CN117076957A - Personnel identity association method and system based on multi-mode information - Google Patents

Personnel identity association method and system based on multi-mode information Download PDF

Info

Publication number
CN117076957A
CN117076957A CN202311329995.1A CN202311329995A CN117076957A CN 117076957 A CN117076957 A CN 117076957A CN 202311329995 A CN202311329995 A CN 202311329995A CN 117076957 A CN117076957 A CN 117076957A
Authority
CN
China
Prior art keywords
identity
association
information
identity information
class
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
Application number
CN202311329995.1A
Other languages
Chinese (zh)
Inventor
高辉
骆健儿
卢君
文卫疆
万俊翔
阳婧
徐佳
易瑶
周元广
李炎
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hunan Intelligent Police Public Security Technology Research Institute Co ltd
Original Assignee
Hunan Intelligent Police Public Security Technology Research Institute Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Hunan Intelligent Police Public Security Technology Research Institute Co ltd filed Critical Hunan Intelligent Police Public Security Technology Research Institute Co ltd
Priority to CN202311329995.1A priority Critical patent/CN117076957A/en
Publication of CN117076957A publication Critical patent/CN117076957A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computing arrangements using knowledge-based models
    • G06N5/02—Knowledge representation; Symbolic representation
    • G06N5/022—Knowledge engineering; Knowledge acquisition
    • G06N5/025—Extracting rules from data

Abstract

The invention provides a personnel identity association method and system based on multi-mode information, and relates to the technical field of personnel identity information association. The method comprises the steps of obtaining multi-mode information of personnel identities under different carrier objects, and performing type division analysis on the multi-mode information to form multi-mode identity type data; carrying out matching analysis on the identity information of different personnel under different carrier objects according to the identity activity information of each personnel on the carrier objects and combining the multi-mode identity type data to form identity matching result data; and establishing an identity association analysis model according to the identity matching result data and the multi-mode identity type data, and carrying out association matching on the personnel identity information in different carrier objects according to the identity association analysis model to form association result data. The method improves the efficiency of identity association and the accuracy of the identity association result through reasonable association mode design.

Description

Personnel identity association method and system based on multi-mode information
Technical Field
The invention relates to the technical field of personnel identity information association, in particular to a personnel identity association method and system based on multi-mode information.
Background
The personnel information of different social network platforms can not be the same as data can not be carried out due to no relevance, so that the invalid occupancy rate of network resources is effectively reduced, the cost of network operation is increased, unified management is inconvenient, targeted information service can not be carried out due to the fact that the identity information is associated with the data information of different social network platforms, and the quality and efficiency of network service are reduced.
Thus, the association of identity information is particularly important. Of course, different social network platforms have modality type identity information data due to the established basic conditions, and thus the complexity of identity association is increased. Most of the identity association modes are completed based on massive analysis of big data, so that on one hand, the time cost and the resource use cost of the identity association are increased, and on the other hand, the efficient and reasonable association analysis mode is not provided, and the accuracy of an association result cannot be improved.
Therefore, designing a personnel identity association method and system based on multi-mode information, through reasonable association mode design, not only is the efficiency of identity association improved, but also the accuracy of the identity association result is improved, and the method and system are the problems to be solved urgently at present.
Disclosure of Invention
The invention aims to provide a personnel identity correlation method based on multi-mode information, which is used for reasonably dividing types based on matching correlation analysis by acquiring the personnel identity multi-mode information under different carrier objects, so that the multi-mode personnel identity information can be effectively utilized, and the analysis efficiency and accuracy of personnel identity correlation can be improved. Meanwhile, based on multi-mode identity type data, a reasonable identity association method model is established, and association of personnel identities on different carrier objects is accurately achieved in a more reasonable and orderly mode. The efficiency and accuracy of identity association are further improved.
The invention also aims to provide a personnel identity association system based on the multi-mode information, which can collect the multi-mode data enough for identity association through the multi-mode information acquisition unit, expand the data types required by the identity association analysis and further improve the accuracy of the identity association. A reasonable identity association analysis hardware system is established by utilizing the matching analysis unit and the identity association analysis unit, so that the identity association analysis is accurately and efficiently carried out, and a necessary hardware foundation is provided for the realization of the identity association.
In a first aspect, the invention provides a personnel identity association method based on multi-modal information, which comprises the steps of obtaining multi-modal information of personnel identities under different carrier objects, and performing type division analysis on the multi-modal information to form multi-modal identity type data; carrying out matching analysis on the identity information of different personnel under different carrier objects according to the identity activity information of each personnel on the carrier objects and combining the multi-mode identity type data to form identity matching result data; and establishing an identity association analysis model according to the identity matching result data and the multi-mode identity type data, and carrying out association matching on the personnel identity information in different carrier objects according to the identity association analysis model to form association result data.
According to the method, reasonable type division based on matching correlation analysis is carried out by acquiring the personnel identity multi-mode information under different carrier objects, so that the multi-mode personnel identity information is effectively utilized, and the analysis efficiency and accuracy of personnel identity correlation are improved. Meanwhile, based on multi-mode identity type data, a reasonable identity association method model is established, and association of personnel identities on different carrier objects is accurately achieved in a more reasonable and orderly mode. The efficiency and accuracy of identity association are further improved.
As a possible implementation manner, acquiring multi-mode information of identities of people under different carrier objects, performing type division analysis on the multi-mode information, and forming multi-mode identity type data, including: the multi-mode information of each personnel identity in different carrier objects is acquired, and the multi-mode information is divided into the following types aiming at identity judgment: the information of whether the multi-mode information is judged in the accuracy judgment is divided into a judging type identity information data set A; the information which is selectively judged in the accuracy judgment of the multi-mode information is divided into a selection type identity information data set B; and analyzing other information except the discrimination type identity information data set A and the selection type identity information data set B in the multi-mode information based on identity recognition to form orientation type identity information data C.
In the invention, two aspects are mainly considered in the type division of the multi-mode information, and on one hand, when identity association is carried out, the association degree can be accurately and reasonably determined by combining data of the types. It will be appreciated that for multimodal identity information on carrier objects, the most important content is to establish the characteristics of a person object so as to quickly identify among a plurality of person objects, and to establish the characteristics of the person object, it is necessary to obtain the characteristic data of the person object, where the important characteristic data includes but is not limited to basic biological characteristics, character characteristics, habits, and so on. Therefore, the characteristics are classified, so that accurate association judgment can be performed through the characteristics when identity association is performed, and the accuracy of association judgment is improved. Another aspect is that the divided type data needs to be considered, so that the analysis efficiency can be improved and the analysis difficulty can be reduced when the identity association analysis is carried out later. Compared with the general analysis form of big data, the multi-mode information is divided into the discrimination class, the selection class and the orientation class, the association judgment can be quickly carried out by means of three different types of data, and if the comparison of a large amount of data is achieved, the discrimination class and the selection class are simpler and quicker in judgment and analysis, and the obtained result has great accuracy probability. And the reasonable description of the characteristics of the personnel object can be realized for the data of the orientation class, so that the efficiency of the association analysis is improved. It should be noted here that, because the identity association not only appears between social network platforms, but also involves identity association of devices and identity association of information, in a similar manner, the effect of half effort of association analysis can be achieved by reasonably classifying multi-modal data of objects, so that the carrier object is not limited to social network platforms, but also can be other objects suitable for the association method.
As a possible implementation manner, in the discrimination class identity information data set a, a single discrimination class identity information set for directly determining the identity of a person is taken as a decisive discrimination class identity information data subsetWherein: />=[/>，/>，…，/>]M is decisive discrimination class identity information data subset +.>The number of discrimination class identity information; collecting other discrimination class identity information except the decisive discrimination class identity information data in the discrimination class identity information data as an auxiliary discrimination class identity information data subset +.>Wherein: />=[/>，/>，…，/>]N is auxiliary discrimination class identity information data subset +.>The number of discrimination class identity information included in the data.
In the present invention, from the viewpoint of identifying the person object in the carrier object, there is also a significant difference in the judgment effect in judging the class identity information. For example, identification type identity information having a strong one-to-one correspondence such as an identity card number, a mobile phone number, iris information, fingerprint information, etc., can be sufficiently judged whether or not it is an identity object that can be associated by only one identity information data. And discrimination class identity information with similar gender, name and the like but with unobvious correspondence needs to be discriminated. Since the two types of identity data have different degrees of influence on the relevance analysis, a distinction is required when performing type classification.
As one possible implementation manner, performing analysis based on identity recognition on other information except the discrimination class identity information data set a and the selection class identity information data set B in the multi-mode information to form orientation class identity information data C, including: in the orientation type identity information data set C, the orientation type identity information of different mode types is separated, the orientation type identity information of each mode type is extracted to form an orientation type identity information data subset containing a plurality of identity characteristics under different mode types, and the orientation type identity information data subset is used for acquiring the orientation type identity information data of different mode typesThe number of the identity features in the subset is that the orientation type identity information data subset is ordered from a more to a less sequence to form a sequence set of the orientation type identity information data subset, namely: c= [，/>，…，/>]Wherein->=[/>，/>，…，/>]U is the sequence number of the orientation class identity information data subset, and v is the number of the identity features in the orientation class identity information data subset with the sequence number u.
In the invention, the orientation type identity information is mainly divided according to the habit, preference and other oriented behaviors of personnel objects. In this way, the behavior information of the tendency of the same person is kept substantially unchanged, since the same person releases the information on different carrier objects. Thus, orientation class identity information containing tendencies is also an important factor affecting identity correlation. Of course, it is more important to divide and determine the behavior information of the tendency, and reasonably analyze the behavior information to determine the characteristics of the person object, for example, the person object prefers to exercise, but the person object prefers to exercise, and the person object has updated orientation characteristics, such as basketball, football, and the like.
As a possible implementation manner, according to the identity activity information of each personnel object on the carrier object and combining the multi-mode identity type data, different personnel identity information under different carrier objects is carried outAnd (3) matching performance analysis, namely forming identity matching result data, comprising the following steps: acquiring identity activity information of personnel objects on carrier objects, and extracting and determining class identity information data subsetsThe words and sentences related to the discrimination identity information in the system are subjected to decisive judgment by carrying out semantic analysis on the words and sentences: determining and marking the discrimination splitting information for judging whether the inverse exists according to the word and sentence semantics; non-decisive marking is carried out on the identification information which can not judge whether the inverse is not judged according to the meaning of the words and sentences; acquiring identity activity information of personnel on a carrier object, and extracting and assisting in distinguishing class identity information data subset +.>The auxiliary judgment is carried out by carrying out semantic analysis on the words and sentences related to the identification information in the method: carrying out frequency statistics on two opposite results for judging identity information according to the judgment result exhibited by the meaning of the word and sentence, and marking, namely: />=[/>//>，/>//>，…，/>//>]，/>Representing the corresponding discrimination identity information Negative probability of the result of (2); acquiring identity activity information of a personnel object on a carrier object, extracting words and sentences related to the selected identity information in the selected identity information data set B, and carrying out selective judgment by carrying out semantic analysis on the words and sentences: determining the selection result of each piece of selection identity information according to the semantics of the words and sentences, carrying out probability statistics on effective decision times on the semantics of the words and sentences for deciding different selection results, and marking, namely: />=[/>//>，/>//>，…，/>//>]，And a negative probability representing the result exhibited by the corresponding selection identity information, i being the number of selection identity information present in the selection identity information data set B.
In the invention, the main purpose of matching analysis of personnel identity is to determine whether the information expressed by the personnel object on the carrier object is correct, and the probability is taken as the accuracy judgment result to reasonably establish basic data for judging whether the information is correct or not. Here, it should be noted that, due to the different carrier objects, the purpose of the operation is different, and thus, the acquired multi-modal data is different. However, when the identity correlation matching analysis is performed, the difference needs to be fully considered, so that when the probability statistics is performed by taking negative probability as a reference, the probability analysis is more reasonable, and after all, the probability of the existence of the identity correlation is high when the data matching comparison cannot be performed completely.
As a possible implementation manner, according to the identity matching result data and the multi-mode identity type data, an identity association analysis model is established, and according to the identity association analysis model, personnel identity information in different carrier objects is associated and matched to form association result data, including: determining two carrier objects to be associated, and setting the two carrier objects as a first associated carrier object and a second associated carrier object respectively; acquiring identity matching result data of a first associated carrier object and identity matching result data of a second associated carrier object; and taking the personnel object as a unit, randomly selecting identity matching result data of one personnel object from the first association carrier object, and acquiring the identity matching result data of the personnel object in the second association carrier object one by one to perform the following association analysis: subset two decisive discrimination class identity information dataThe identification information in (a) carries out association analysis: if the identification information with the same decisive mark exists and the identification information identification results are the same, the two personnel objects are directly subjected to identity association; otherwise, judging the identity information data subset based on assistance +. >And selecting a probability association analysis of the class identity information data set B and the orientation class identity information data set C, and forming association result data.
In the invention, the provided association analysis mode is to match the definitive discrimination type identity information, and the definitive discrimination type identity information is the discrimination of the inverse type or not, so that if the data capable of being matched exist, whether the association is to be carried out can be accurately judged, for example, after the identity numbers of the two identity cards are determined, if the identity numbers are consistent, the identity can be accurately determined as the same person. Therefore, the matching of the definitive discrimination class identity information can also greatly reduce the analysis quantity and improve the analysis efficiency. Of course, since the modal data have differences in different carrier objects, when the definitive discrimination class identity information cannot complete the matching analysis, comprehensive consideration needs to be performed based on other types of data.
As a possible implementation manner, based on assistance, the identity information data subset is distinguishedSelecting a probabilistic association analysis of the class identity information dataset B and the oriented class identity information dataset C and forming association result data, comprising: acquiring auxiliary discrimination class identity information data subset +. >Performing overall negative probability analysis to obtain judging class negative association probability +.>The method comprises the steps of carrying out a first treatment on the surface of the Acquiring a selection class identity information data set B of a personnel object under two carrier objects, and performing negative probability analysis based on the coincidence degree to obtain a selection class negative association probability +.>The method comprises the steps of carrying out a first treatment on the surface of the Acquiring an orientation class identity information data set C of a personnel object under two carrier objects, and performing negative probability analysis based on the coincidence degree of orientation class identity characteristics to obtain orientation class negative association probability ∈>The method comprises the steps of carrying out a first treatment on the surface of the Association probability of combining discriminant and negative>Selection class negative association probability->Orientation class negative association probability ++>Obtaining a negative association probability of two person objects according to the following formula->：
；
Setting a negative judgment threshold alpha, and acquiring all negative association probabilities of the person objects determined under the first association carrier object and the person objects corresponding to the second association carrier object one by oneIf there is a negative association probability +.>If the negative association probability value of the person object selected by the first association carrier object in the second association carrier object is less than or equal to alpha, determining the person object with the largest negative association probability value as the same association person, and carrying out identity association; otherwise, judging that the personnel object selected from the first association carrier object has no identity association in the second association carrier object, and continuously repeating the association judgment of establishing an identity association analysis model from the personnel object selected from the first association carrier object.
In the invention, when judging by utilizing other types of identity information data except the decisive judging type identity information, the influence of the types of identity information on the judging accuracy needs to be comprehensively considered by considering that any type of identity information data cannot be accurately determined as the decisive judging type identity information. The data among different carrier objects cannot achieve ideal full correspondence due to the difference of multi-mode data, and the data which does not achieve correlation analysis also has the possibility of correlation, so that the matching analysis is mainly carried out by carrying out negative probability calculation during analysis. The negative judgment threshold value can be determined based on historical analysis data so as to improve the reliability of correlation analysis based on the negative correlation probability, and avoid that the correlation error and unreasonable caused by correlating two personnel identities without correlation per se reduce the accuracy of the correlation analysis.
As a possible implementation manner, acquiring auxiliary discrimination class identity information data subsets of personnel objects under two carrier objectsPerforming overall negative probability analysis to obtain judging class negative association probability +. >Comprising: acquiring the discriminant class negative association probability +.f of the kth personnel object under the selected second association carrier object according to the following formula>：Wherein->Here, k is the probability of making a negative association +.>Person object under the second associated carrier object determined at the time of calculation.
In the invention, for obtaining the negative association probability of the auxiliary discrimination class identity information, because the auxiliary discrimination class identity information can provide unified data for all different carrier objects due to the network environment and policy requirements, for example, basic data of personnel objects can be collected by basically all network platforms, so that the basic data on different carrier objects can be unified, and therefore, when the calculation of the negative association probability is carried out, the impression of other used data on results is not needed to be considered, and the calculation of the negative association probability is very accurate.
As a possible implementation, the selection class identity information data set B of the person object under the two carrier objects is acquired, a negative probability analysis based on the degree of coincidence is performed,obtaining selection class negative association probabilitiesComprising the following steps: acquiring the total number of the selection identity information of the selection class identity information data set B of the personnel object under the first association carrier object and the kth personnel object selected under the second association carrier object, determining the coincidence rate x and the coincidence quantity r of the selection identity information, and determining the selection class negative association probability +. >：
；
Acquiring an orientation class identity information data set C of a personnel object under two carrier objects, and performing negative probability analysis based on the coincidence degree of orientation class identity characteristics to obtain orientation class negative association probabilityComprising: acquiring an orientation type identity information data set C of a person object under a first associated carrier object and a kth person object selected under a second associated carrier object, determining a superposed orientation type identity data subset, and sequencing the superposed orientation type identity data subsets according to the order of the number of identity features in the orientation type identity information data subsets from more to less to form a superposed orientation type identity information data set>Wherein: />=[/>，/>，…，/>]Z represents the coincident sequential label; probability statistics is carried out on the identity characteristics overlapped in each overlapped orientation type identity information data subset to form overlapped probability +.>The method comprises the steps of carrying out a first treatment on the surface of the Determining an orientation class negative association probability according to>：/>。
In the present invention, a selection class negative association probability of a selection class identity information dataset B is providedOrientation class negative association probability of orientation class identity information dataset C>A calculation method. For the data set B of the selection type identity information, the coincidence rate of the selection type identity information represents the volume of the selection type identity information capable of realizing the relevance analysis, a reasonable analysis is made, even if the calculation of the negative probability of the selection type identity information in different carrier objects cannot be completely carried out, the overall negative probability is very low in the case of being capable of correlating the identity information, otherwise, the negative probability of the selection type identity information with the coincidence property is very high, and then the negative probability trend of all the selection type identity information can be shown, so that the negative correlation probability of the selection type is improved when the selection type is carried out >During calculation of (2), the negative probability is amplified according to the duty ratio of the superposition number, so that more accurate correlation analysis is facilitated. Of course, the orientation class negative correlation probability for the orientation class identity information dataset C>It will be appreciated that the more identity features that occur, the greater the accuracy of the correlation analysis. Therefore, when negative probability analysis is performed, the probability calculation is performed after the identification features are ranked, and reasonable probability prediction is realized while the rationality of probability calculation is fully considered.
In a second aspect, the present invention provides a personnel identity association system based on multi-mode information, which is applied to the personnel identity association method based on multi-mode information in the first aspect, and includes: the multi-mode information acquisition unit is used for acquiring multi-mode information of different personnel objects in different carrier objects, and performing type division to form multi-mode identity type data; the matching analysis unit is used for acquiring the multi-mode identity type data of the multi-mode information acquisition unit, carrying out matching analysis and forming identity matching result data; the identity association analysis unit is used for carrying out association matching by combining the identity matching result data of the matching analysis unit and the multi-mode identity type data of the multi-mode information acquisition unit to complete identity association.
In the invention, the system can collect the multi-mode data which is enough for identity association through the multi-mode information acquisition unit, expands the data types required by the analysis of the identity association, and further improves the accuracy of the identity association. A reasonable identity association analysis hardware system is established by utilizing the matching analysis unit and the identity association analysis unit, so that the identity association analysis is accurately and efficiently carried out, and a necessary hardware foundation is provided for the realization of the identity association.
The personnel identity correlation system and method based on the multi-mode information provided by the invention have the beneficial effects that:
according to the method, the reasonable type division based on the matching correlation analysis is carried out by acquiring the personnel identity multi-mode information under different carrier objects, so that the multi-mode personnel identity information is effectively utilized, and the analysis efficiency and accuracy of personnel identity correlation are improved. Meanwhile, based on multi-mode identity type data, a reasonable identity association method model is established, and association of personnel identities on different carrier objects is accurately achieved in a more reasonable and orderly mode. The efficiency and accuracy of identity association are further improved.
The system can collect enough multi-mode data for identity association through the multi-mode information acquisition unit, expands the data types required by identity association analysis, and further improves the accuracy of identity association. A reasonable identity association analysis hardware system is established by utilizing the matching analysis unit and the identity association analysis unit, so that the identity association analysis is accurately and efficiently carried out, and a necessary hardware foundation is provided for the realization of the identity association.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments of the present invention will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and should not be considered as limiting the scope, and other related drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a step diagram of a personnel identity association method based on multi-mode information according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be described below with reference to the accompanying drawings in the embodiments of the present invention.
The personnel information of different social network platforms can not be the same as data can not be carried out due to no relevance, so that the invalid occupancy rate of network resources is effectively reduced, the cost of network operation is increased, unified management is inconvenient, targeted information service can not be carried out due to the fact that the identity information is associated with the data information of different social network platforms, and the quality and efficiency of network service are reduced.
Thus, the association of identity information is particularly important. Of course, different social network platforms have modality type identity information data due to the established basic conditions, and thus the complexity of identity association is increased. Most of the identity association modes are completed based on massive analysis of big data, so that on one hand, the time cost and the resource use cost of the identity association are increased, and on the other hand, the efficient and reasonable association analysis mode is not provided, and the accuracy of an association result cannot be improved.
Referring to fig. 1, an embodiment of the present invention provides a method for associating personnel identities based on multimodal information. According to the method, the reasonable type division based on the matching correlation analysis is carried out by acquiring the personnel identity multi-mode information under different carrier objects, so that the multi-mode personnel identity information is effectively utilized, and the analysis efficiency and accuracy of personnel identity correlation are improved. Meanwhile, based on multi-mode identity type data, a reasonable identity association method model is established, and association of personnel identities on different carrier objects is accurately achieved in a more reasonable and orderly mode. The efficiency and accuracy of identity association are further improved.
The personnel identity association method based on the multi-mode information specifically comprises the following steps:
s1: and acquiring multi-mode information of personnel identities under different carrier objects, and performing type division analysis on the multi-mode information to form multi-mode identity type data.
The method for obtaining the multi-modal information of the personnel identity under different carrier objects, performing type division analysis on the multi-modal information, and forming multi-modal identity type data comprises the following steps: the multi-mode information of each personnel identity in different carrier objects is acquired, and the multi-mode information is divided into the following types aiming at identity judgment: the information of whether the multi-mode information is judged in the accuracy judgment is divided into a judging type identity information data set A; the information which is selectively judged in the accuracy judgment of the multi-mode information is divided into a selection type identity information data set B; and analyzing other information except the discrimination type identity information data set A and the selection type identity information data set B in the multi-mode information based on identity recognition to form orientation type identity information data C.
The type division of the multi-mode information mainly considers two aspects, namely, when identity association is carried out, the association degree can be accurately and reasonably determined by combining data of the types. It will be appreciated that for multimodal identity information on carrier objects, the most important content is to establish the characteristics of a person object so as to quickly identify among a plurality of person objects, and to establish the characteristics of the person object, it is necessary to obtain the characteristic data of the person object, where the important characteristic data includes but is not limited to basic biological characteristics, character characteristics, habits, and so on. Therefore, the characteristics are classified, so that accurate association judgment can be performed through the characteristics when identity association is performed, and the accuracy of association judgment is improved. Another aspect is that the divided type data needs to be considered, so that the analysis efficiency can be improved and the analysis difficulty can be reduced when the identity association analysis is carried out later. Compared with the general analysis form of big data, the multi-mode information is divided into the discrimination class, the selection class and the orientation class, the association judgment can be quickly carried out by means of three different types of data, and if the comparison of a large amount of data is achieved, the discrimination class and the selection class are simpler and quicker in judgment and analysis, and the obtained result has great accuracy probability. And the reasonable description of the characteristics of the personnel object can be realized for the data of the orientation class, so that the efficiency of the association analysis is improved. It should be noted here that, because the identity association not only appears between social network platforms, but also involves identity association of devices and identity association of information, in a similar manner, the effect of half effort of association analysis can be achieved by reasonably classifying multi-modal data of objects, so that the carrier object is not limited to social network platforms, but also can be other objects suitable for the association method.
In the discrimination class identity information data set A, a single discrimination class identity information set for directly determining the identity of a person is taken as a decisive discrimination class identity information data subsetWherein: />=[/>，/>，…，/>]M is decisive discrimination class identity information data subset +.>The number of discrimination class identity information; collecting other discrimination class identity information except the decisive discrimination class identity information data in the discrimination class identity information data as an auxiliary discrimination class identity information data subset +.>Wherein:=[/>，/>，…，/>]n is auxiliary discrimination class identity information data subset +.>The number of discrimination class identity information included in the data.
From the viewpoint of identifying the person object in the carrier object, there is also a significant difference in the judgment effect in the judgment type identity information. For example, identification type identity information having a strong one-to-one correspondence such as an identity card number, a mobile phone number, iris information, fingerprint information, etc., can be sufficiently judged whether or not it is an identity object that can be associated by only one identity information data. And discrimination class identity information with similar gender, name and the like but with unobvious correspondence needs to be discriminated. Since the two types of identity data have different degrees of influence on the relevance analysis, a distinction is required when performing type classification.
Analyzing other information except the discrimination type identity information data set A and the selection type identity information data set B in the multi-mode information based on identity recognition to form orientation type identity information data C, wherein the method comprises the following steps: in the orientation type identity information data set C, separating orientation type identity information of different mode types, extracting identity characteristics of the orientation type identity information of each mode type to form an orientation type identity information data subset containing a plurality of identity characteristics under different mode types, and sequencing the orientation type identity information data subset according to the sequence of the number of the identity characteristics in the different orientation type identity information data subset from more to less to form a sequence set of the orientation type identity information data subset, namely: c= [，/>，…，/>]Wherein->=[/>，/>，…，/>]U is the sequence number of the orientation class identity information data subset, and v is the number of the identity features in the orientation class identity information data subset with the sequence number u.
The orientation type identity information is mainly divided according to habit, preference and other orientation behaviors of personnel and objects. In this way, the behavior information of the tendency of the same person is kept substantially unchanged, since the same person releases the information on different carrier objects. Thus, orientation class identity information containing tendencies is also an important factor affecting identity correlation. Of course, it is more important to divide and determine the behavior information of the tendency, and reasonably analyze the behavior information to determine the characteristics of the person object, for example, the person object prefers to exercise, but the person object prefers to exercise, and the person object has updated orientation characteristics, such as basketball, football, and the like.
S2: and carrying out matching analysis on the identity information of different personnel under different carrier objects according to the identity activity information of each personnel on the carrier objects and combining the multi-mode identity type data to form identity matching result data.
According to the identity activity information of each personnel object on the carrier object and combining the multi-mode identity type data, carrying out matching analysis on the identity information of different personnel under different carrier objects to form identity matching result data, wherein the method comprises the following steps: acquiring identity activity information of personnel objects on carrier objects, and extracting and determining class identity information data subsetsThe words and sentences related to the discrimination identity information in the system are subjected to decisive judgment by carrying out semantic analysis on the words and sentences: determining and marking the discrimination splitting information for judging whether the inverse exists according to the word and sentence semantics; non-decisive marking is carried out on the identification information which can not judge whether the inverse is not judged according to the meaning of the words and sentences; acquiring identity activity information of personnel on a carrier object, and extracting and assisting in distinguishing class identity information data subset +.>The auxiliary judgment is carried out by carrying out semantic analysis on the words and sentences related to the identification information in the method: carrying out frequency statistics on two opposite results for judging identity information according to the judgment result exhibited by the meaning of the word and sentence, and marking, namely: / >=[/>//>，/>//>，…，/>//>]，/>A negative probability representing a result exhibited by the corresponding discrimination identity information; acquiring identity activity information of a personnel object on a carrier object, extracting words and sentences related to the selected identity information in the selected identity information data set B, and carrying out selective judgment by carrying out semantic analysis on the words and sentences: determining the selection result of each piece of selection identity information according to the semantics of the words and sentences, carrying out probability statistics on effective decision times on the semantics of the words and sentences for deciding different selection results, and marking, namely: />=[/>//>，/>//>，…，/>//>]，/>And a negative probability representing the result exhibited by the corresponding selection identity information, i being the number of selection identity information present in the selection identity information data set B.
The main purpose of the matching analysis of the personnel identity is to determine whether the information expressed by the personnel object on the carrier object is correct or not, and the probability is taken as the accuracy judgment result to reasonably establish basic data for judging whether the information is correct or not. Here, it should be noted that, due to the different carrier objects, the purpose of the operation is different, and thus, the acquired multi-modal data is different. However, when the identity correlation matching analysis is performed, the difference needs to be fully considered, so that when the probability statistics is performed by taking negative probability as a reference, the probability analysis is more reasonable, and after all, the probability of the existence of the identity correlation is high when the data matching comparison cannot be performed completely.
S3: and establishing an identity association analysis model according to the identity matching result data and the multi-mode identity type data, and carrying out association matching on the personnel identity information in different carrier objects according to the identity association analysis model to form association result data.
The method comprises the following steps: determining two carrier objects to be associated, and setting the two carrier objects as a first associated carrier object and a second associated carrier object respectively; acquiring identity matching result data of a first associated carrier object and identity matching result data of a second associated carrier object; and taking the personnel object as a unit, randomly selecting identity matching result data of one personnel object from the first association carrier object, and acquiring the identity matching result data of the personnel object in the second association carrier object one by one to perform the following association analysis: subset two decisive discrimination class identity information dataThe identification information in (a) carries out association analysis: if there is a judgment with the same decisive flagIdentity information is distinguished, and the identity information is judged to be the same, so that identity association is directly carried out on two personnel objects; otherwise, judging the identity information data subset based on assistance +. >And selecting a probability association analysis of the class identity information data set B and the orientation class identity information data set C, and forming association result data.
The provided association analysis mode is to match the definitive discrimination class identity information, and the definitive discrimination class identity information is the discrimination of the inverse class or not, so that if the data capable of being matched exist, whether the association is to be carried out or not can be accurately judged, for example, after the identity numbers of the definitive discrimination class identity information and the definitive discrimination class identity information are determined, if the identity numbers are consistent, the identity numbers can be accurately determined as the same person. Therefore, the matching of the definitive discrimination class identity information can also greatly reduce the analysis quantity and improve the analysis efficiency. Of course, since the modal data have differences in different carrier objects, when the definitive discrimination class identity information cannot complete the matching analysis, comprehensive consideration needs to be performed based on other types of data.
Discrimination of class identity information data subsets based on assistanceSelecting a probabilistic association analysis of the class identity information dataset B and the oriented class identity information dataset C and forming association result data, comprising: acquiring auxiliary discrimination class identity information data subset +. >Performing overall negative probability analysis to obtain judging class negative association probability +.>The method comprises the steps of carrying out a first treatment on the surface of the Acquiring a selection class identity information data set B of a personnel object under two carrier objects, and performing negative probability analysis based on the coincidence degree to obtain a selection class negative association probability +.>The method comprises the steps of carrying out a first treatment on the surface of the Acquiring an orientation class identity information data set C of a personnel object under two carrier objects, and performing negative probability analysis based on the coincidence degree of orientation class identity characteristics to obtain orientation class negative association probabilityThe method comprises the steps of carrying out a first treatment on the surface of the Association probability of combining discriminant and negative>Selection class negative association probability->Orientation class negative association probability ++>Obtaining a negative association probability of two person objects according to the following formula->：
；
Setting a negative judgment threshold alpha, and acquiring all negative association probabilities of the person objects determined under the first association carrier object and the person objects corresponding to the second association carrier object one by oneIf there is a negative association probability +.>If the negative association probability value of the person object selected by the first association carrier object in the second association carrier object is less than or equal to alpha, determining the person object with the largest negative association probability value as the same association person, and carrying out identity association; otherwise, judging that the personnel object selected from the first association carrier object has no identity association in the second association carrier object, and continuing to repeatedly perform the analysis of establishing identity association from the personnel object selected from the first association carrier object And (5) judging the relevance of the model.
When judging by using other types of identity information data except the definitive judging type identity information, the influence of the types of identity information on the judging accuracy needs to be comprehensively considered by considering that any type of identity information data cannot be accurately determined as the definitive judging type identity information. The data among different carrier objects cannot achieve ideal full correspondence due to the difference of multi-mode data, and the data which does not achieve correlation analysis also has the possibility of correlation, so that the matching analysis is mainly carried out by carrying out negative probability calculation during analysis. The negative judgment threshold value can be determined based on historical analysis data so as to improve the reliability of correlation analysis based on the negative correlation probability, and avoid that the correlation error and unreasonable caused by correlating two personnel identities without correlation per se reduce the accuracy of the correlation analysis.
Acquiring auxiliary discrimination class identity information data subsets of personnel objects under two carrier objectsPerforming overall negative probability analysis to obtain judging class negative association probability +.>Comprising: acquiring the discriminant class negative association probability +.f of the kth personnel object under the selected second association carrier object according to the following formula >：/>Wherein, the method comprises the steps of, wherein,here, k is the probability of making a negative association +.>Person object under the second associated carrier object determined at the time of calculation.
For the acquisition of the negative association probability of the auxiliary discrimination class identity information, because the auxiliary discrimination class identity information can provide unified data for all different carrier objects due to the requirements of network environment and policy, for example, basic data of personnel objects can be acquired by basically all network platforms, so that the basic data on different carrier objects can be unified, and therefore, when the calculation of the negative association probability is carried out, the impression of other used data on results is not required to be considered, and the calculation of the negative association probability is quite accurate.
Acquiring a selection class identity information data set B of a personnel object under two carrier objects, and performing coincidence-based negative probability analysis to obtain a selection class negative association probabilityComprising the following steps: acquiring the total number of the selection identity information of the selection class identity information data set B of the personnel object under the first association carrier object and the kth personnel object selected under the second association carrier object, determining the coincidence rate x and the coincidence quantity r of the selection identity information, and determining the selection class negative association probability according to the following formula ：/>The method comprises the steps of carrying out a first treatment on the surface of the Acquiring an orientation class identity information data set C of a personnel object under two carrier objects, and performing negative probability analysis based on the coincidence degree of orientation class identity characteristics to obtain orientation class negative association probability ∈>Comprising: acquiring an orientation class identity information data set C of a person object under a first associated carrier object and a kth person object selected under a second associated carrier object, determining an overlapped orientation class identity data subset, and carrying out the sequence of the overlapped orientation class identity data subset according to the number of identity features in the orientation class identity information data subset from more to less on the orientation class identity information data subsetLine ordering to form coincidence orientation type identity information dataset +.>Wherein: />=[/>，/>，…，/>]Z represents the coincident sequential label; probability statistics is carried out on the identity characteristics overlapped in each overlapped orientation type identity information data subset to form overlapped probability +.>The method comprises the steps of carrying out a first treatment on the surface of the Determining an orientation class negative association probability according to>：。
Providing a class-of-selection negative association probability for a class-of-selection identity information dataset BOrientation class negative association probability of orientation class identity information dataset C>A calculation method. For the selection class identity information data set B, the coincidence ratio of the selection identity information represents the volume of the selection class identity information capable of realizing the association analysis, and a reasonable analysis is made, even if the negative probability of the selection class identity information in different carrier objects cannot be completely calculated, the selection class identity information is integrated under the condition that the association of the identity information is realized The negative probability of the selection class identity information with coincidence property can show the negative probability trend of all the selection class identity information, so the selection class negative association probability +.>During calculation of (2), the negative probability is amplified according to the duty ratio of the superposition number, so that more accurate correlation analysis is facilitated. Of course, the orientation class negative correlation probability for the orientation class identity information dataset C>It will be appreciated that the more identity features that occur, the greater the accuracy of the correlation analysis. Therefore, when negative probability analysis is performed, the probability calculation is performed after the identification features are ranked, and reasonable probability prediction is realized while the rationality of probability calculation is fully considered.
The invention also provides a personnel identity association system based on the multi-mode information, which is applied to the personnel identity association method based on the multi-mode information, and comprises the following steps: the multi-mode information acquisition unit is used for acquiring multi-mode information of different personnel objects in different carrier objects, and performing type division to form multi-mode identity type data; the matching analysis unit is used for acquiring the multi-mode identity type data of the multi-mode information acquisition unit, carrying out matching analysis and forming identity matching result data; the identity association analysis unit is used for carrying out association matching by combining the identity matching result data of the matching analysis unit and the multi-mode identity type data of the multi-mode information acquisition unit to complete identity association.
The system can collect enough multi-mode data for identity association through the multi-mode information acquisition unit, expands the data types required by identity association analysis, and further improves the accuracy of identity association. A reasonable identity association analysis hardware system is established by utilizing the matching analysis unit and the identity association analysis unit, so that the identity association analysis is accurately and efficiently carried out, and a necessary hardware foundation is provided for the realization of the identity association.
In summary, the system and the method for associating personnel identities based on multi-mode information provided by the embodiment of the invention have the beneficial effects that:
according to the method, the reasonable type division based on the matching correlation analysis is carried out by acquiring the personnel identity multi-mode information under different carrier objects, so that the multi-mode personnel identity information is effectively utilized, and the analysis efficiency and accuracy of personnel identity correlation are improved. Meanwhile, based on multi-mode identity type data, a reasonable identity association method model is established, and association of personnel identities on different carrier objects is accurately achieved in a more reasonable and orderly mode. The efficiency and accuracy of identity association are further improved.
The system can collect enough multi-mode data for identity association through the multi-mode information acquisition unit, expands the data types required by identity association analysis, and further improves the accuracy of identity association. A reasonable identity association analysis hardware system is established by utilizing the matching analysis unit and the identity association analysis unit, so that the identity association analysis is accurately and efficiently carried out, and a necessary hardware foundation is provided for the realization of the identity association.
In the present invention, "at least one" means one or more, and "a plurality" means two or more. "at least one of" or the like means any combination of these items, including any combination of single item(s) or plural items(s). For example, at least one (one) of a, b, or c may represent: a, b, c, a-b, a-c, b-c, or a-b-c, wherein a, b, c may be single or plural.
It should be understood that, in various embodiments of the present invention, the sequence numbers of the foregoing processes do not mean the order of execution, and the order of execution of the processes should be determined by the functions and internal logic thereof, and should not constitute any limitation on the implementation process of the embodiments of the present invention.
Those of ordinary skill in the art will appreciate that the elements and method steps of the examples described in connection with the embodiments disclosed herein can be implemented as electronic hardware, or as a combination of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, and are not repeated herein.
In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the units is merely a logical function division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a read-only memory (ROM), a random access memory (random access memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The foregoing examples illustrate only a few embodiments of the invention, which are described in detail and are not to be construed as limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention. Accordingly, the scope of protection of the present invention is to be determined by the appended claims.

Claims (10)

1. A method for associating identities of persons based on multimodal information, comprising:
acquiring multi-modal information of personnel identities under different carrier objects, and performing type division analysis on the multi-modal information to form multi-modal identity type data;
carrying out matching analysis on different personnel identity information under different carrier objects according to the identity activity information of each personnel object on the carrier objects and combining the multi-mode identity type data to form identity matching result data;
and establishing an identity association analysis model according to the identity matching result data and the multi-mode identity type data, and carrying out association matching on the personnel identity information in different carrier objects according to the identity association analysis model to form association result data.
2. The method for associating personnel identity based on multi-modal information according to claim 1, wherein the steps of obtaining multi-modal information of personnel identity under different carrier objects, performing type division analysis of the multi-modal information, and forming multi-modal identity type data include:
acquiring multi-mode information of each personnel identity in different carrier objects, and dividing the multi-mode information into the following identity judgment types:
the information which is judged whether to be inverted or not in the accuracy judgment of the multi-mode information is divided into a judging type identity information data set A;
the information which is selectively judged in the accuracy judgment of the multi-mode information is divided into a selection type identity information data set B;
and analyzing other information in the multi-mode information except the discrimination type identity information data set A and the selection type identity information data set B based on identity recognition to form orientation type identity information data C.
3. The method according to claim 2, wherein in the discrimination class identity information data set a, a single discrimination class identity information set for directly determining the identity of a person is defined as a definitive discrimination class identity information data subset Wherein:
=[/>，/>，…，/>]m is the decisive discrimination class identity information data subset +.>The number of the discrimination class identity information;
collecting other discrimination class identity information except the decisive discrimination class identity information data in the discrimination class identity information data into an auxiliary discrimination class identity information data subsetWherein:
=[/>，/>，…，/>]n is the auxiliary discrimination class identity information data subset +.>The number of the discrimination class identity information.
4. A person identity associating method based on multi-modal information according to claim 3, wherein the step of performing identity-based analysis on other information in the multi-modal information than the discrimination class identity information data set a and the selection class identity information data set B to form orientation class identity information data C includes:
in the orientation identity information data set C, orientation identity information of different modality types is separated, and the orientation identity information of each modality type is extracted to form an orientation identity information data subset containing a plurality of identity features under different modality types, and the orientation identity information data subset is ordered according to the order of the number of the identity features in the different orientation identity information data subsets from more to less, so as to form an order set of the orientation identity information data subset, namely:
C=[，/>，…，/>]Wherein->=[/>，/>，…，/>]U is the sequence number of the orientation type identity information data subset, and v is the number of the identity features in the orientation type identity information data subset with the sequence number of u.
5. The method for associating person identities based on multi-modal information according to claim 4, wherein said performing matching analysis on different person identity information under different carrier objects according to the identity activity information of each person object on the carrier object and in combination with the multi-modal identity type data to form identity matching result data includes:
acquiring the identity activity information of the personnel object on the carrier object, and extracting the identity information data subset of the decisive discrimination classThe words and sentences related to the discrimination identity information in the system are subjected to decisive judgment by carrying out semantic analysis on the words and sentences:
determining and marking the discrimination splitting information for judging whether the inverse exists according to the word and sentence semantics;
non-decisive marking is carried out on the identification information which can not judge whether the inverse is not judged according to the meaning of the words and sentences;
acquiring identity activity information of the personnel object on the carrier object, and extracting a class identity information data subset for auxiliary discrimination The auxiliary judgment is carried out by carrying out semantic analysis on the words and sentences related to the identification information in the method:
carrying out frequency statistics on two opposite results for judging identity information according to the judgment result exhibited by the meaning of the word and sentence, and marking, namely:，/>a negative probability representing a result exhibited by the corresponding discrimination identity information;
acquiring the identity activity information of the personnel object on the carrier object, extracting words and sentences related to the selected identity information in the selected identity information data set B, and carrying out selective judgment by carrying out semantic analysis on the words and sentences:
determining a selection result of each selection identity information according to the semantics of the words and sentences, carrying out probability statistics on effective decision times on the semantics of the words and sentences which decide different selection results, and marking, namely:
，/>and a negative probability representing the result exhibited by the corresponding selection identity information, i being the number of the selection identity information in the selection identity information data set B.
6. The method for associating personal identity based on multi-modal information according to claim 5, wherein the steps of establishing an identity association analysis model according to the identity matching result data and the multi-modal identity type data, and associating and matching the personal identity information in different carrier objects according to the identity association analysis model to form association result data include:
Determining two carrier objects to be associated, and setting the two carrier objects as a first associated carrier object and a second associated carrier object respectively;
acquiring the identity matching result data of the first associated carrier object and the identity matching result data of the second associated carrier object;
and taking the personnel object as a unit, randomly selecting the identity matching result data of one personnel object from the first association carrier object, and acquiring the identity matching result data of the personnel object in the second association carrier object one by one to perform the following association analysis:
subset two said decisive discrimination class identity information dataThe identification information in (a) carries out association analysis:
if the identification information with the same decisive mark exists and the identification information identification results are the same, the two personnel objects are directly subjected to identity association;
otherwise, carrying out discrimination of class identity information data subset based on the assistanceAnd (3) probability association analysis of the selected class identity information data set B and the oriented class identity information data set C, and forming association result data.
7. The multimodal information based personnel identity correlation method of claim 6 wherein the auxiliary discrimination class identity information based data subset -probability association analysis of said selection class identity information dataset B and said orientation class identity information dataset C and forming association result data comprising:
acquiring the auxiliary discrimination class identity information data subsets of personnel objects under two carrier objectsPerforming overall negative probability analysis to obtain judging class negative association probability +.>；
Acquiring the selection class identity information data sets B of the personnel objects under the two carrier objects, and performing coincidence-based negative probability analysis to obtain selection class negative association probability；
Acquiring the orientation class identity information data sets C of the personnel objects under the two carrier objects, and performing negative probability analysis based on the coincidence degree of orientation class identity characteristics to acquire orientation class negative association probability；
Combining the discriminant class negative association probabilitiesSaid selection class negative association probability +.>Said orientation class negative association probability +.>Obtaining a negative association probability of two person objects according to the following formula->：
；
Setting a negative judgment threshold alpha, and acquiring all negative association probabilities of the person objects determined under the first association carrier object and the person objects corresponding to the second association carrier object one by one If there is said negative association probability with the maximum negative association probability value +.>If the negative association probability value of the person object selected by the first association carrier object in the second association carrier object is less than or equal to alpha, determining the person object with the largest negative association probability value as the same association person, and carrying out identity association;
otherwise, judging that the personnel object selected from the first association carrier object has no identity association in the second association carrier object, and continuing to repeatedly perform association judgment for establishing an identity association analysis model from the personnel object selected from the first association carrier object.
8. The multimodal information based personnel identity correlation method of claim 7 wherein the acquiring of the auxiliary discrimination class identity information data subset of personnel objects under two of the carrier objectsPerforming overall negative probability analysis to obtain judging class negative association probability +.>Comprising:
acquiring the discriminant class negative association probability of the kth personnel object under the selected second association carrier object according to the following：
Wherein, the method comprises the steps of, wherein,
，
here, k is the probability of making a negative associationAnd calculating the personnel object under the determined second associated carrier object.
9. The method for associating personnel identity based on multimodal information according to claim 8, wherein the acquiring the data sets B of the selected class identity information of the personnel objects under the two carrier objects performs a negative probability analysis based on the coincidence degree to obtain a selected class negative association probabilityComprising the following steps:
acquiring the total number of the selection identity information of the selection class identity information data set B of the personnel object under the first association carrier object and the kth personnel object selected under the second association carrier object, determining the coincidence rate x and the coincidence quantity r of the selection identity information, and determining the selection class negative association probability according to the following formula：
；
Acquiring the orientation class identity information data sets C of the personnel objects under the two carrier objects, and performing negative probability analysis based on the coincidence degree of orientation class identity characteristics to acquire orientation class negative association probabilityComprising:
acquiring the orientation class identity information data set C of a person object under the first associated carrier object and a kth person object selected under the second associated carrier object, determining the overlapped orientation class identity data subset, sequencing the overlapped orientation class identity data subset according to the sequence of the number of identity features in the orientation class identity information data subset from more to less to form an overlapped orientation class identity information data set Wherein:
=[/>，/>，…，/>]z represents the coincident sequential label;
probability statistics is carried out on the identity characteristics overlapped in each overlapped orientation type identity information data subset to form the overlapped probability of each overlapped orientation type identity information data subset；
Determining the orientation class negative association probability according to：
。
10. A person identity association system based on multi-modal information, which is applied to the person identity association method based on multi-modal information as set forth in any one of claims 1 to 9, and includes:
the multi-mode information acquisition unit is used for acquiring multi-mode information of different personnel objects in different carrier objects, and performing type division to form multi-mode identity type data;
the matching analysis unit is used for acquiring the multi-mode identity type data of the multi-mode information acquisition unit, carrying out matching analysis and forming identity matching result data;
and the identity association analysis unit is used for carrying out association matching by combining the identity matching result data of the matching analysis unit and the multi-mode identity type data of the multi-mode information acquisition unit to complete identity association.
CN202311329995.1A 2023-10-16 2023-10-16 Personnel identity association method and system based on multi-mode information Pending CN117076957A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311329995.1A CN117076957A (en) 2023-10-16 2023-10-16 Personnel identity association method and system based on multi-mode information

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311329995.1A CN117076957A (en) 2023-10-16 2023-10-16 Personnel identity association method and system based on multi-mode information

Publications (1)

Publication Number Publication Date
CN117076957A true CN117076957A (en) 2023-11-17

Family

ID=88704598

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311329995.1A Pending CN117076957A (en) 2023-10-16 2023-10-16 Personnel identity association method and system based on multi-mode information

Country Status (1)

Country Link
CN (1) CN117076957A (en)

Citations (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100228692A1 (en) * 2009-03-03 2010-09-09 Honeywell International Inc. System and method for multi-modal biometrics
EP2523149A2 (en) * 2011-05-11 2012-11-14 Tata Consultancy Services Ltd. A method and system for association and decision fusion of multimodal inputs
CN104699784A (en) * 2015-03-13 2015-06-10 苏州思必驰信息科技有限公司 Data searching method and device based on interactive input
US20170024899A1 (en) * 2014-06-19 2017-01-26 Bae Systems Information & Electronic Systems Integration Inc. Multi-source multi-modal activity recognition in aerial video surveillance
CN108646914A (en) * 2018-04-27 2018-10-12 安徽斛兵信息科技有限公司 A kind of multi-modal affection data collection method and device
CN112069484A (en) * 2020-11-10 2020-12-11 中国科学院自动化研究所 Multi-mode interactive information acquisition method and system
CN113222775A (en) * 2021-05-28 2021-08-06 北京理工大学 User identity correlation method integrating multi-mode information and weight tensor
WO2021159756A1 (en) * 2020-09-04 2021-08-19 平安科技（深圳）有限公司 Method for response obligation detection based on multiple modes, and system and apparatus
CN113297397A (en) * 2021-05-12 2021-08-24 山东大学 Information matching method and system based on hierarchical multi-mode information fusion
CN114270337A (en) * 2019-06-17 2022-04-01 得麦股份有限公司 System and method for personalized and multi-modal context-aware human-machine dialog
CN115131881A (en) * 2022-06-09 2022-09-30 厦门瑞为信息技术有限公司 Millimeter wave security check method and device based on multi-mode non-inductive identity authentication
CN116049743A (en) * 2022-12-14 2023-05-02 深圳市仰和技术有限公司 Cognitive recognition method based on multi-modal data, computer equipment and storage medium
CN116797895A (en) * 2023-06-16 2023-09-22 浙江大学 Judicial scene-oriented multi-mode fusion identity authentication method, medium and device

Patent Citations (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100228692A1 (en) * 2009-03-03 2010-09-09 Honeywell International Inc. System and method for multi-modal biometrics
EP2523149A2 (en) * 2011-05-11 2012-11-14 Tata Consultancy Services Ltd. A method and system for association and decision fusion of multimodal inputs
US20170024899A1 (en) * 2014-06-19 2017-01-26 Bae Systems Information & Electronic Systems Integration Inc. Multi-source multi-modal activity recognition in aerial video surveillance
CN104699784A (en) * 2015-03-13 2015-06-10 苏州思必驰信息科技有限公司 Data searching method and device based on interactive input
CN108646914A (en) * 2018-04-27 2018-10-12 安徽斛兵信息科技有限公司 A kind of multi-modal affection data collection method and device
CN114270337A (en) * 2019-06-17 2022-04-01 得麦股份有限公司 System and method for personalized and multi-modal context-aware human-machine dialog
WO2021159756A1 (en) * 2020-09-04 2021-08-19 平安科技（深圳）有限公司 Method for response obligation detection based on multiple modes, and system and apparatus
CN112069484A (en) * 2020-11-10 2020-12-11 中国科学院自动化研究所 Multi-mode interactive information acquisition method and system
CN113297397A (en) * 2021-05-12 2021-08-24 山东大学 Information matching method and system based on hierarchical multi-mode information fusion
CN113222775A (en) * 2021-05-28 2021-08-06 北京理工大学 User identity correlation method integrating multi-mode information and weight tensor
CN115131881A (en) * 2022-06-09 2022-09-30 厦门瑞为信息技术有限公司 Millimeter wave security check method and device based on multi-mode non-inductive identity authentication
CN116049743A (en) * 2022-12-14 2023-05-02 深圳市仰和技术有限公司 Cognitive recognition method based on multi-modal data, computer equipment and storage medium
CN116797895A (en) * 2023-06-16 2023-09-22 浙江大学 Judicial scene-oriented multi-mode fusion identity authentication method, medium and device

Similar Documents

Publication Publication Date Title
CN109189991B (en) Duplicate video identification method, device, terminal and computer readable storage medium
WO2019218514A1 (en) Method for extracting webpage target information, device, and storage medium
CN109299271B (en) Training sample generation method, text data method, public opinion event classification method and related equipment
CN106469181B (en) User behavior pattern analysis method and device
Petkos et al. A soft frequent pattern mining approach for textual topic detection
CN109284353B (en) Medical case retrieval method, device, computer equipment and storage medium
CN111898642A (en) Key point detection method and device, electronic equipment and storage medium
Fisichella et al. Unsupervised public health event detection for epidemic intelligence
CN110457481A (en) A kind of method, apparatus, equipment and the storage medium of disaggregated model training
CN110929525B (en) Network loan risk behavior analysis and detection method, device, equipment and storage medium
Silvestre et al. Feature selection for clustering categorical data with an embedded modelling approach
CN110737821B (en) Similar event query method, device, storage medium and terminal equipment
Shamir et al. Leveraging pattern recognition consistency estimation for crowdsourcing data analysis
CN112035620A (en) Question-answer management method, device, equipment and storage medium of medical query system
CN103488707B (en) A kind of method that candidate categories are searched for based on Greedy strategy and heuritic approach
CN112966072A (en) Case prediction method and device, electronic device and storage medium
CN106910135A (en) User recommends method and device
CN112270191A (en) Method and device for extracting work order text theme
CN113254651B (en) Method and device for analyzing referee document, computer equipment and storage medium
CN113111159A (en) Question and answer record generation method and device, electronic equipment and storage medium
CN113032524A (en) Trademark infringement identification method, terminal device and storage medium
CN109462635B (en) Information pushing method, computer readable storage medium and server
CN109144999B (en) Data positioning method, device, storage medium and program product
CN117076957A (en) Personnel identity association method and system based on multi-mode information
CN113724779B (en) SNAREs protein identification method, system, storage medium and equipment based on machine learning technology

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