Accès libre

Using Network Embedding to Obtain a Richer and More Stable Network Layout for a Large Scale Bibliometric Network

Ting Chen
Ting Chen
, 
Guopeng Li
Guopeng Li
, 
Qiping Deng
Qiping Deng
 et 
Xiaomei Wang
Xiaomei Wang
   | 08 déc. 2020
Journal of Data and Information Science's Cover Image
À propos de cet article
Article précédent
Article suivant
Citez
Article Category: Research Paper
Publié en ligne: 08 déc. 2020
Pages: 154 - 177
Reçu: 01 juil. 2020
Accepté: 20 oct. 2020
DOI: https://doi.org/10.2478/jdis-2021-0006
Mots clés
© 2021 Ting Chen et al., published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Figure 1

Three base maps created by OpenOrd with different cutting edges strategies and parameters. A is the map with 47,294 highly cited papers and all 3.6 million co-citation relationships, and OpenOrd cutting edge parameter is 0.8. B is the map with top 15 weighted relationships of each paper, and cutting edge parameter is 0.85. C is the map also with top 15 weighted relationships but cutting edge parameter is 0.9.
Three base maps created by OpenOrd with different cutting edges strategies and parameters. A is the map with 47,294 highly cited papers and all 3.6 million co-citation relationships, and OpenOrd cutting edge parameter is 0.8. B is the map with top 15 weighted relationships of each paper, and cutting edge parameter is 0.85. C is the map also with top 15 weighted relationships but cutting edge parameter is 0.9.

Figure 2

The classical process of network clustering and visualization.
The classical process of network clustering and visualization.

Figure 3

The flowchart of the proposed visualization approach.
The flowchart of the proposed visualization approach.

Figure 4

Illustrations of visualization of highly cited paper co-citation network by different embedding models with t-SNE.
Illustrations of visualization of highly cited paper co-citation network by different embedding models with t-SNE.

Figure 5

Research fronts overlapped on the classic co-citation map A and Node2Vec map with a zoom-in view.
Research fronts overlapped on the classic co-citation map A and Node2Vec map with a zoom-in view.

Figure 6

Layouts of nine Node2Vec maps with the same parameters.
Layouts of nine Node2Vec maps with the same parameters.

Figure 7

Nine OpenOrd maps with top 15 edges of each nodes and cutting edge parameter = 0.9.
Nine OpenOrd maps with top 15 edges of each nodes and cutting edge parameter = 0.9.

Detail of top 10 ESI fronts of each discipline overlaid in Figure 5.

ID&FieldResearch front keywords#paper
993, Social sciences, generalPolice agency directed randomized field trial; police legitimacy invariant; police legitimacy; police officer body-worn cameras (bwcs)30
39, Clinical medicineTriple-negative essential thrombocythemia patients; myeloproliferative neoplasm-associated myelofibrosis; primary myelofibrosis50
514, Computer scienceMultigranulation fuzzy decision-theoretic rough set; triangular fuzzy decision-theoretic rough sets; intuitionistic fuzzy decision-theoretic rough sets; multigranulation decision-theoretic rough sets;47
108, PhysicsFrozen quantum coherence; measuring quantum coherence; genuine quantum coherence; quantum coherence; quantum thermodynamics49
90, Biology & biochemistryReversible m(6)a rna methylation; rna m(6)a methylation; m(6)a rna methylation; m(6)a rna modification controls cell fate transition; viral m(6)a rna methylomes42
309, EngineeringParticle swarm optimization-based maximum power point tracking algorithm; maximum power point tracking control techniques; maximum power point tracking (mppt) techniques; observe (p&o) maximum power point tracking (mppt) algorithm; maximum power point tracking techniques36
69, GeosciencesGreenland ice sheet surface mass balance contribution; 1900–2015 greenland ice sheet surface mass balance; potential antarctic ice sheet retreat driven; northeast greenland ice sheet triggered; simple antarctic ice sheet model43
63, Chemistry3d porous crystalline polyimide covalent organic frameworks; crystalline 2d covalent organic frameworks; homochiral 2d porous covalent organic frameworks; large-pore crystalline polyimide covalent organic frameworks; highly crystalline covalent organic frameworks45
107, Materials scienceEquiatomic high-entropy alloy crmnfeconi; equiatomic high-entropy alloys; single-phase high-entropy alloy crmnfeconi; nanocrystalline cocrfemnni high-entropy alloy; single-phase high-entropy alloys37
71, Space scienceSmall kepler planets; kepler transit candidates xvii; potentially habitable planets orbiting m dwarfs estimated; earth-size planets orbiting sun-like stars31

Silhouette Coefficient of research fronts on five maps.

Number of RFCo-citation map ACo-citation map BCo-citation map CN2V & tSNEDW & tSNELine & tSNE
100.8070.710.730.9550.8390.637
500.6330.580.60.7210.5430.438
1000.5890.520.560.6560.4370.370
2000.5290.430.420.5700.3530.306
5000.4070.290.290.4380.1930.190
1,0000.3020.180.270.3130.0470.054
10,096−0.227−0.26−0.26−0.279−0.469−0.455

Models (used in this paper) categorized by building and learning method.

ModelBuilding context of nodesLearning method
DeepWalkTruncated Random WalksSkip-gram & Hierarchical Softmax
Line1-hop and 2-hop NeighborsSkip-gram & Negative Sampling
Node2VecBiased Truncated Random WalksSkip-gram & Negative Sampling
eISSN:
2543-683X
Langue:
Anglais
Périodicité:
4 fois par an
Sujets de la revue:
Computer Sciences, Information Technology, Project Management, Databases and Data Mining
RSS Feed de la revue