dedup-isc-ft-v107-score
float64
0.32
1
uid
stringlengths
32
32
text
stringlengths
0
32.5k
paper_id
stringlengths
1
14
original_image_filename
stringlengths
5
222
0.402953
fc1b75fee30f47b49559e245c60ad29a
E-mail <cit.>. (a) Degree distributions in original and after rewiring networks, (b) Maximum and minimum degrees vs. the number of rewiring in Degree, BP, and SP Non-Preserving. The above three lines show the maximum degrees. The below three lines show the minimum degrees. Violet, green, and light blue denote Degree, BP, SP Non-Preserving. Orange denotes the original degree distribution.
2012.14712
email_multi_deg.jpg
0.468918
a11db168ff5646f889d3e594f460a164
Hamster <cit.>. Comparison of the robustness index R_hub, the size approximate of FVS, and the degree-degree correlation coefficient r vs. the number of rewiring. (Left: a, c, e) Rewirings in Preserving. Violet, green, and light blue solid lines denote the result by Degree, BP, and SP Preserving, respectively. The red dot line indicates a baseline of the conventional best. (Right: b, d, f) Rewirings in Non-Preserving. Violet, green, and light blue solid lines denote the result by Degree, BP, and SP Non-Preserving, respectively.
2012.14712
hamster_multi.jpg
0.430438
e6eb5cc52ff14b39b9655a53503bff1d
Hamster <cit.>. (a) Degree distributions in original and after rewiring networks, (b) Maximum and minimum degrees vs. the number of rewiring in Degree, BP, and SP Non-Preserving. The above three lines show the maximum degrees. The below three lines show the minimum degrees. Violet, green, and light blue denote Degree, BP, SP Non-Preserving. Orange denotes the original degree distribution.
2012.14712
hamster_multi_deg.jpg
0.455561
b6de9e7c6422485da75f02f6260b8c08
Japanese <cit.>. Comparison of the robustness index R_hub, the size approximate of FVS, and the degree-degree correlation coefficient r vs. the number of rewiring. (Left: a, c, e) Rewirings in Preserving. Violet, green, and light blue solid lines denote the result by Degree, BP, and SP Preserving, respectively. The red dot line indicates a baseline of the conventional best. (Right: b, d, f) Rewirings in Non-Preserving. Violet, green, and light blue solid lines denote the result by Degree, BP, and SP Non-Preserving, respectively.
2012.14712
japanese_multi.jpg
0.446536
ace67b1ece1c4a33b809d77d940792a3
Japanese <cit.>. (a) Degree distributions in original and after rewiring networks, (b) Maximum and minimum degrees vs. the number of rewiring in Degree, BP, and SP Non-Preserving. The above three lines show the maximum degrees. The below three lines show the minimum degrees. Violet, green, and light blue denote Degree, BP, SP Non-Preserving. Orange denotes the original degree distribution.
2012.14712
japanese_multi_deg.jpg
0.421282
93b88fe3c76d47efb1b00e5844f16208
OpenFlights <cit.>. Comparison of the robustness index R_hub, the size approximate of FVS, and the degree-degree correlation coefficient r vs. the number of rewiring. (Left: a, c, e) Rewirings in Preserving. Violet, green, and light blue solid lines denote the result by Degree, BP, and SP Preserving, respectively. The red dot line indicates a baseline of the conventional best. (Right: b, d, f) Rewirings in Non-Preserving. Violet, green, and light blue solid lines denote the result by Degree, BP, and SP Non-Preserving, respectively.
2012.14712
openflights_multi.jpg
0.487112
53b54fc3b8f94d95a3ba353dc40607be
OpenFlights <cit.>. (a) Degree distributions in original and after rewiring networks, (b) Maximum and minimum degrees vs. the number of rewiring in Degree, BP, and SP Non-Preserving. The above three lines show the maximum degrees. The below three lines show the minimum degrees. Violet, green, and light blue denote Degree, BP, SP Non-Preserving. Orange denotes the original degree distribution.
2012.14712
openflights_multi_deg.jpg
0.4643
b176364546364da1806e566278590c01
PolBlogs <cit.>. Comparison of the robustness index R_hub, the size approximate of FVS, and the degree-degree correlation coefficient r vs. the number of rewiring. (Left: a, c, e) Rewirings in Preserving. Violet, green, and light blue solid lines denote the result by Degree, BP, and SP Preserving, respectively. The red dot line indicates a baseline of the conventional best. (Right: b, d, f) Rewirings in Non-Preserving. Violet, green, and light blue solid lines denote the result by Degree, BP, and SP Non-Preserving, respectively.
2012.14712
polblogs_multi.jpg
0.473537
0f351da7675847139a47ace6175fe0e9
PolBlogs <cit.>. (a) Degree distributions in original and after rewiring networks, (b) Maximum and minimum degrees vs. the number of rewiring in Degree, BP, and SP Non-Preserving. The above three lines show the maximum degrees. The below three lines show the minimum degrees. Violet, green, and light blue denote Degree, BP, SP Non-Preserving. Orange denotes the original degree distribution.
2012.14712
polblogs_multi_deg.jpg
0.460749
3a20a78c6f6f4a6cb296390424bf2d1b
Power Grid <cit.>. Comparison of the robustness index R_hub, the size approximate of FVS, and the degree-degree correlation coefficient r vs. the number of rewiring. (Left: a, c, e) Rewirings in Preserving. Violet, green, and light blue solid lines denote the result by Degree, BP, and SP Preserving, respectively. The red dot line indicates a baseline of the conventional best. (Right: b, d, f) Rewirings in Non-Preserving. Violet, green, and light blue solid lines denote the result by Degree, BP, and SP Non-Preserving, respectively.
2012.14712
power_multi.jpg
0.44366
7a52e95b35ce4cf08ec9309488807309
Power Grid <cit.>. (a) Degree distributions in original and after rewiring networks, (b) Maximum and minimum degrees vs. the number of rewiring in Degree, BP, and SP Non-Preserving. The above three lines show the maximum degrees. The below three lines show the minimum degrees. Violet, green, and light blue denote Degree, BP, SP Non-Preserving. Orange denotes the original degree distribution.
2012.14712
power_multi_deg.jpg
0.427283
50777aceb32643c0a34cc6fb4e79721e
Yeast <cit.>. Comparison of the robustness index R_hub, the size approximate of FVS, and the degree-degree correlation coefficient r vs. the number of rewiring. (Left: a, c, e) Rewirings in Preserving. Violet, green, and light blue solid lines denote the result by Degree, BP, and SP Preserving, respectively. The red dot line indicates a baseline of the conventional best. (Right: b, d, f) Rewirings in Non-Preserving. Violet, green, and light blue solid lines denote the result by Degree, BP, and SP Non-Preserving, respectively.
2012.14712
yeast_multi.jpg
0.398732
009e0cb858354558aa9014bda586bf2d
Yeast <cit.>. (a) Degree distributions in original and after rewiring networks, (b) Maximum and minimum degrees vs. the number of rewiring in Degree, BP, and SP Non-Preserving. The above three lines show the maximum degrees. The below three lines show the minimum degrees. Violet, green, and light blue denote Degree, BP, SP Non-Preserving. Orange denotes the original degree distribution.
2012.14712
yeast_multi_deg.jpg
0.401146
25cb57889b834da096eb26ec487e7357
Aerial UAV
2012.14713
AerialDroneEnergy2.jpg
0.450228
ff5cf9734c164aaeb5085e2de368e984
Average CPU usage
2012.14713
CPUusageN.jpg
0.459216
e6b4dc9cc82c46639b115f35ede3d003
Classification of applications running on distributed computing resources.
2012.14713
Classification.jpg
0.417279
b1b8590504b94c37b1b1b58d0fa7229e
2012.14713
CollaborativeUnderwater-MasterAirN.jpg
0.391355
86996c602f4a4b0c8c36024f32b76928
2012.14713
CollaborativeUnderwater-MasterUnderwaterN.jpg
0.426453
f6209e4dc332449085f3eda1e9d8f2af
Baseline for collaborative processing
2012.14713
CollaborativeUnderwaterBaselineN.jpg
0.42566
e90f0327e2b7421a82a0f50136871afc
Waterproof encasing
2012.14713
Cooperation1.jpg
0.441723
533e4faa39db445c8137cdb57ca5f797
Cloudlet cooperation
2012.14713
Cooperation2.jpg
0.454831
9998767ecbdc4325bfe09da11e998123
Edge access
2012.14713
Dos-Edge.jpg
0.425484
3f365bd228de4df2974c1a51d28ccdb7
Aerial UAV cloudlet
2012.14713
Drone1.jpg
0.427921
e56f8f3219974b1b9a98af7fd7ff329d
Underwater UAV cloudlet
2012.14713
Drone2.jpg
0.392523
328ca7dc31ad4ea9a3788ce1d78b4945
Ground UAV cloudlet
2012.14713
Drone3.jpg
0.394878
b237af105fbc402089b49a7c1c1c0a02
Energy consumption of S5
2012.14713
EnergyDischarge-S5Rev1.jpg
0.490906
bf8278b28dab425da6e5a49ffa550ded
Energy consumption of RP4
2012.14713
EnergyDischargeRaspberryRev1.jpg
0.531405
d80a3e15b13049ada60f1919759341a0
GEESE System Overview
2012.14713
GEESE-System.jpg
0.399341
544b955c2cd845ae83a740a1660365ba
Ground UAV
2012.14713
GroundDroneEnergy2.jpg
0.50018
3906698ad2ac49ee9eb0a81a147c777c
Average memory usage
2012.14713
MEMusageN.jpg
0.446567
2b123a9d352142849bee9ce1d5191c92
Performance of RP4 to handle workloads.
2012.14713
Raspberry-LoadRev1.jpg
0.459061
c55ed9646f1b4d38b443cc7bf5848fce
Performance of S5 to handle workloads.
2012.14713
Smartphone-LoadRev1.jpg
0.436214
4978f67b4eb141b597f34b18476fb8ce
Sucess rate
2012.14713
SucessRatesN.jpg
0.481743
cb129693a65842be86b9e22d40bc32df
UAVs access
2012.14713
Tres-UAVs.jpg
0.404652
09df2df566204ac08dd948dadd5ea8c5
Underwater UAV
2012.14713
UnderwaterDroneEnergy2.jpg
0.461593
f374673a1be24e8385401341bae270e9
Cloud access
2012.14713
Uno-cloud.jpg
0.398796
6fd00839365c41b680324eebab0ef28c
Transportation of cloudlets to the edge to handle the workload
2012.14713
UseCaseNew.jpg
0.466419
df750fbe8ead4c26b5f2d34887cd2c69
Molecular abundances relative to H_2 as a function of radius in a model of IRC+10216 that assumes 2.5% of interstellar UV photons are able to penetrate to 10^15 cm without being extinguished by dust grains. The solid curves refer to a `smooth' outflow, that is one in which the density distribution follows a 1/r^2 distribution. Dashed lines refer to the clumpy model. Colour code: Green, H_2O; Red, OH; Black, O.
2012.14715
fig1.jpg
0.534175
c694a4b898d14d81ac8e9a338116fd88
As Fig. <ref>, except for a porous, clumpy, mass distribution in the CSE. In particular, these results are for a one-component distribution of mass, taking up a fraction 0.1 of the total volume of the CSE. The solid curves refer to the smooth flow, dashed curves to the one-component model. Colour code: Green, H_2O; Red, OH; Black, O.
2012.14715
fig2.jpg
0.483268
77a8dc6221d441deaece7ddfbc9ed586
As Fig. <ref>, except that the results are for a one-component distribution of mass, taking up a fraction f_vol = 0.01 of the total volume of the CSE. The solid curves refer to the smooth flow, dashed curves to the one-component model. Colour code: Green, H_2O; Red, OH; Black, O.
2012.14715
fig3.jpg
0.467247
8299456df00045daab474819328b62f2
Effective radial extinction in magnitudes of internally generated UV photons for mass-loss rates of 10^-6 M_⊙ yr^-1 (solid lines), 2 × 10^-6 M_⊙ yr^-1 (dashed lines), and 2 × 10^-5 M_⊙ yr^-1 (dotted lines). The cases of smooth, one-component and two-component (weighted) outflows are color coded.
2012.14715
fig4.jpg
0.530239
c2a0eab68b2142e3bd97148a036e6cc2
Water abundance in a model of IRC+10216 that assumes a porous, one component, clump mass distribution in the CSE with f_vol = 0.1 and includes a source of internal UV photons from a stellar blackbody at a temperature of 2330 K. The solid curves refer to the smooth flow, dashed curves to the one-component model. Colour code: Green, H_2O; Red, OH; Black, O.
2012.14715
fig5.jpg
0.465504
860c7e21a08d4c7cb0ac5047beb2ff63
As Fig. <ref>, but for a mass-loss rate of 10^-6 M_⊙ yr^-1 and a source of internal UV photons from a stellar blackbody at a temperature of 2330 K. Here f_vol = 0.01. The solid curves refer to the smooth flow, dashed curves to the one-component model. Colour code: Green, H_2O; Red, OH; Black, O.
2012.14715
fig6.jpg
0.482171
aa789d50ca98485e91bf331305a7a964
Framework of DRL empowered communication network resource control and allocation
2012.14716
DRL.jpg
0.534983
75c09fc9858a419593bab01ebc7aeaf8
Experimental results of the energy efficiency versus the number of BS antennas
2012.14716
EE_BS.jpg
0.464766
cbc73ae6782d4b39b5230055da85d6f6
Experimental results of the energy efficiency versus the number of IoT devices
2012.14716
EE_IoT.jpg
0.40739
f782fa5c66e843f3ba95b9d381af8ff3
Proposed architecture of the AI and IRS-driven wireless communication in 6G IoT
2012.14716
architecture.jpg
0.451454
5af2858a0ace40e9bc0e14e7ab4dbda6
Scenario of AI and IRS empowered communication for 6G IoT
2012.14716
senario.jpg
0.466781
d78d6d472c274a248f6a5542a621b695
The logic and the main notations employed for the dynamical description of an increasing gauge coupling. The two different curves for τ≥ - τ_1 correspond to different values of δ≪ 1. Note that q(δ,γ) = δ/γ [see also Eq. (<ref>)].
2012.14720
FIGURE1.jpg
0.457977
bba4680166734d9a913daba6e11df765
We illustrate the main notations employed when discussing the case of the decreasing gauge coupling. As in the case of Fig. <ref> the two different curves for τ≥ - τ_1 correspond to different values of δ ≪ 1. Note that q( δ , γ ) = δ / γ [see also Eq. (<ref>)].
2012.14720
FIGURE2.jpg
0.486379
cb4e1ed264e14fbfa00f39623056a06e
The stochastic background of binary mergers. In grey we show the 1-σ region consistent with NANOGrav signal.
2012.14721
OmegaGW_Fit_Nano.jpg
0.450979
3278eb05a8fe43bc91779d9b9be796df
Limits on the abundance of PBHs coming from the non observations of μ distortions in the CMB and from large scale structure considerations. We can see that for f_ NL≳3, PBHs with the mass and abundance appropiate to fit NANOGrav are not constrained by μ distortions. However, for these abundances, the formation of clusters imposes M>2× 10^11. The region in black corresponds to the fits shown in Fig. <ref>
2012.14721
mu_distortions_limit.jpg
0.428774
74143c589bad4aa0897eec91a781357c
Visualization of node representation on ACM-3.
2012.14722
ACM_3_visualization.jpg
0.410252
4247978c500c4005822872731b676ce9
Visualization of node representation on ACM-5. Each point indicates a paper and its color denotes the published area.
2012.14722
ACM_5_visualization.jpg
0.435937
30951a41bcc7416facd2328ef052655a
Visualization of node representation on IMDB.
2012.14722
IMDB_visualization.jpg
0.431781
7bc658f116dd43acbbe633e7a0578170
Visualization of the learned attention scores.
2012.14722
case_study_attention.jpg
0.560422
0e8de2e9b4cd491598670a185e3a0df4
Explanation of the hybrid micro/macro level convolution.
2012.14722
hybrid_convolution.jpg
0.449714
fab5e3e2128649b7a76ca3e571db88c3
Effects of the components in the proposed model.
2012.14722
merge_ablation_study.jpg
0.402536
518fb939a3754914891ff514234247dc
Framework of the proposed model.
2012.14722
model.jpg
0.431211
95aab11b42f0447ba89946404509013b
Parameter Sensitivity of the proposed model on IMDB.
2012.14722
parameter_sensitivity.jpg
0.4308
9d975e8852b94999b3fe35b0094ddb7e
Electric discharge in an annular gap. The inner ring represents the cathode emitting electrons at a constant current density and the outer ring represents the grounded anode.The left half is showing the electric field and the right half is showing the electric potential.
2012.14724
Discharge.jpg
0.43686
ef4aa333883c4768bd057b6defe97d8d
Hall effect in an annular channel, streamlines depict the mean velocity of species. Heavy ions (blue) move in axial direction, accelerated by the electric field. Electrons (red) move in azimuthal direction accelerated by the 𝐄×𝐁-drift.
2012.14724
HallEffect.jpg
0.438937
e5c1f135415f4c97bb0efb8fd0847e08
Schematic structure of a picFoam simulation case.
2012.14724
OFstructure.jpg
0.462449
a84d84c4557048bc9b06227657f4dc68
Particle-in-Cell scheme. The classic computation cycle<cit.> is depicted using boxes with rounded off edges, additional boxes show further steps performed by picFoam.
2012.14724
PICscheme.jpg
0.56103
11223e31f8f54235a842b2b05ed374d8
Scheme of a bounded plasma with external RLC circuit in one dimension.
2012.14724
RLCcircuit.jpg
0.588055
3647d8caecba4f649404a8ae61c8d90b
Behavior of an RLC circuit with R = 1 Ω, L = 0.04 H and C = 1 μF interacting with an argon plasma of density n_e = n_i = 10^15 m^-3; comparison between XPDP1 and picFoam.
2012.14724
RLCplasma.jpg
0.459043
b04ba5b50114401d9ea0f54f3b8ca9fc
Comparison of the theoretical progression of current Eq.<ref> in an RLC circuit with R = 1 Ω, L = 1 μH and C = 5 μF to the boundary model implementations in picFoam and PDP1 with no plasma and a permittivity of ϵ=10^20.
2012.14724
RLCtheory.jpg
0.416302
9caa53f841a24b13ba360e97bafdea7c
Final velocity of an electron accelerated through various potential differences.
2012.14724
acceleration-plot.jpg
0.446016
fd8e4162c6674733b3209fa6d08b4f05
Ionization of argon species in collision with electrons at a number density of 10^22 m^-3 using a fixed cross section 3.5e^-20 m^2 corresponding to the initial electron energy of 100 eV.
2012.14724
ionization.jpg
0.626545
42bbcd83f6b74c65a1f6ec18955ece96
Leapfrog scheme. Initially the velocities of the particles are set back by half a time step, resulting in the temporal leap frogging of spatial and velocity data.
2012.14724
leapfrogScheme.jpg
0.434626
bd0bc5f6ad8649a4a9eff63798adae47
Initial charge density over the domain length of the simulation in section <ref> at N^C_P=50, here showing noise when using the cell average method compared to the volume weighting method.
2012.14724
oscillation-noise.jpg
0.387131
653e158d204f4536beee293324091cf4
Plot of the particles total kinetic ∑ 0.5 γ m_e |𝐮|^2 and the corresponding electric fields total potential energy ∑ 0.5 ϵ_0 |𝐄_𝐜|^2/ V_c over time. Two cycles through the equilibrium position correspond to the plasma frequency ω_pe, with its period T=2 π/ω_pe.
2012.14724
oscillation-plot.jpg
0.513168
479a01193b9b4a12b91eab77740508da
Thermalisation of species argon and neon trough binary collisions, simulated with three particle weight differences using the correction scheme of Nanbu and Yonemura<cit.>.
2012.14724
plotBinaryColl.jpg
0.50651
f29b9b602c9240dc9c99b7f0d55eaee1
Thermalisation of electrons and argon ions trough relativistic Coulomb collisions<cit.>, simulated with three particle weight differences using the correction scheme of Sentoku and Kemp<cit.>.
2012.14724
plotCoulomb.jpg
0.539546
278f4d86825c4fbca1747e458450e46f
L_2 norm of the relative error in the charge density field plotted over the number of particles N_P per cell.
2012.14724
plotNoise.jpg
0.442952
7d27c0b6a8aa4653a8827c2a845a35ee
Plot of the potential between the plasma source and the collector for all simulated cases (τ=0.1,1,5). The crosses show the theoretical values calculated from Eq.<ref> and Eq.<ref>.
2012.14724
potentialPlots.jpg
0.492829
d5077f3454494e1782ce3c54dd46ffff
Power spectra of the Fourier transformation of the electron's x-coordinate in time. The inner graph shows the zoom around the tips.
2012.14724
rotation-FFT.jpg
0.450587
5523966ebfdc4eadb92f522d82f818dc
Trajectory of an electron after completing 1000 loops in a constant magnetic field B_z = 1 T.
2012.14724
rotation-plot.jpg
0.489468
539a29de1ddc4cea84d7cf0db6d5aef3
Illustration of one tetrahedron in the cell decomposition of hexagonal cell and of the barycentric description of a particles position.
2012.14724
tetDecompose.jpg
0.461587
68248277e285424ba0add28f5888aeb7
performance for various noise levels p and mixing parameter values λ∈{0, 0.5, 1}.
2012.14727
ablation-plot.jpg
0.435616
ef8f24adb7b84f6cb6da5f8e7ace42b6
Overview of the model. The model first computes edge random walks on two neighborhoods of a given edge (u, v). Each neighbourhood walks are aggregated into S_u, S_v. Both are combined with the edge features f_uv using an Encoder module, which results into the edge embedding vector h_uv. The loss function consists of two parts: structural loss (L_cos) and feature reconstruction loss (L_MSE).
2012.14727
attre2vec-model-v3.jpg
0.41025
5650a20ffda54374a9a6c569cbf15a53
Decoder module architecture
2012.14727
decoder.jpg
0.430871
d779bbf3d9d849b7a4d2b31095725eb5
Encoder module architecture
2012.14727
encoder.jpg
0.428036
df1bedcb96004146bf716f3f617abfd8
2-D representations of ideal and noisy graph edges using with λ∈{0, 0.5, 1}.
2012.14727
initial-figure-v2.jpg
0.420856
738da52cc29748c8b62bc8be18755c5a
Architecture of the MLP(ℳ,ℳ).
2012.14727
mlp2.jpg
0.446951
9831bd7b1252419d9796c57558554109
Architecture of the MLP(ℳ,ℳ,ℱ).
2012.14727
mlp3.jpg
0.403806
d9a791a3f4a546dba12a1e65778e5b47
Average latency with peers classified by client types.
2012.14728
AverageLatencyPerClientType.jpg
0.428873
b721d2d85a974bab8c55dec244be4f0a
Average time connected to peers from each client.
2012.14728
AverageOfConnectedTimePerClientType.jpg
0.435376
b02d30d6311a4bf2b5773f355e246c82
Average connections to peers classified by client types.
2012.14728
AverageOfConnectionsPerClientType.jpg
0.432778
a81277ff860f423690f0a5b469e6b18d
Average disconnections to peers classified by client.
2012.14728
AverageOfDisconnectionsPerClientType.jpg
0.432492
7f7294a5fe1c4189a48ae544abdf596e
Number of Beacon Blocks Received from each Peer.
2012.14728
BeaconBlockMessagePerClient.jpg
0.424312
63d996e45f1b44ba8fc75da0e4d239a1
Number of connections established with each peer.
2012.14728
ConnectionsWithPeers.jpg
0.420269
723584c3a2f14f74a955f40900e90bf3
Number of disconnections established with each peer.
2012.14728
DisconnectionsWithPeers.jpg
0.389368
4d397bbb45354f8abf86bba47b846094
Connection latency with each peer.
2012.14728
LatencyWithPeers.jpg
0.377158
115f1a99632d48ef96cda8f3f18dd75c
Average of messages received on the BeaconAggregateAndProof topic from peers filtered by clients.
2012.14728
MessageAverageFromBeaconAggregateProof.jpg
0.461673
feafc63655f344a5b2ca1cb009cad9b2
Average of messages received on the BeaconBlock topic from peers filtered by clients.
2012.14728
MessageAverageFromBeaconBlock.jpg
0.396866
f9b76d39f2204b8bb7a53c96d0323edf
Number of messages received on the BeaconAggregateAndProof topic from peers filtered by clients.
2012.14728
MessagesFromBeaconAggregateProof.jpg
0.371065
2f752387c57742f284544fedc4f10e19
Number of messages received on the BeaconBlock topic from peers filtered by clients.
2012.14728
MessagesFromBeaconBlock.jpg
0.460563
41a3d633bcc049998612afa90d2f87a9
Number of peers connected from each client.
2012.14728
PeersPerClient.jpg
0.4083
69b86708c44646988f14342f80eb2bdf
Number of peers classified by country of origin.
2012.14728
PeersPerCountries.jpg