Skip to main content

Table 3 Effects of temporal factor and structural factor on cascade prediction

From: Information cascades prediction with attention neural network

 MSE (t = 1*)MSE (t = 2*)MSE (t = 3*)
Twitter
 Features-linear (no \(\varvec{T}\))4.1063.8233.715
 Features-linear (no \(\varvec{S}\))3.9763.6403.524
 Features-linear3.8213.5113.423
 Proposed (no \(\varvec{T}\))3.7723.5033.328
 Proposed (no \(\varvec{S}\))3.7163.5403.407
 Proposed (time series \(\varvec{T}\))3.8093.6213.463
 Proposed2.6092.3492.300
AMiner
 Features-linear (no \(\varvec{T}\))2.6212.4072.092
 Features-linear (no \(\varvec{S}\))2.5612.3121.986
 Features-linear2.4292.1361.880
 Proposed (no \(\varvec{T}\))2.4112.0501.799
 Proposed (no \(\varvec{S}\))2.3072.1861.838
 Proposed (time series \(\varvec{T}\))2.4572.1291.906
 Proposed2.1721.6721.534
  1. p.s. t = 1*, where ‘*’ denotes hour for Twitter (year for AMiner)