CN111368213A - Method and system for detecting overlapped community structure of civil aviation passenger relationship network - Google Patents

Method and system for detecting overlapped community structure of civil aviation passenger relationship network Download PDF

Info

Publication number
CN111368213A
CN111368213A CN202010142875.0A CN202010142875A CN111368213A CN 111368213 A CN111368213 A CN 111368213A CN 202010142875 A CN202010142875 A CN 202010142875A CN 111368213 A CN111368213 A CN 111368213A
Authority
CN
China
Prior art keywords
passenger
nodes
civil aviation
node
relationship network
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
CN202010142875.0A
Other languages
Chinese (zh)
Inventor
杜航原
白亮
王文剑
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.)
Shanxi University
Original Assignee
Shanxi University
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 Shanxi University filed Critical Shanxi University
Priority to CN202010142875.0A priority Critical patent/CN111368213A/en
Publication of CN111368213A publication Critical patent/CN111368213A/en
Pending legal-status Critical Current

Links

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
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9536—Search customisation based on social or collaborative filtering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/01—Social networking
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/14—Travel agencies

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Databases & Information Systems (AREA)
  • General Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Strategic Management (AREA)
  • Primary Health Care (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention provides a method and a system for detecting an overlapped community structure of a civil aviation passenger relationship network, which can accurately and quickly obtain a detection result of the overlapped community structure in the civil aviation passenger relationship network. The method comprises the following steps: constructing a civil aviation passenger relationship network through the association relationship among passengers; determining an aggregation factor, a mutual exclusion factor and a representation degree of each passenger node in a civil aviation passenger relationship network, wherein the representation degree is determined by the aggregation factor and the mutual exclusion factor; selecting K passenger nodes with the largest representativeness as core passenger nodes of each community structure, wherein other passenger nodes are non-core passenger nodes, and initializing the community membership degree of the core passenger nodes; and determining the membership degree of each non-core passenger node about each community structure to obtain an overlapped community structure detection result of the civil aviation passenger relationship network. The invention relates to the technical field of civil aviation data analysis.

Description

Method and system for detecting overlapped community structure of civil aviation passenger relationship network
Technical Field
The invention relates to the technical field of civil aviation data analysis, in particular to a method and a system for detecting an overlapped community structure of a civil aviation passenger relationship network.
Background
With the rapid development of the modern construction of the society, the pursuit of the people for good life is increasing day by day, and the requirements for travel service and experience are diversified. More and more passengers select a convenient and rapid travel mode of boarding an airplane, so that a great deal of passenger travel data stored in various information systems of various large airlines show explosive growth. How to fully utilize the data resources to deeply plough the passenger value, expand the passenger resources, excavate the potential demands of the passengers, and provide high-quality personalized services for the passengers so as to stimulate the income improvement is a huge opportunity and challenge faced by the civil aviation industry nowadays. In recent years, complex network research gradually shows important academic and application values in civil aviation passenger data analysis, passenger value analysis, potential high-value passenger mining, new passenger growth prediction, travel preference understanding of different passengers, family discovery, even intelligent passenger data management and the like are carried out on the basis of a civil aviation passenger relationship network.
The community structure is an important attribute of a complex network, and in the complex network composed of civil aviation passenger data, the community structure is expressed in the form of a passenger family group, a passenger colleague group, a passenger friend group, a tourist group organized by a travel agency, and the like. Accurate discovery and identification of the community structures are beneficial to an airline company to discover implicit rules and evolution trends in the relationship of passengers, so that the service quality is improved and the product recommendation is optimized for the passengers. Compared with a general complex network, the community structure in the civil aviation passenger relationship network also has an overlapping characteristic, namely, passenger nodes in the network can belong to more than one community at the same time. For example, a traveler may belong to multiple groups of friends or multiple groups of guests at the same time. The overlapping characteristic further increases the community discovery difficulty of the civil aviation passenger relationship network, so that designing an efficient overlapping community discovery method for the civil aviation passenger relationship network is a significant and challenging technical problem.
Disclosure of Invention
The invention aims to provide a method and a system for detecting an overlapped community structure of a civil aviation passenger relationship network, so as to solve the problem that the overlapped community structure in the civil aviation passenger relationship network is difficult to accurately obtain in the prior art.
In order to solve the above technical problem, an embodiment of the present invention provides a method for detecting an overlapping community structure of a civil aviation passenger relationship network, including:
constructing a civil aviation passenger relationship network through the association relationship among passengers;
determining an aggregation factor, a mutual exclusion factor and a representation degree of each passenger node in a civil aviation passenger relationship network, wherein the representation degree is determined by the aggregation factor and the mutual exclusion factor;
selecting K passenger nodes with the largest representativeness as core passenger nodes of each community structure, wherein other passenger nodes are non-core passenger nodes, and initializing the community membership degree of the core passenger nodes; k is the number of community structures contained in the civil aviation passenger relationship network;
and determining the membership degree of each non-core passenger node about each community structure to obtain an overlapped community structure detection result of the civil aviation passenger relationship network.
Further, the constructing of the civil aviation passenger relationship network through the association relationship among the passengers comprises:
acquiring passenger order data, wherein each passenger in the order data corresponds to a passenger node in a civil aviation passenger relationship network;
judging whether two passengers appear in the same order and buy the air tickets of the same flight, if so, a connecting edge exists between the passenger nodes corresponding to the two passengers;
the civil aviation passenger relation network G (V, E) is constructed by the passenger nodes and the connecting edges among the passenger nodes, wherein V is (V ═ E1,v2,…,vm) Representing a set of all passenger nodes in a civil aviation passenger relationship network G (V, E), ViThe number of ith passenger nodes in the civil aviation passenger relationship network G (V, E) is represented, i is more than or equal to 1 and less than or equal to m, m is the number of passenger nodes in the civil aviation passenger relationship network G (V, E), and E is equal to (E)1,e2,…,en) Representing a set of all connected edges in a civil aviation passenger relationship network G (V, E), ElAnd l is more than or equal to 1 and less than or equal to n, and n is the number of all connected edges in the civil aviation passenger relationship network G (V, E).
Further, the determining the aggregation factor of each passenger node in the civil aviation passenger relationship network includes:
taking the number of connected edges connected with a first passenger node as the importance of the first passenger node, wherein the first passenger node is any passenger node in a civil aviation passenger relationship network;
determining similarity between each passenger node in civil aviation passenger relationship network and passenger nodes with direct edge-connecting relationship, wherein the passenger nodes viPassenger node v with direct edge relationjThe similarity between them is expressed as:
s(vi,vj)＝|NBi∩NBj|
wherein, s (v)i,vj) Representing passenger nodes viAnd passenger node vjSimilarity between them, NBiRepresentation and passenger node viPassenger node set, NG, with direct edge-to-edge relationshipjRepresentation and passenger node vjPassenger node set with direct edge-to-edge relationship, | NBi∩NBjI denotes NBiAnd NGjNumber of public passenger nodes in;
and determining an aggregation factor of each passenger node according to the importance of each passenger node and the similarity between each passenger node and other passenger nodes, wherein the aggregation factor is used for describing the consistence of the internal association of the community structure.
Further, passenger node v in civil aviation passenger relationship networkiThe polymerization factor of (a) is represented as:
Figure BDA0002399697950000035
wherein L isiRepresenting passenger nodes viPolymerization factor of diRepresenting passenger nodes viThe importance of (a) to (b),
Figure BDA0002399697950000031
representing passenger nodes viAnd NBiMaximum facies between passenger nodes in (1)Similarity.
Further, the mutual exclusion factor is used for describing sparsity of association outside the community structure;
passenger node v in civil aviation passenger relation networkiThe mutual exclusion factor of (2) is expressed as:
Figure BDA0002399697950000032
wherein, PiFor passenger nodes viThe mutual exclusion factor(s) of (c),
Figure BDA0002399697950000033
represents NBiThe middle importance degree is higher than the passenger node viPassenger node vjAnd passenger node viThe maximum similarity between them.
Further, passenger node v in civil aviation passenger relationship networkiThe degree of representation of (d) is expressed as:
Figure BDA0002399697950000034
wherein R isiFor passenger nodes viDegree of representation of (c).
Further, the representation degree is used for reflecting the capacity of each passenger node to become the core of the community structure to which the passenger node belongs;
the K passenger nodes with the largest representativeness are selected as core passenger nodes of each community structure, other passenger nodes are non-core passenger nodes, and the community membership degree of the initialized core passenger nodes comprises the following steps:
all passenger nodes in the civil aviation passenger relationship network are sorted in descending order according to the aggregation factors, and the sorted passenger node set is recorded as
Figure BDA0002399697950000041
For any 2 passenger nodes therein
Figure BDA0002399697950000042
And
Figure BDA0002399697950000043
satisfies the following conditions: if m>i>j>0, then
Figure BDA0002399697950000044
And
Figure BDA0002399697950000045
satisfies the polymerization factor of Li<Lj；
From a ranked set of passenger nodes
Figure BDA0002399697950000046
Selecting K passenger nodes with the maximum substitution table as core passenger nodes of each community structure, and recording the node as C ═ C1,c2,…,cK) Wherein c iskThe core passenger nodes of the kth community structure in the civil aviation passenger relationship network are represented, and K is more than or equal to 1 and less than or equal to K, and the serial numbers of the core passenger nodes in the civil aviation passenger relationship network are represented;
and initializing the membership degree of each core passenger node with respect to the community structure to which the core passenger node belongs to a preset value, wherein each core passenger node in the civil aviation passenger relationship network determines a community structure, and the membership degree is used for reflecting the overlapping characteristic of the community structures in the civil aviation passenger relationship network.
Furthermore, each non-core passenger node can simultaneously belong to a plurality of community structures represented by any core passenger node with aggregation factors higher than that of the non-core passenger node;
the membership of each non-core passenger node with respect to the community structure is expressed as:
Figure BDA0002399697950000047
Figure BDA0002399697950000048
wherein m isi,kRepresenting ith passenger node v in civil aviation passenger relationship networkiAffiliation with respect to kth Community StructureDegree of membership; intermediate variable m'i，kIs composed of
Figure BDA0002399697950000049
In the form of a short-hand writing of (1),
Figure BDA00023996979500000410
representing passenger nodes
Figure BDA00023996979500000411
And a polymerization factor higher than
Figure BDA00023996979500000412
A passenger node
Figure BDA00023996979500000413
The degree of similarity between the two images,
Figure BDA00023996979500000414
representing passenger nodes
Figure BDA00023996979500000415
Membership for the kth community structure.
Further, the determining the membership degree of each non-core passenger node with respect to each community structure to obtain the detection result of the overlapping community structure of the civil aviation passenger relationship network includes:
and taking a set formed by core passenger nodes representing all community structures, and an obtained set formed by all non-core passenger nodes and membership degrees of all community structures as an overlapped community structure detection result of the civil aviation passenger relationship network.
The embodiment of the invention also provides a detection system for an overlapped community structure of a civil aviation passenger relationship network, which comprises the following steps:
the building module is used for building a civil aviation passenger relationship network through the incidence relationship among passengers;
the first determination module is used for determining an aggregation factor, a mutual exclusion factor and a representation degree of each passenger node in the civil aviation passenger relationship network, wherein the representation degree is determined by the aggregation factor and the mutual exclusion factor;
the selecting module is used for selecting the K passenger nodes with the largest representativeness as core passenger nodes of each community structure, and other passenger nodes are non-core passenger nodes and initialize the community membership degree of the core passenger nodes; k is the number of community structures contained in the civil aviation passenger relationship network;
and the second determining module is used for determining the membership degree of each non-core passenger node about each community structure to obtain the detection result of the overlapped community structure of the civil aviation passenger relationship network.
The technical scheme of the invention has the following beneficial effects:
in the scheme, a civil aviation passenger relationship network is constructed through the association relationship among passengers; determining an aggregation factor, a mutual exclusion factor and a representation degree of each passenger node in a civil aviation passenger relationship network, wherein the representation degree is determined by the aggregation factor and the mutual exclusion factor; selecting K passenger nodes with the largest representativeness as core passenger nodes of each community structure, wherein other passenger nodes are non-core passenger nodes, and initializing the community membership degree of the core passenger nodes; and determining the membership degree of each non-core passenger node about each community structure to obtain an overlapped community structure detection result of the civil aviation passenger relationship network. Therefore, the essential characteristics of dense internal association and sparse external association of the community structure in the civil aviation passenger relationship network are respectively described through the aggregation factor and the mutual exclusion factor, the affiliation between the passenger node and the community structure is described by utilizing the membership degree, the overlapping characteristic of the community structure in the civil aviation passenger relationship network can be effectively reflected, and the detection result of the overlapping community structure in the civil aviation passenger relationship network can be accurately and quickly obtained.
Drawings
Fig. 1 is a schematic flow chart of a method for detecting an overlapping community structure of a civil aviation passenger relationship network according to an embodiment of the present invention;
fig. 2 is a detailed flowchart of a method for detecting an overlapping community structure of a civil aviation passenger relationship network according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of an overlapping community structure detection system of a civil aviation passenger relationship network according to an embodiment of the present invention.
Detailed Description
In order to make the technical problems, technical solutions and advantages of the present invention more apparent, the following detailed description is given with reference to the accompanying drawings and specific embodiments.
The invention provides a method and a system for detecting an overlapped community structure of a civil aviation passenger relationship network, aiming at the problem that the existing overlapped community structure in the civil aviation passenger relationship network is difficult to accurately obtain.
Example one
As shown in fig. 1 and fig. 2, the method for detecting an overlapping community structure of a civil aviation passenger relationship network according to the embodiment of the present invention includes:
s101, constructing a civil aviation passenger relationship network through the association relationship among passengers;
s102, determining an aggregation factor, a mutual exclusion factor and a representation degree of each passenger node in a civil aviation passenger relationship network, wherein the representation degree is determined by the aggregation factor and the mutual exclusion factor;
s103, selecting K passenger nodes with the largest representativeness as core passenger nodes of each community structure, wherein other passenger nodes are non-core passenger nodes, and initializing the community membership degree of the core passenger nodes; k is the number of community structures contained in the civil aviation passenger relationship network;
and S104, determining the membership degree of each non-core passenger node with respect to each community structure to obtain the detection result of the overlapped community structure of the civil aviation passenger relationship network.
The method for detecting the overlapped community structure of the civil aviation passenger relationship network comprises the steps of constructing the civil aviation passenger relationship network through the incidence relationship among passengers; determining an aggregation factor, a mutual exclusion factor and a representation degree of each passenger node in a civil aviation passenger relationship network, wherein the representation degree is determined by the aggregation factor and the mutual exclusion factor; selecting K passenger nodes with the largest representativeness as core passenger nodes of each community structure, wherein other passenger nodes are non-core passenger nodes, and initializing the community membership degree of the core passenger nodes; and determining the membership degree of each non-core passenger node about each community structure to obtain an overlapped community structure detection result of the civil aviation passenger relationship network. Therefore, the essential characteristics of dense internal association and sparse external association of the community structure in the civil aviation passenger relationship network are respectively described through the aggregation factor and the mutual exclusion factor, the affiliation between the passenger node and the community structure is described by utilizing the membership degree, the overlapping characteristic of the community structure in the civil aviation passenger relationship network can be effectively reflected, and the detection result of the overlapping community structure in the civil aviation passenger relationship network can be accurately and quickly obtained.
In a specific implementation manner of the foregoing method for detecting an overlapping community structure of a civil aviation passenger relationship network, the constructing a civil aviation passenger relationship network through an association relationship between passengers further includes:
acquiring passenger order data, wherein each passenger in the order data corresponds to a passenger node in a civil aviation passenger relationship network;
judging whether two passengers appear in the same order and buy the air tickets of the same flight, if so, a connecting edge exists between the passenger nodes corresponding to the two passengers;
the civil aviation passenger relation network G (V, E) is constructed by the passenger nodes and the connecting edges among the passenger nodes, wherein V is (V ═ E1,v2,…,vm) Representing a set of all passenger nodes in a civil aviation passenger relationship network G (V, E), ViThe number of ith passenger nodes in the civil aviation passenger relationship network G (V, E) is represented, i is more than or equal to 1 and less than or equal to m, m is the number of passenger nodes in the civil aviation passenger relationship network G (V, E), and E is equal to (E)1,e2,…,en) Representing a set of all connected edges in a civil aviation passenger relationship network G (V, E), ElAnd l is more than or equal to 1 and less than or equal to n, and n is the number of all connected edges in the civil aviation passenger relationship network G (V, E).
In this embodiment, for example, passenger order data in the fourth quarter of 2018 may be obtained through an interface provided by a passenger order database of an airline company, where the obtained passenger order data is composed of 64487 passenger orders and includes 173482 travel records of 87638 passengers, and the order data includes: order number, flight number taken by the passenger, passenger ID number, passenger name, passenger gender, ticket number, etc.
In this embodiment, if two passengers appear in the same order and purchase tickets for the same flight, there is a connecting edge between the passenger nodes corresponding to the two passengers, and the civil aviation passenger relationship network G (V, E) is constructed by 87638 passenger nodes and the connecting edges between them, where m is 87638 and n is 154396.
In this embodiment, each passenger node corresponds to one passenger in the order data, and includes 4 attributes of a passenger ID number, a passenger name, a passenger gender, and a ticket-buying frequency, where the ticket-buying frequency is an accumulated ticket-buying frequency of the passenger in the passenger order data.
In a specific implementation of the foregoing method for detecting an overlapping community structure of a civil aviation passenger relationship network, further, the determining an aggregation factor of each passenger node in the civil aviation passenger relationship network includes:
taking the number of connected edges connected with a first passenger node as the importance of the first passenger node, wherein the first passenger node is any passenger node in a civil aviation passenger relationship network;
determining similarity between each passenger node in civil aviation passenger relationship network and passenger nodes with direct edge-connecting relationship, wherein the passenger nodes viPassenger node v with direct edge relationjThe similarity between them is expressed as:
s(vi,vj)＝|NBi∩NBj| (1)
in the formula (1), s (v)i,vj) Representing passenger nodes viAnd passenger node vjThe similarity is used for describing the correlation between two passenger nodes in the civil aviation passenger relationship network; NBiRepresentation and passenger node viA passenger node set with direct edge-connecting relation; NGjRepresentation and passenger node vjWith a direct connectionA set of passenger nodes for edge relationships; | NBi∩NBjI denotes NBiAnd NGjNumber of public passenger nodes in;
and determining an aggregation factor of each passenger node according to the importance of each passenger node and the similarity between each passenger node and other passenger nodes, wherein the aggregation factor is used for describing the maximum cohesion of a certain passenger node to other passenger nodes in a community structure to which the certain passenger node is possibly attached, namely the consistence of the internal association of the community structure.
In the specific implementation manner of the overlapped community structure detection method of the civil aviation passenger relationship network, further, the passenger node v in the civil aviation passenger relationship networkiThe polymerization factor of (a) is represented as:
Figure BDA0002399697950000081
in the formula (2), LiRepresenting passenger nodes viPolymerization factor of diRepresenting passenger nodes viThe importance of (a) to (b),
Figure BDA0002399697950000082
representing passenger nodes viAnd NBiThe maximum similarity between passenger nodes in (1).
In a specific implementation of the foregoing method for detecting an overlapping community structure of a civil aviation passenger relationship network, further, the mutual exclusion factor is used to reflect a maximum correlation between a certain passenger node and a passenger node outside a community structure to which the passenger node may belong, that is, sparsity of external correlation of the community structure; passenger node v in civil aviation passenger relation networkiThe mutual exclusion factor of (2) is expressed as:
Figure BDA0002399697950000083
in the formula (3), PiFor passenger nodes viThe mutual exclusion factor(s) of (c),
Figure BDA0002399697950000084
represents NBiThe middle importance degree is higher than the passenger node viPassenger node vjAnd passenger node viThe maximum similarity between them.
In the specific implementation manner of the overlapped community structure detection method of the civil aviation passenger relationship network, further, the passenger node v in the civil aviation passenger relationship networkiThe degree of representation of (d) is expressed as:
Figure BDA0002399697950000085
in the formula (4), RiFor passenger nodes viDegree of representation of (c).
In a specific implementation of the method for detecting an overlapping community structure of a civil aviation passenger relationship network, the representation is further used for reflecting the ability of each passenger node to become the core of the community structure to which the passenger node belongs;
the K passenger nodes with the largest representativeness are selected as core passenger nodes of each community structure, other passenger nodes are non-core passenger nodes, and the community membership degree of the initialized core passenger nodes comprises the following steps:
all passenger nodes in the civil aviation passenger relationship network are sorted in descending order according to the aggregation factors, and the sorted passenger node set is recorded as
Figure BDA0002399697950000091
For any 2 passenger nodes therein
Figure BDA0002399697950000092
And
Figure BDA0002399697950000093
satisfies the following conditions: if m>i>j>0, then
Figure BDA0002399697950000094
And
Figure BDA0002399697950000095
satisfies the polymerization factor of Li<Lj；
From a ranked set of passenger nodes
Figure BDA0002399697950000096
Selecting K passenger nodes with the maximum substitution table as core passenger nodes of each community structure, and recording the node as C ═ C1,c2,…,cK) Wherein c iskThe core passenger node represents the kth community structure in the civil aviation passenger relationship network, K is more than or equal to 1 and less than or equal to K, the serial number of the core passenger node in the civil aviation passenger relationship network is represented, and for example, K is 500, which is the number of community structures contained in the civil aviation passenger relationship network;
and initializing the membership degree of each core passenger node relative to the community structure to which the core passenger node belongs to a preset value (for example, 1), wherein each core passenger node in the civil aviation passenger relationship network determines a community structure, and the membership degree represents the membership relationship between the passenger node and the community structure and is used for reflecting the overlapping characteristic of the community structures in the civil aviation passenger relationship network.
In the specific implementation of the method for detecting the overlapping community structure of the civil aviation passenger relationship network, further, each non-core passenger node may be simultaneously affiliated to a plurality of community structures represented by any core passenger node with aggregation factors higher than that of the non-core passenger node;
as shown in fig. 2, for the passenger nodes sorted in descending order according to the aggregation factor, according to the aggregation factor and the similarity of the passenger nodes, the membership degree of each non-core passenger node with respect to each community structure is calculated by the recursive calculation process shown in equations (5) and (6), and is expressed as:
Figure BDA0002399697950000097
Figure BDA0002399697950000098
in the formulae (5) and (6), mi,kRepresenting ith passenger node v in civil aviation passenger relationship networkiMembership for a kth community structure; intermediate variable m'i，kIs composed of
Figure BDA0002399697950000101
In the form of a short-hand writing of (1),
Figure BDA0002399697950000102
representing passenger nodes
Figure BDA0002399697950000103
And a polymerization factor higher than
Figure BDA0002399697950000104
A passenger node
Figure BDA0002399697950000105
The degree of similarity between the two images,
Figure BDA0002399697950000106
representing passenger nodes
Figure BDA0002399697950000107
Membership for the kth community structure.
In this embodiment, the set formed by the core passenger nodes representing the community structures obtained in S103, the set formed by the non-core passenger nodes and the membership degrees thereof with respect to the community structures obtained in S104 are used as the detection result of the overlapping community structure of the civil aviation passenger relationship network, and the result is output to the data analyst of the civil aviation enterprise for analyzing the passenger value, predicting the passenger growth, analyzing the passenger trip preference, and the like, so as to help the civil aviation enterprise improve the service quality and optimize the product recommendation.
Next, to verify the validity of the overlapped community structure detection method of the civil aviation passenger relationship network provided by the present invention, 5 existing overlapped community structure detection methods are selected for comparison: local matching method (LFM), LINK (LINK) -Based Algorithm, Community overlap propagation Algorithm (COPRA), party discovery (Clique Finder) Algorithm, and density peak-Based overlap Community discovery Algorithm (Overlapping Community Detection Algorithm Based on intensity discovery Peaks, OCDDP), where LFM, LINK, COPRA, CFinder, OCDDP, and the overlap Community structure Detection method of civil aviation relational network provided by the present invention are used to evaluate the Community overlap structure Detection result of passenger order data in the fourth quarter of 2018 of a certain aviation enterprise, and the results are shown in table 1:
TABLE 1 evaluation results of detection results of overlapping communities
Figure BDA0002399697950000108
As can be seen from the results in table 1, the method for detecting the overlapping community structure of the civil aviation passenger relationship network provided by the invention can obtain the detection result of the overlapping community structure with higher accuracy than the existing method when detecting the overlapping community structure of the civil aviation passenger relationship network, and has relatively higher execution efficiency.
Example two
The invention also provides a specific implementation mode of the overlapped community structure detection system of the civil aviation passenger relationship network, and the overlapped community structure detection system of the civil aviation passenger relationship network provided by the invention corresponds to the specific implementation mode of the overlapped community structure detection method of the civil aviation passenger relationship network, and the purpose of the invention can be realized by executing the flow steps in the specific implementation mode of the method, so the explanation in the specific implementation mode of the overlapped community structure detection method of the civil aviation passenger relationship network is also suitable for the specific implementation mode of the overlapped community structure detection system of the civil aviation passenger relationship network provided by the invention, and the details are not repeated in the following specific implementation modes of the invention.
As shown in fig. 3, an embodiment of the present invention further provides a system for detecting an overlapping community structure of a civil aviation passenger relationship network, including:
the building module 11 is used for building a civil aviation passenger relationship network through the incidence relationship among passengers;
the first determining module 12 is configured to determine an aggregation factor, a mutual exclusion factor, and a representation degree of each passenger node in the civil aviation passenger relationship network, where the representation degree is determined by the aggregation factor and the mutual exclusion factor;
the selecting module 13 is used for selecting the K passenger nodes with the largest representativeness as core passenger nodes of each community structure, and other passenger nodes are non-core passenger nodes and initialize the community membership degrees of the core passenger nodes; k is the number of community structures contained in the civil aviation passenger relationship network;
and the second determining module 14 is configured to determine a membership degree of each non-core passenger node with respect to each community structure, and output an overlapped community structure detection result of the civil aviation passenger relationship network.
The overlapped community structure detection system of the civil aviation passenger relationship network disclosed by the embodiment of the invention constructs the civil aviation passenger relationship network through the incidence relation among passengers; determining an aggregation factor, a mutual exclusion factor and a representation degree of each passenger node in a civil aviation passenger relationship network, wherein the representation degree is determined by the aggregation factor and the mutual exclusion factor; selecting K passenger nodes with the largest representativeness as core passenger nodes of each community structure, wherein other passenger nodes are non-core passenger nodes, and initializing the community membership degree of the core passenger nodes; and determining the membership degree of each non-core passenger node about each community structure to obtain an overlapped community structure detection result of the civil aviation passenger relationship network. Therefore, the essential characteristics of dense internal association and sparse external association of the community structure in the civil aviation passenger relationship network are respectively described through the aggregation factor and the mutual exclusion factor, the affiliation between the passenger node and the community structure is described by utilizing the membership degree, the overlapping characteristic of the community structure in the civil aviation passenger relationship network can be effectively reflected, and the detection result of the overlapping community structure in the civil aviation passenger relationship network can be accurately and quickly obtained.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the appended claims.

Claims (10)

1. A detection method for overlapped community structure of civil aviation passenger relationship network is characterized by comprising the following steps:
constructing a civil aviation passenger relationship network through the association relationship among passengers;
determining an aggregation factor, a mutual exclusion factor and a representation degree of each passenger node in a civil aviation passenger relationship network, wherein the representation degree is determined by the aggregation factor and the mutual exclusion factor;
selecting K passenger nodes with the largest representativeness as core passenger nodes of each community structure, wherein other passenger nodes are non-core passenger nodes, and initializing the community membership degree of the core passenger nodes; k is the number of community structures contained in the civil aviation passenger relationship network;
and determining the membership degree of each non-core passenger node about each community structure to obtain an overlapped community structure detection result of the civil aviation passenger relationship network.
2. The method for detecting the overlapped community structure of the civil aviation passenger relationship network according to claim 1, wherein the step of constructing the civil aviation passenger relationship network through the association relationship among passengers comprises the following steps:
acquiring passenger order data, wherein each passenger in the order data corresponds to a passenger node in a civil aviation passenger relationship network;
judging whether two passengers appear in the same order and buy the air tickets of the same flight, if so, a connecting edge exists between the passenger nodes corresponding to the two passengers;
the civil aviation passenger relation network G (V, E) is constructed by the passenger nodes and the connecting edges among the passenger nodes, wherein V is (V ═ E1,v2,…,vm) Representing a set of all passenger nodes in a civil aviation passenger relationship network G (V, E), ViThe number of ith passenger nodes in the civil aviation passenger relationship network G (V, E) is represented, i is more than or equal to 1 and less than or equal to m, m is the number of passenger nodes in the civil aviation passenger relationship network G (V, E), and E is equal to (E)1,e2,…,en) Representing all connected edges in civil aviation passenger relationship network G (V, E)Set of constituents, elAnd l is more than or equal to 1 and less than or equal to n, and n is the number of all connected edges in the civil aviation passenger relationship network G (V, E).
3. The method for detecting the overlapped community structure of the civil aviation passenger relationship network according to claim 1, wherein the determining the aggregation factor of each passenger node in the civil aviation passenger relationship network comprises:
taking the number of connected edges connected with a first passenger node as the importance of the first passenger node, wherein the first passenger node is any passenger node in a civil aviation passenger relationship network;
determining similarity between each passenger node in civil aviation passenger relationship network and passenger nodes with direct edge-connecting relationship, wherein the passenger nodes viPassenger node v with direct edge relationjThe similarity between them is expressed as:
s(vi,vj)＝|NBi∩NBj|
wherein, s (v)i,vj) Representing passenger nodes viAnd passenger node vjSimilarity between them, NBiRepresentation and passenger node viPassenger node set, NG, with direct edge-to-edge relationshipjRepresentation and passenger node vjPassenger node set with direct edge-to-edge relationship, | NBi∩NBjI denotes NBiAnd NGjNumber of public passenger nodes in;
and determining an aggregation factor of each passenger node according to the importance of each passenger node and the similarity between each passenger node and other passenger nodes, wherein the aggregation factor is used for describing the consistence of the internal association of the community structure.
4. The method of claim 3, wherein passenger nodes v in the civil aviation passenger relationship networkiThe polymerization factor of (a) is represented as:
Figure FDA0002399697940000021
wherein L isiRepresenting passenger nodes viPolymerization factor of diRepresenting passenger nodes viThe importance of (a) to (b),
Figure FDA0002399697940000022
representing passenger nodes viAnd NBiThe maximum similarity between passenger nodes in (1).
5. The method for detecting the overlapped community structure of the civil aviation passenger relationship network as claimed in claim 4, wherein the mutual exclusion factor is used for describing sparsity of the external association of the community structure;
passenger node v in civil aviation passenger relation networkiThe mutual exclusion factor of (2) is expressed as:
Figure FDA0002399697940000023
wherein, PiFor passenger nodes viThe mutual exclusion factor(s) of (c),
Figure FDA0002399697940000024
represents NBiThe middle importance degree is higher than the passenger node viPassenger node vjAnd passenger node viThe maximum similarity between them.
6. The method of claim 5, wherein passenger nodes v in the civil aviation passenger relationship networkiThe degree of representation of (d) is expressed as:
Figure FDA0002399697940000025
wherein R isiFor passenger nodes viDegree of representation of (c).
7. The method for detecting the overlapped community structure of the civil aviation passenger relationship network as claimed in claim 1, wherein the representation degree is used for reflecting the capability of each passenger node to become the core of the community structure to which the passenger node belongs;
the K passenger nodes with the largest representativeness are selected as core passenger nodes of each community structure, other passenger nodes are non-core passenger nodes, and the community membership degree of the initialized core passenger nodes comprises the following steps:
all passenger nodes in the civil aviation passenger relationship network are sorted in descending order according to the aggregation factors, and the sorted passenger node set is recorded as
Figure FDA0002399697940000031
For any 2 passenger nodes therein
Figure FDA0002399697940000032
And
Figure FDA0002399697940000033
satisfies the following conditions: if m>i>j>0, then
Figure FDA0002399697940000034
And
Figure FDA0002399697940000035
satisfies the polymerization factor of Li<Lj；
From a ranked set of passenger nodes
Figure FDA0002399697940000036
Selecting K passenger nodes with the maximum substitution table as core passenger nodes of each community structure, and recording the node as C ═ C1,c2,…,cK) Wherein c iskThe core passenger nodes of the kth community structure in the civil aviation passenger relationship network are represented, and K is more than or equal to 1 and less than or equal to K, and the serial numbers of the core passenger nodes in the civil aviation passenger relationship network are represented;
and initializing the membership degree of each core passenger node with respect to the community structure to which the core passenger node belongs to a preset value, wherein each core passenger node in the civil aviation passenger relationship network determines a community structure, and the membership degree is used for reflecting the overlapping characteristic of the community structures in the civil aviation passenger relationship network.
8. The method for detecting the overlapped community structures of the civil aviation passenger relationship network according to claim 1, wherein each non-core passenger node can simultaneously belong to a plurality of community structures represented by any core passenger node with aggregation factors higher than that of the non-core passenger node;
the membership of each non-core passenger node with respect to the community structure is expressed as:
Figure FDA0002399697940000037
Figure FDA0002399697940000038
wherein m isi,kRepresenting ith passenger node v in civil aviation passenger relationship networkiMembership for a kth community structure; intermediate variable m'i,kIs composed of
Figure FDA0002399697940000039
In the form of a short-hand writing of (1),
Figure FDA00023996979400000310
representing passenger nodes
Figure FDA00023996979400000311
And a polymerization factor higher than
Figure FDA00023996979400000312
A passenger node
Figure FDA00023996979400000313
The degree of similarity between the two images,
Figure FDA00023996979400000314
representing passenger nodes
Figure FDA00023996979400000315
Membership for the kth community structure.
9. The method for detecting the overlapping community structure of the civil aviation passenger relationship network according to claim 1, wherein the step of determining the membership degree of each non-core passenger node with respect to each community structure to obtain the detection result of the overlapping community structure of the civil aviation passenger relationship network comprises the steps of:
and taking a set formed by core passenger nodes representing all community structures, and an obtained set formed by all non-core passenger nodes and membership degrees of all community structures as an overlapped community structure detection result of the civil aviation passenger relationship network.
10. An overlapping community structure detection system for a civil aviation passenger relationship network, comprising:
the building module is used for building a civil aviation passenger relationship network through the incidence relationship among passengers;
the first determination module is used for determining an aggregation factor, a mutual exclusion factor and a representation degree of each passenger node in the civil aviation passenger relationship network, wherein the representation degree is determined by the aggregation factor and the mutual exclusion factor;
the selecting module is used for selecting the K passenger nodes with the largest representativeness as core passenger nodes of each community structure, and other passenger nodes are non-core passenger nodes and initialize the community membership degree of the core passenger nodes; k is the number of community structures contained in the civil aviation passenger relationship network;
and the second determining module is used for determining the membership degree of each non-core passenger node about each community structure to obtain the detection result of the overlapped community structure of the civil aviation passenger relationship network.
CN202010142875.0A 2020-03-04 2020-03-04 Method and system for detecting overlapped community structure of civil aviation passenger relationship network Pending CN111368213A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010142875.0A CN111368213A (en) 2020-03-04 2020-03-04 Method and system for detecting overlapped community structure of civil aviation passenger relationship network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010142875.0A CN111368213A (en) 2020-03-04 2020-03-04 Method and system for detecting overlapped community structure of civil aviation passenger relationship network

Publications (1)

Publication Number Publication Date
CN111368213A true CN111368213A (en) 2020-07-03

Family

ID=71211769

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010142875.0A Pending CN111368213A (en) 2020-03-04 2020-03-04 Method and system for detecting overlapped community structure of civil aviation passenger relationship network

Country Status (1)

Country Link
CN (1) CN111368213A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112163786A (en) * 2020-10-19 2021-01-01 科技谷（厦门）信息技术有限公司 Civil aviation passenger personal influence assessment method based on big data and pagerank algorithm
CN112163785A (en) * 2020-10-19 2021-01-01 科技谷（厦门）信息技术有限公司 Civil aviation passenger personal influence assessment method based on big data and neural network

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104166731A (en) * 2014-08-29 2014-11-26 河海大学常州校区 Discovering system for social network overlapped community and method thereof
CN105654337A (en) * 2015-12-25 2016-06-08 中国民航信息网络股份有限公司 Civil aviation passenger value assessment method
CN108198084A (en) * 2017-12-22 2018-06-22 山西大学 A kind of complex network is overlapped community discovery method

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104166731A (en) * 2014-08-29 2014-11-26 河海大学常州校区 Discovering system for social network overlapped community and method thereof
CN105654337A (en) * 2015-12-25 2016-06-08 中国民航信息网络股份有限公司 Civil aviation passenger value assessment method
CN108198084A (en) * 2017-12-22 2018-06-22 山西大学 A kind of complex network is overlapped community discovery method

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
邵酉辰: "基于复杂网络的民航旅客重叠社区发现研究", 《中国优秀硕士学位论文全文数据库 基础科技辑》 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112163786A (en) * 2020-10-19 2021-01-01 科技谷（厦门）信息技术有限公司 Civil aviation passenger personal influence assessment method based on big data and pagerank algorithm
CN112163785A (en) * 2020-10-19 2021-01-01 科技谷（厦门）信息技术有限公司 Civil aviation passenger personal influence assessment method based on big data and neural network
CN112163786B (en) * 2020-10-19 2024-05-28 科技谷（厦门）信息技术有限公司 Civil aviation passenger personal influence assessment method based on big data and pagerank algorithm

Similar Documents

Publication Publication Date Title
Zhang et al. Community detection in networks with node features
CN103812872B (en) A kind of network navy behavioral value method and system based on mixing Di Li Cray process
CN106682172A (en) Keyword-based document research hotspot recommending method
CN111368213A (en) Method and system for detecting overlapped community structure of civil aviation passenger relationship network
CN108898244B (en) Digital signage position recommendation method coupled with multi-source elements
CN112347372A (en) Method for service promotion of financial enterprise based on user portrait scheme
CN111222847A (en) Open-source community developer recommendation method based on deep learning and unsupervised clustering
CN108959577B (en) Entity matching method and computer program based on non-dominant attribute outlier detection
CN111985119B (en) HarmonySE and CBAM-based architecture trade-off analysis method
CN106126523A (en) A kind of counterfeit money Crime Information analyzes system and the method for analysis
Yan et al. Identification of secondary functional areas and functional structure analysis based on multisource geographic data
CN109885797B (en) Relational network construction method based on multi-identity space mapping
CN113284369A (en) Prediction method for actual measurement airway data based on ADS-B
Cai et al. The mining of urban hotspots based on multi-source location data fusion
CN106611339B (en) Seed user screening method, and product user influence evaluation method and device
CN106202347A (en) A kind of device excavated with useful data for data quality management
CN115878877A (en) Concept drift-based visual detection method for access crawler of aviation server
CN108090132B (en) Community overlapping division method integrating average division distance and structural relationship of labels
CN110442800A (en) A kind of semi-supervised community discovery method of aggregators attribute and graph structure
CN116189008A (en) Remote sensing image change detection method based on fixed point number quantification
Mishra et al. Improving the efficacy of clustering by using far enhanced clustering algorithm
Chu et al. Hierarchical Method for Mining a Prevailing Flight Pattern in Airport Terminal Airspace
CN108615056A (en) A kind of tree enhancing Naive Bayes Classification method based on decomposable asymmetric choice net score function
CN114529096A (en) Social network link prediction method and system based on ternary closure graph embedding
CN111125541B (en) Method for acquiring sustainable multi-cloud service combination for multiple users

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
RJ01 Rejection of invention patent application after publication

Application publication date: 20200703