TensorCast: Forecasting with Context using Coupled Tensors

Miguel Araújo, Pedro Ribeiro and Christos Faloutsos

2017

Abstract

Given an heterogeneous social network, can we forecast its future? Can we predict who will start using a given hashtag on twitter? Can we leverage side information, such as who retweets or follows whom, to improve our membership forecasts? We present TensorCast, a novel method that forecasts time-evolving networks more accurately than current state of the art methods by incorporating multiple data sources in coupled tensors. TensorCast is (a) scalable, being linearithmic on the number of connections; (b) effective, achieving over 20% improved precision on top-1000 forecasts of community members; (c) general, being applicable to data sources with different structure. We run our method on multiple real-world networks, including DBLP and a Twitter temporal network with over 310 million non-zeros, where we predict the evolution of the activity of the use of political hashtags.

Publication in PDF format

pdf

Software

software TensorCast

Journal/Conference/Book

IEEE International Conference on Data Mining

Awards/Notice

Best Paper Award

Reference (text)

Miguel Araújo, Pedro Ribeiro and Christos Faloutsos . TensorCast: Forecasting with Context using Coupled Tensors. (Best Paper Award) Proceedings of the IEEE International Conference on Data Mining (ICDM), IEEE, New Orleans, USA, November, 2017.

Bibtex

@inproceedings{ribeiro-ICDM2017,
  author = {Miguel Araújo and  Pedro Ribeiro and Christos Faloutsos },
  title = {TensorCast: Forecasting with Context using Coupled Tensors},
  booktitle = {IEEE International Conference on Data Mining},
  publisher = {IEEE},
  month = {November},
  year = {2017}
}