CN108039068B - Weighted aviation network community structure division method based on flight delay propagation - Google Patents
Weighted aviation network community structure division method based on flight delay propagation Download PDFInfo
- Publication number
- CN108039068B CN108039068B CN201810009291.9A CN201810009291A CN108039068B CN 108039068 B CN108039068 B CN 108039068B CN 201810009291 A CN201810009291 A CN 201810009291A CN 108039068 B CN108039068 B CN 108039068B
- Authority
- CN
- China
- Prior art keywords
- community
- node
- flight delay
- weighted
- network model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 30
- 239000011159 matrix material Substances 0.000 claims description 26
- 230000010006 flight Effects 0.000 claims description 15
- 230000003111 delayed effect Effects 0.000 claims description 14
- 238000010276 construction Methods 0.000 claims description 9
- 230000002159 abnormal effect Effects 0.000 abstract description 4
- 230000000694 effects Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 238000007621 cluster analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013551 empirical research Methods 0.000 description 1
- JGPMMRGNQUBGND-UHFFFAOYSA-N idebenone Chemical compound COC1=C(OC)C(=O)C(CCCCCCCCCCO)=C(C)C1=O JGPMMRGNQUBGND-UHFFFAOYSA-N 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 230000007306 turnover Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G5/00—Traffic control systems for aircraft, e.g. air-traffic control [ATC]
-
- 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
-
- G06Q50/40—
Abstract
The invention discloses a weighted aviation network community structure division method based on flight delay propagation, which belongs to the technical field of aerospace, and is characterized in that a proper weight index is selected according to characteristics of airport flight delay conditions, airline distances among airports and the like to construct a community division algorithm of a weighted aviation network, so that key airport nodes and key communities with most influence in the network are searched, the positions of the nodes in the communities and the influence of the nodes on the flight delay propagation inside and outside the communities are quantitatively explained, and the technical problems that the influence of key airports and communities on the flight delay propagation in the whole aviation network cannot be analyzed from the aspect of flight delay in the conventional network community division method, so that the positioning of individual airports is abnormal, and the community division is inaccurate are solved.
Description
Technical Field
The invention belongs to the technical field of aerospace, and particularly relates to a weighted aviation network community structure division method based on flight delay propagation.
Background
In recent years, with the rapid development of the aviation industry, the demand of air transportation is continuously increased, and the problem of flight delay caused by the demand is more and more concerned. Research shows that airports of different grades and the degree of association between the airports have different effects on flight delay propagation in the network.
At present, a method for dividing a community structure of an airport group mainly focuses on a hierarchical clustering method in sociology, and by evaluating a collaborative development strategy of the airport group, the correlation characteristics of the airport group are further researched, and a representative method of the method comprises the following steps: cluster analysis, DEA analysis, etc.; or the airport nodes are divided according to the topological structure of the network by using a graph division algorithm in computer science, and the main method is a graph division method and the like. In general, different airports have different effects on the delay spread of flights in the network due to different runway numbers, airport levels and passenger turnover. If the nodes are not distinguished, the key airport nodes with the most influence in the network and the influence effect on other airports cannot be identified, so that flight delay cannot be controlled in a targeted manner. Therefore, the clustering analysis method only considering the point degree and the physical distance inevitably causes the situations of mismatching of the point degree and the betweenness, inconsistency of the high-connectivity node and the central node, and the like. When these anomalies occur, the position of the node in the community, the effect on the network behavior, and the influence on other nodes inside and outside the community cannot be quantitatively described. Thereby causing inaccuracies in community partitioning. On the other hand, through empirical research on the aviation network, the importance of the weight to the understanding of the internal structure of the network, the unbalanced distribution of the weight on the edge and the hierarchical structure of the network are found. This shows that even for the same topology, the network structure characteristics are affected due to the different weights on the edges. Therefore, the unlicensed network is not suitable for dividing the network community structure, and an appropriate weight index needs to be selected for different network behaviors.
The prior art has the following defects: selecting proper weight indexes to carry out community division on the network from the angle of flight delay propagation; the existing method cannot effectively consider and distinguish flight delay conditions of different airports, so that the positioning of individual airports is abnormal, and the network community structure obtained by division is not accurate and comprehensive.
Disclosure of Invention
The invention aims to provide a weighted aviation network community structure division method based on flight delay propagation, and solves the technical problem of providing hardware system support and system architecture for community division of flight delay.
In order to achieve the purpose, the invention adopts the following technical scheme:
a weighted aviation network community structure division method based on flight delay propagation comprises the following steps:
step 1: establishing a data center server and a client server, wherein the data center server is communicated with the client server through the Internet;
establishing a database module for storing data, a weighted flight delay network model construction module for creating a weighted flight delay network model and a community division algorithm module for constructing a community division algorithm model in a data center server;
step 2: setting each airport to be a node, wherein a connecting line between the two nodes is a route, and each route is an edge;
the method comprises the following steps that a manager inputs the number of nodes, the number of delayed flights corresponding to each node, the number of edges and passenger flow information corresponding to all the edges through a client server, the client server produces network parameters, the client server transmits the network parameters to a database module for storage, and the network parameters comprise the number of nodes, the number of delayed flights corresponding to the nodes, the number of edges and the passenger flow corresponding to the edges;
and step 3: the weighted flight delay network model building module reads the network parameters in the database module and builds the weighted flight delay network model through the following steps:
step A: the weighted flight delay network model construction module sets the total number of nodes in the weighted flight delay network model to be N and the number of initial edges to be E according to network parameters; setting the node v and the node u as two arbitrary nodes in the weighted flight delay network model, fvRepresenting the number of delayed flights of node v, fuRepresenting the number of delayed flights of node u, dvuRepresents the distance between node v and node u;
and B: calculating a delay correlation coefficient r between a node v and a node uvu:
And C: setting the edge weight evuFor weighting the edge weights, e, connecting node v and node u in the flight delay network modelvuHas a weight of wvuW is calculated by the following formulavuThe value of (c):
step D: repeatedly executing the step B and the step C, calculating delay correlation coefficients between two nodes corresponding to all edges and weights corresponding to all edges, and constructing a weighted flight delay network model;
and 4, step 4: the community division algorithm module reads the weighted flight delay network model constructed by the weighted flight delay network model construction module, and constructs the community division algorithm model through the following steps:
step E: initializing a weighted flight delay network model, setting each node in the weighted flight delay network model as a community, and setting an initial modularity value of the network as Q which is 0;
if there is an edge connection between node v and node u, the edge weight evu=wvu(ii) a If there is no edge connection between node v and node u, the edge weight e vu0; setting each node to correspond to an auxiliary vector a;
step F: setting the weight of node v to kvK is calculated according to the following formulavThe value of (c):
setting an intermediate quantity m, calculating the value of m according to the following formula:
setting an auxiliary vector a of a node vvA is calculated according to the following formulavThe value of (c):
av=kvrvu/2m;
setting the weight of node u to kuK is calculated according to the following formulauThe value of (c):
setting an auxiliary vector a of a node uuA is calculated according to the following formulauThe value of (c):
au=kurrvu/2m;
step G: initializing a Δ Q matrix whose elements satisfy Δ Qvu=2(evu-avau) Initially, the Δ Q matrix is a sparse matrix;
step H: obtaining the maximum element of each row from the initialized delta Q matrix, forming a maximum pile H, and finding out the maximum delta Q from the maximum pile HvuAnd merging the corresponding community v and community u: if the community v is smaller than the community u, marking the merged community as u; if the community v is larger than the community u, marking the merged community as v;
step I: assuming that the community v is smaller than the community u, the matrix Δ Q, the maximum heap H and the auxiliary vector a are updated according to the following method:
step S1: update matrix Δ Q: removing the v-th row and the u-th column, and updating the elements of the u-th row and the u-th column: setting the community Q as any one except the community v and the community u in the weighted flight delay network model, and if the community Q is connected with the community v and the community u, determining delta Q'uq=ΔQvq+ΔQuq(ii) a Δ Q 'if Community Q is connected only to Community v and is not connected to Community u'vq=ΔQvq-2auaq(ii) a Δ Q 'if Community Q is connected to Community u only and is not connected to Community v'uq=ΔQuq-2avaq;ΔQ′uqThe updated delta Q is set as delta Q';
step S2: updating the maximum stack H according to the delta Q';
step S3: the elements of the auxiliary vector a are updated according to the following formula: a isu=av+auWherein a'uAn auxiliary vector corresponding to the updated community u; a'v=0,a′vAn auxiliary vector corresponding to the updated community v; recording the modularity value Q which is Q + delta Q after combination;
step S4: repeating steps S1 through S3 until all elements Δ Q in the Δ Q matrixvuFrom positive to negative when all elements are Δ QvuWhen all the groups are negative, a Q function peak value is obtained, and the corresponding community division is the optimal division mode;
and 5: analyzing key nodes and key communities according to community division results obtained in the step 4, wherein the average value of the community internal side rights represents the compactness of the community internal connection; the external community boundaries represent the mutual influence among communities; the sum of the weights of the edges represents how closely the inter-community influence is.
The modularity value Q is a modular modularity function representing the degree of modularity of the network.
The invention relates to a weighted aviation network community structure division method based on flight delay propagation, which solves the problem that the conventional aviation network community structure division method cannot select proper weight indexes to carry out community division on a network from the perspective of flight delay propagation; the method has the technical problems that flight delay conditions of different airports cannot be effectively considered and distinguished, the positioning of individual airports is abnormal, and the divided network community structure is not accurate and comprehensive enough, redefines network weight from the perspective of flight delay propagation, considers the influence conditions (such as the number of delayed flights) of different airport nodes on delay propagation and the distance between airports, and divides the community structure of the aviation network airport group; the invention constructs a reasonable weighted flight delay network model, divides communities on the basis of the model, and more accurately searches key airport nodes and community structures which influence delay propagation.
Drawings
FIG. 1 is a general flow chart of the present invention;
FIG. 2 is a flow chart of step 2 through step 3 of the present invention;
fig. 3 is a flow chart of step 4 of the present invention.
Detailed Description
1-3, a method for dividing a weighted airline network community structure based on flight delay propagation includes the following steps:
step 1: establishing a data center server and a client server, wherein the data center server is communicated with the client server through the Internet;
establishing a database module for storing data, a weighted flight delay network model construction module for creating a weighted flight delay network model and a community division algorithm module for constructing a community division algorithm model in a data center server;
step 2: setting each airport to be a node, wherein a connecting line between the two nodes is a route, and each route is an edge;
the method comprises the following steps that a manager inputs the number of nodes, the number of delayed flights corresponding to each node, the number of edges and passenger flow information corresponding to all the edges through a client server, the client server produces network parameters, the client server transmits the network parameters to a database module for storage, and the network parameters comprise the number of nodes, the number of delayed flights corresponding to the nodes, the number of edges and the passenger flow corresponding to the edges;
and step 3: the weighted flight delay network model building module reads the network parameters in the database module and builds the weighted flight delay network model through the following steps:
step A: the weighted flight delay network model construction module sets the total number of nodes in the weighted flight delay network model to be N and the number of initial edges to be E according to network parameters; setting the node v and the node u as two arbitrary nodes in the weighted flight delay network model, fvRepresenting the number of delayed flights of node v, fuRepresenting the number of delayed flights of node u, dvuRepresents the distance between node v and node u;
and B: calculating a delay correlation coefficient r between a node v and a node uvu:
And C: setting the edge weight evuFor weighting the edge weights, e, connecting node v and node u in the flight delay network modelvuHas a weight of wvuW is calculated by the following formulavuThe value of (c):
step D: repeatedly executing the step B and the step C, calculating delay correlation coefficients between two nodes corresponding to all edges and weights corresponding to all edges, and constructing a weighted flight delay network model;
where u and v represent nodes in the network, each node is initially a community, and the elements of the matrix represent the degree of association between the nodes (similar to an adjacency matrix, where connected nodes correspond to the element r in the matrix)vu);
As the matrix is continuously updated, u and v slowly become communities, are not independent nodes any more, and are denser, at the moment, u and v represent the communities u and v, and the matrix elements represent the degree of association between the two communities.
And 4, step 4: the community division algorithm module reads the weighted flight delay network model constructed by the weighted flight delay network model construction module, and constructs the community division algorithm model through the following steps:
step E: initializing a weighted flight delay network model, setting each node in the weighted flight delay network model as a community, and setting an initial modularity value of the network as Q which is 0;
if there is an edge connection between node v and node u, the edge weight evu=wvu(ii) a If there is no edge connection between node v and node u, the edge weight e vu0; setting each node to correspond to an auxiliary vector a;
step F: setting the weight of node v to kvK is calculated according to the following formulavThe value of (c):
setting an intermediate quantity m, calculating the value of m according to the following formula:
setting an auxiliary vector a of a node vvA is calculated according to the following formulavThe value of (c):
av=kvrvu/2m;
setting the weight of node u to kuK is calculated according to the following formulauThe value of (c):
setting an auxiliary vector a of a node uuA is calculated according to the following formulauThe value of (c):
au=kurvu/2m;
step G: initializing a Δ Q matrix whose elements satisfy Δ Qvu=2(evu-avau) Initially, the Δ Q matrix is a sparse matrix;
step H: obtaining the maximum element of each row from the initialized delta Q matrix, forming a maximum pile H, and finding out the maximum delta Q from the maximum pile HvuAnd merging the corresponding community v and community u: if the community v is smaller than the community u, marking the merged community as u; if the community v is larger than the community u, marking the merged community as v;
step I: assuming that the community v is smaller than the community u, the matrix Δ Q, the maximum heap H and the auxiliary vector a are updated according to the following method:
step S1: update matrix Δ Q: removing the v-th row and the u-th column, and updating the elements of the u-th row and the u-th column: setting the community Q as any one except the community v and the community u in the weighted flight delay network model, and if the community Q is connected with the community v and the community u, determining delta Q'uq=ΔQvq+ΔQuq(ii) a Δ Q 'if Community Q is connected only to Community v and is not connected to Community u'vq=ΔQvq-2auaq(ii) a Δ Q 'if Community Q is connected to Community u only and is not connected to Community v'uq=ΔQuq-2avaq;ΔQ′uqThe updated delta Q is set as delta Q';
step S2: updating the maximum stack H according to the delta Q';
step S3: the elements of the auxiliary vector a are updated according to the following formula: a isu=av+auWherein a'uAn auxiliary vector corresponding to the updated community u; a isv=0,avAn auxiliary vector corresponding to the updated community v; recording the modularity value Q which is Q + delta Q after combination;
step S4: repeating steps S1 through S3 until all elements Δ Q in the Δ Q matrixvuFrom positive to negative when all elements are Δ QvuWhen all the groups are negative, a Q function peak value is obtained, and the corresponding community division is the optimal division mode;
and 5: analyzing key nodes and key communities according to community division results obtained in the step 4, wherein the average value of the community internal side rights represents the compactness of the community internal connection; the external community boundaries represent the mutual influence among communities; the sum of the weights of the edges represents how closely the inter-community influence is.
The modularity value Q is a modular modularity function representing the degree of modularity of the network.
The invention relates to a weighted aviation network community structure division method based on flight delay propagation, which solves the problem that the conventional aviation network community structure division method cannot select proper weight indexes to carry out community division on a network from the perspective of flight delay propagation; the method has the technical problems that flight delay conditions of different airports cannot be effectively considered and distinguished, the positioning of individual airports is abnormal, and the divided network community structure is not accurate and comprehensive enough, redefines network weight from the perspective of flight delay propagation, considers the influence conditions (such as the number of delayed flights) of different airport nodes on delay propagation and the distance between airports, and divides the community structure of the aviation network airport group; the invention constructs a reasonable weighted flight delay network model, divides communities on the basis of the model, and more accurately searches key airport nodes and community structures which influence delay propagation.
Claims (2)
1. A weighted aviation network community structure division method based on flight delay propagation is characterized by comprising the following steps: the method comprises the following steps:
step 1: establishing a data center server and a client server, wherein the data center server is communicated with the client server through the Internet;
establishing a database module for storing data, a weighted flight delay network model construction module for creating a weighted flight delay network model and a community division algorithm module for constructing a community division algorithm model in a data center server;
step 2: setting each airport to be a node, wherein a connecting line between the two nodes is a route, and each route is an edge;
the method comprises the following steps that a manager inputs the number of nodes, the number of delayed flights corresponding to each node, the number of edges and passenger flow information corresponding to all the edges through a client server, the client server produces network parameters, the client server transmits the network parameters to a database module for storage, and the network parameters comprise the number of nodes, the number of delayed flights corresponding to the nodes, the number of edges and the passenger flow corresponding to the edges;
and step 3: the weighted flight delay network model building module reads the network parameters in the database module and builds the weighted flight delay network model through the following steps:
step A: the weighted flight delay network model construction module sets the total number of nodes in the weighted flight delay network model to be N and the number of initial edges to be E according to network parameters; setting the node v and the node u as two arbitrary nodes in the weighted flight delay network model, fvRepresenting the number of delayed flights of node v, fuRepresenting the number of delayed flights of node u, dvuRepresents the distance between node v and node u;
and B: calculating a delay correlation coefficient r between a node v and a node uvu:
And C: setting the edge weight evuFor weighting the edge weights, e, connecting node v and node u in the flight delay network modelvuHas a weight of wvuW is calculated by the following formulavuThe value of (c):
step D: repeatedly executing the step B and the step C, calculating delay correlation coefficients between two nodes corresponding to all edges and weights corresponding to all edges, and constructing a weighted flight delay network model;
and 4, step 4: the community division algorithm module reads the weighted flight delay network model constructed by the weighted flight delay network model construction module, and constructs the community division algorithm model through the following steps:
step E: initializing a weighted flight delay network model, setting each node in the weighted flight delay network model as a community, and setting an initial modularity value of the network as Q which is 0;
if there is an edge connection between node v and node u, the edge weight evu=wvu(ii) a If there is no edge connection between node v and node u, the edge weight evu0; setting each node to correspond to an auxiliary vector a;
step F, setting the weight of the node v as kvK is calculated according to the following formulavThe value of (c):
setting an intermediate quantity m, calculating the value of m according to the following formula:
setting the auxiliary direction of the node vQuantity avA is calculated according to the following formulavThe value of (c):
av=kvrvu/2m;
setting the weight of node u to kuK is calculated according to the following formulauThe value of (c):
setting an auxiliary vector a of a node uuA is calculated according to the following formulauThe value of (c):
au=kurvu/2m;
step G: initializing a Δ Q matrix whose elements satisfy Δ Qvu=2(evu-avau) Initially, the Δ Q matrix is a sparse matrix;
step H: obtaining the maximum element of each row from the initialized delta Q matrix, forming a maximum pile H, and finding out the maximum delta Q from the maximum pile HvuAnd merging the corresponding community v and community u: if the community v is smaller than the community u, marking the merged community as u; if the community v is larger than the community u, marking the merged community as v;
step I: assuming that the community v is smaller than the community u, the matrix Δ Q, the maximum heap H and the auxiliary vector a are updated according to the following method:
step S1: update matrix Δ Q: setting the community Q as any one of communities except the community v and the community u in the weighted flight delay network model, and if the community Q is connected with the community v and the community u, delta Q'uq=ΔQvq+ΔQuq(ii) a Δ Q 'if Community Q is connected only to Community v and is not connected to Community u'vq=ΔQvq-2auaq(ii) a Δ Q 'if Community Q is connected to Community u only and is not connected to Community v'uq=ΔQuq-2avaq;ΔQ′uqThe updated delta Q is set as delta Q';
step S2: updating the maximum stack H according to the delta Q';
step S3: updating elements of auxiliary vector a according to formula a'u=av+auWherein a'uAn auxiliary vector corresponding to the updated community u; a'v=0,a′vAn auxiliary vector corresponding to the updated community v; recording the modularity value Q which is Q + delta Q after combination;
step S4: repeating steps S1 through S3 until all elements Δ Q in the Δ Q matrixvuFrom positive to negative when all elements are Δ QvuWhen all the groups are negative, a Q function peak value is obtained, and the corresponding community division is the optimal division mode;
and 5: analyzing key nodes and key communities according to community division results obtained in the step 4, wherein the average value of the community internal side rights represents the compactness of the community internal connection; the external community boundaries represent the mutual influence among communities; the sum of the weights of the edges represents how closely the inter-community influence is.
2. The method as claimed in claim 1, wherein the modularity value Q is a modular modularization function representing the modularization degree of the network.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810009291.9A CN108039068B (en) | 2018-01-05 | 2018-01-05 | Weighted aviation network community structure division method based on flight delay propagation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810009291.9A CN108039068B (en) | 2018-01-05 | 2018-01-05 | Weighted aviation network community structure division method based on flight delay propagation |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108039068A CN108039068A (en) | 2018-05-15 |
CN108039068B true CN108039068B (en) | 2021-08-24 |
Family
ID=62098874
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810009291.9A Active CN108039068B (en) | 2018-01-05 | 2018-01-05 | Weighted aviation network community structure division method based on flight delay propagation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108039068B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109190700B (en) * | 2018-08-27 | 2020-06-12 | 北京航空航天大学 | Quantitative analysis method for aviation delay propagation |
CN111191843B (en) * | 2019-12-30 | 2023-07-18 | 南京航空航天大学 | Airport delay prediction method based on time sequence network propagation dynamics equation |
CN111506716B (en) * | 2020-04-15 | 2023-04-25 | 腾讯科技(深圳)有限公司 | Question-answer data processing method and device and computer readable storage medium |
CN112509382B (en) * | 2020-11-13 | 2021-10-22 | 南京航空航天大学 | Method for analyzing flight delay propagation rule |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105813235A (en) * | 2014-12-31 | 2016-07-27 | 中国电信股份有限公司 | Mobile terminal client community division method and division system |
CN106301888A (en) * | 2016-07-27 | 2017-01-04 | 西安电子科技大学 | Based on core node and the network community division method of community's convergence strategy |
WO2017108133A1 (en) * | 2015-12-23 | 2017-06-29 | Swiss Reinsurance Company Ltd. | Automated, reactive flight-delay risk-transfer system and method thereof |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9384668B2 (en) * | 2012-05-09 | 2016-07-05 | Singularity University | Transportation using network of unmanned aerial vehicles |
US9348333B1 (en) * | 2014-12-08 | 2016-05-24 | Amazon Technologies, Inc. | Modular air delivery |
-
2018
- 2018-01-05 CN CN201810009291.9A patent/CN108039068B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105813235A (en) * | 2014-12-31 | 2016-07-27 | 中国电信股份有限公司 | Mobile terminal client community division method and division system |
WO2017108133A1 (en) * | 2015-12-23 | 2017-06-29 | Swiss Reinsurance Company Ltd. | Automated, reactive flight-delay risk-transfer system and method thereof |
CN106301888A (en) * | 2016-07-27 | 2017-01-04 | 西安电子科技大学 | Based on core node and the network community division method of community's convergence strategy |
Non-Patent Citations (1)
Title |
---|
基于基础延误预测的航班计划优化研究;吴薇薇 等;《交通运输系统工程与信息》;20161231;第16卷(第6期);第189-195页 * |
Also Published As
Publication number | Publication date |
---|---|
CN108039068A (en) | 2018-05-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108039068B (en) | Weighted aviation network community structure division method based on flight delay propagation | |
Tudisco et al. | Node and layer eigenvector centralities for multiplex networks | |
Li et al. | A comparative analysis of evolutionary and memetic algorithms for community detection from signed social networks | |
CN112925989B (en) | Group discovery method and system of attribute network | |
Donetti et al. | Improved spectral algorithm for the detection of network communities | |
CN105700549A (en) | Unmanned plane multi-track planning method based on sequence ecological niche PSO (particle swarm optimization) algorithm | |
CN101404071B (en) | Electronic circuit fault diagnosis neural network method based on grouping particle swarm algorithm | |
CN101694652A (en) | Network resource personalized recommended method based on ultrafast neural network | |
CN110164129B (en) | Single-intersection multi-lane traffic flow prediction method based on GERNN | |
CN102571431B (en) | Group concept-based improved Fast-Newman clustering method applied to complex network | |
Zadeh et al. | An efficient metamodel-based multi-objective multidisciplinary design optimization framework | |
CN114172942A (en) | Cooperative task allocation and track optimization method for multi-unmanned aerial vehicle assisted Internet of things | |
Amelio et al. | An evolutionary and local refinement approach for community detection in signed networks | |
CN112765415A (en) | Link prediction method based on relational content joint embedding convolution neural network | |
CN111121784A (en) | Unmanned reconnaissance aircraft route planning method | |
Amrit et al. | Design strategies for multi-objective optimization of aerodynamic surfaces | |
Yang et al. | A closer look at the adversarial robustness of deep equilibrium models | |
CN114556364A (en) | Neural architecture search based on similarity operator ordering | |
CN112989526A (en) | Aviation network key node identification method based on kernel extreme learning machine | |
CN116956740A (en) | User preference model construction method, system and application for plane truss structure | |
Lin et al. | Dynamical representation learning for Ethereum transaction network via non-negative adaptive latent factorization of tensors | |
CN116862023A (en) | Robust federal learning abnormal client detection method based on spectral clustering | |
CN104657429A (en) | Complex-network-based technology-driven product innovation method | |
Zhao et al. | Research and comparison on identification and prediction methods of air traffic network congestion | |
CN110941767A (en) | Network community detection countermeasure enhancement method based on multi-similarity integration |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |