Hierarchical Incomplete Multi-source Feature Learning for Spatiotemporal Event Forecasting
Liang Zhao*, VT; Jieping Ye, University of Michigan at Ann Arbor; Feng Chen, SUNY Albany; Chang-Tien Lu, Virginia Tech; Naren Ramakrishnan, Virginia Tech
Forecasting signiﬁcant societal events is an interesting and challenging problem as it taking into consideration multiple aspects of a society, including its economics, politics, and culture. Traditional forecasting methods based on a single data source ﬁnd it hard to cover all these aspects comprehensively, thus limiting model performance. Multi-source event forecasting has proven promising but still suﬀers from several challenges, including 1) geographical hierarchies in multi-source data features, 2) missing values, and 3) characterization of structured feature sparsity. This paper proposes a novel feature learning model that concurrently addresses all the above challenges. Speciﬁcally, given multi-source data from diﬀerent geographical levels, we design a new fore-casting model by characterizing the lower-level features’ dependence on higher-level features. To handle the correlations amidst structured feature sets and deal with missing values among the coupled features, we propose a novel feature learning model based on an Nth-order strong hierarchy and fused-overlapping group Lasso. An eﬃcient algorithm is developed to optimize model parameters and ensure global optima. Extensive experiments on 10 datasets in diﬀerent domains demonstrate the eﬀectiveness and eﬃciency of the proposed model.
Filed under: Dimensionality Reduction