity.
4. The method of claim 1, wherein S300 comprises:
calculating the trustworthiness KT of the knowledge entity by:
KT=αRR+αCC+αTT+αSS+αAA,
αR+αC+αT+αS+αA=1,
wherein alpha isRIs the weight, alpha, of the reliability characteristic value R of the knowledge entityCIs the weight, alpha, of the correctness feature C of the knowledge entityTIs the weight, alpha, of the time-dependent characteristic value T of the knowledge entitySIs the weight, alpha, of the security feature value S of the knowledge entityAIs the weight of the accessibility feature value A of the knowledge entity.
5. The method of claim 4, wherein S300 further comprises:
and when the credibility of the knowledge entity is calculated, carrying out credible attribute weight configuration and credible evidence configuration according to user requirements, wherein the weights of the five credible attributes are all 0.2 under default configuration, and all credible evidences are used for credible evaluation of knowledge.
6. The method of claim 1, wherein S300 comprises:
calculating the confidence C of the knowledge model byr:
Cr=α1R'+α2C'+α3T'+α4S'+α5A',
α1+α2+α3+α4+α5=1,
Wherein alpha is1Is the weight, alpha, of the reliability characteristic value R' of the knowledge model2Is the weight, alpha, of the correctness feature C' of the knowledge model3Is the weight, alpha, of the time-dependent characteristic value T' of the knowledge model4Is the weight, alpha, of the security feature value S' of the knowledge model5Is the weight of the accessibility feature value H of the knowledge model.
7. The method of claim 6, wherein S300 further comprises:
performing iterative incremental calculation on the credibility of the knowledge model, wherein the credibility attribute characteristic value of newly added and deleted knowledge entities is less than 1,2k,Ck,Tk,Sk,Ak>. includes:
(1) measuring the reliability of the newly added and deleted knowledge entities: calculating the reliability characteristic values of the added and deleted knowledge entities by the following formula:
wherein N is the number of the original knowledge entities, R is the average reliability of the original knowledge entities, R' is the average reliability of the newly added knowledge entities, RiReliability characteristic value of the ith knowledge entity;
(2) measuring the correctness of the addition and deletion of the knowledge entities: calculating the correctness characteristic values of the added and deleted knowledge entities by the following formula:
wherein N is the number of the original knowledge entities, C is the average correctness of the original knowledge entities, C' is the average correctness of the newly added knowledge entities, CiThe correctness characteristic value of the ith knowledge entity is obtained;
(3) measuring timeliness of adding and deleting knowledge entities: calculating the timeliness characteristic values of the added and deleted knowledge entities by the following formula:
T=min(T,T'),T'=min(T1,T2,...,Tk),
wherein T is the timeliness of the original knowledge model, T' is the minimum timeliness of the newly added knowledge entity, TkThe timeliness of the kth newly added knowledge entity;
(4) measuring the safety of adding and deleting knowledge entities: calculating the security characteristic values of the added and deleted knowledge entities by the following formula:
S=min(S,S'),S'=min(S1,S2,...,Sk),
wherein S is the security of the original knowledge model, S' is the minimum security of the newly added knowledge entity, SkThe safety characteristic value of the kth newly added knowledge entity is obtained;
(5) measuring the accessibility of the added and deleted knowledge entities: the accessibility feature value of the newly added knowledge entity is calculated by:
the accessibility feature value of the deleted knowledge entity is calculated by:
where θ is the accessibility threshold, k is the number of knowledge entities, AmThe accessibility feature value for the mth deleted knowledge entity.
8. The method according to any one of claims 4-7, wherein S400 comprises:
the credibility of knowledge is divided into five grades, which are respectively: the method comprises the following steps of classifying the credibility level of a knowledge entity according to rules, wherein the credibility level comprises an unknown level, a existence level, a confirmation level, a utility level and an evaluation level, and the classification comprises the following steps:
1) if the evidence missing rate of the knowledge entity is more than 80 percent, and the evidence missing rate is not common knowledge and axiom, or the calculated credibility is less than 0.2, defining the credibility level of the knowledge entity as unknown level;
2) if the evidence of knowledge credibility exists, the calculated credibility is less than 0.5, but no credible attribute which can be verified exists, the credibility level of the knowledge entity is defined as unknown level;
3) if the credible attribute of the knowledge entity has verifiable credible evidence and the characteristic value of the credible attribute is greater than 0.5, defining the credible grade of the knowledge entity as a practical grade;
4) if the evidence of the knowledge entity also comprises the evidence of the successful application case on the basis of meeting the condition of the confirmation level, defining the credibility level of the knowledge entity as a practical level;
5) and if the evidence of the knowledge entity also comprises evidence which passes the credibility analysis of the independent authority verification and analysis organization, defining the credibility level of the knowledge entity as an evaluation level.
9. A credible evaluation system of a knowledge graph in the urban area is characterized by comprising the following steps:
the evidence obtaining module is used for obtaining credible evidences of a knowledge obtaining stage, a knowledge reasoning stage and a knowledge verification stage from the credible evidence model of the domain knowledge;
the first calculation module is used for calculating the credible attribute characteristic values of the knowledge entity and the knowledge model based on the acquired credible evidence and a calculation method of the credible attribute characteristic values in the domain knowledge credible attribute model, wherein the credible attributes comprise reliability, correctness, timeliness, safety and accessibility;
the second calculation module is used for calculating the credibility of the knowledge entity based on the credibility attribute characteristic value of the knowledge entity or calculating the credibility of the knowledge model based on the credibility attribute characteristic value of the knowledge model;
and the grade dividing module is used for dividing the credibility grade of the knowledge entity or the knowledge model based on the credibility of the knowledge entity or the knowledge model and the support degree of the credibility evidence of the life cycle of the knowledge model to the credibility of the knowledge entity.
10. The system of claim 9, wherein the knowledge entity comprises knowledge concepts, knowledge instances, knowledge attributes, and knowledge relationships, and wherein computing the trustworthy attribute feature values for the knowledge entity comprises:
(1) measuring the reliability of the knowledge entity: calculating a reliability characteristic value R of the kth knowledge entity by:
β=max(eds)*max(edt)-min(eds)*min(edf),
wherein e isdsAs evidence of data origin, edfAs evidence of data type, edtFor the data time evidence, beta is a normalization factor, and N is the number of data sources;
(2) measuring the correctness of the knowledge entity: the correctness feature value C of the k-th knowledge entity is calculated by:
C=a1edq+a2(era*err)+a3max(euv,etr),
edq=(1-edd)*(1-edr)*edc+ecp*ech*eem+ete*eec,
wherein e isdqAs evidence of data quality, euvFor the user to visit the evidence, eraFor reasoning about the proof of algorithm performance, errFor reasoning about the evidence of the result, etrVerifying the reported evidence for a third party, a1,a2,a3The weights of the knowledge acquisition stage, the knowledge inference stage and the knowledge verification stage are respectively set as default average values 1/3, eddAs evidence of data loss rate, edrAs evidence of data redundancy rate, edcFor proof of concept matching confidence, ecpExtracting evidence for a concept, echFor proof of concept hierarchy relationship, eemFor element fusion matching probability, eteFor triple formation confidence evidence, eecExtracting algorithm confidence evidence;
(3) measuring the timeliness of the knowledge entity: calculating the timeliness characteristic value T of the k-th knowledge entity by the following formula:
T=max(edt,eee),
wherein e isdtAs evidence of data time, eeeEditing evidence for an expert;
(4) measuring the safety of the knowledge entity: calculating a security feature value S for the kth knowledge entity by:
S=1-erd*(1-euf),
wherein e isrdAs evidence of the degree of risk, eufFeeding back evidence for the user;
(5) measuring accessibility of a knowledge entity: the accessibility feature value a of the k-th knowledge entity is calculated by:
A=max(edm),
wherein e isdmIs field evidence.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110890875.3A CN113792114A (en) | 2021-08-04 | 2021-08-04 | Credible evaluation method and system for urban field knowledge graph |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110890875.3A CN113792114A (en) | 2021-08-04 | 2021-08-04 | Credible evaluation method and system for urban field knowledge graph |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113792114A true CN113792114A (en) | 2021-12-14 |
Family
ID=79181387
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110890875.3A Pending CN113792114A (en) | 2021-08-04 | 2021-08-04 | Credible evaluation method and system for urban field knowledge graph |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113792114A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114239741A (en) * | 2021-12-21 | 2022-03-25 | 中国人民解放军国防科技大学 | Medical data classification method based on evidence reasoning classifier and related equipment |
-
2021
- 2021-08-04 CN CN202110890875.3A patent/CN113792114A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114239741A (en) * | 2021-12-21 | 2022-03-25 | 中国人民解放军国防科技大学 | Medical data classification method based on evidence reasoning classifier and related equipment |
CN114239741B (en) * | 2021-12-21 | 2024-03-29 | 中国人民解放军国防科技大学 | Medical data classification method and related equipment based on evidence reasoning classifier |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111612039B (en) | Abnormal user identification method and device, storage medium and electronic equipment | |
WO2019019630A1 (en) | Anti-fraud identification method, storage medium, server carrying ping an brain and device | |
CN114598539B (en) | Root cause positioning method and device, storage medium and electronic equipment | |
CN111563103B (en) | Method and system for detecting data blood relationship | |
Huang et al. | Technology life cycle analysis: From the dynamic perspective of patent citation networks | |
CN111309822A (en) | User identity identification method and device | |
CN113849702B (en) | Method and device for determining target data, electronic equipment and storage medium | |
CN111428095B (en) | Graph data quality verification method and graph data quality verification device | |
CN117078048A (en) | Digital twinning-based intelligent city resource management method and system | |
CN115378988A (en) | Data access abnormity detection and control method and device based on knowledge graph | |
CN113792114A (en) | Credible evaluation method and system for urban field knowledge graph | |
EP3901791A1 (en) | Systems and method for evaluating identity disclosure risks in synthetic personal data | |
CN112950359A (en) | User identification method and device | |
CN117194382A (en) | Middle-stage data processing method and device, electronic equipment and storage medium | |
CN115952150A (en) | Multi-source heterogeneous data fusion method and device | |
CN114495137A (en) | Bill abnormity detection model generation method and bill abnormity detection method | |
CN114493853A (en) | Credit rating evaluation method, credit rating evaluation device, electronic device and storage medium | |
Abdallah et al. | A Data Collection Quality Model for Big Data Systems | |
CN113792749A (en) | Time series data abnormity detection method, device, equipment and storage medium | |
CN117539920B (en) | Data query method and system based on real estate transaction multidimensional data | |
CN111625727A (en) | Information processing method and device for social relationship data and storage medium | |
CN117745080B (en) | Multi-factor authentication-based data access control and security supervision method and system | |
Chen et al. | Operational scenario definition in traffic simulation-based decision support systems: Pattern recognition using a clustering algorithm | |
CN114722061B (en) | Data processing method and device, equipment and computer readable storage medium | |
CN113723522B (en) | Abnormal user identification method and device, electronic equipment and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20211214 |