Learning user-specific latent influence and susceptibility from information cascades

22 Jan 2015  ·  Wang Yongqing, Shen Hua-Wei, Liu Shenghua, Cheng Xue-Qi ·

Predicting cascade dynamics has important implications for understanding information propagation and launching viral marketing. Previous works mainly adopt a pair-wise manner, modeling the propagation probability between pairs of users using n^2 independent parameters for n users. Consequently, these models suffer from severe overfitting problem, specially for pairs of users without direct interactions, limiting their prediction accuracy. Here we propose to model the cascade dynamics by learning two low-dimensional user-specific vectors from observed cascades, capturing their influence and susceptibility respectively. This model requires much less parameters and thus could combat overfitting problem. Moreover, this model could naturally model context-dependent factors like cumulative effect in information propagation. Extensive experiments on synthetic dataset and a large-scale microblogging dataset demonstrate that this model outperforms the existing pair-wise models at predicting cascade dynamics, cascade size, and "who will be retweeted".

PDF Abstract
No code implementations yet. Submit your code now

Categories


Social and Information Networks Physics and Society

Datasets


  Add Datasets introduced or used in this paper