Proximal Graphical Event Models

Event datasets include events that occur irregularly over the timeline and are prevalent in numerous domains. We introduce proximal graphical event models (PGEM) as a representation of such datasets. PGEMs belong to a broader family of models that characterize relationships between various types of events, where the rate of occurrence of an event type depends only on whether or not its parents have occurred in the most recent history. The main advantage over the state of the art models is that they are entirely data driven and do not require additional inputs from the user, which can require knowledge of the domain such as choice of basis functions or hyperparameters in graphical event models. We theoretically justify our learning of optimal windows for parental history and the choices of parental sets, and the algorithm are sound and complete in terms of parent structure learning. We present additional efficient heuristics for learning PGEMs from data, demonstrating their effectiveness on synthetic and real datasets.

[1]  Philip A. Schrodt,et al.  Conflict and Mediation Event Observations (CAMEO): A New Event Data Framework for the Analysis of Foreign Policy Interactions , 2002 .

[2]  Christopher Meek.  Toward Learning Graphical and Causal Process Models , 2014, CI@UAI.

[3]  Puyang Xu,et al.  A Model for Temporal Dependencies in Event Streams , 2011, NIPS.

[4]  R. Dahlhaus.  Graphical interaction models for multivariate time series1 , 2000 .

[5]  Le Song,et al.  Learning Triggering Kernels for Multi-dimensional Hawkes Processes , 2013, ICML.

[6]  Thore Graepel,et al.  Poisson-Networks: A Model for Structured Poisson Processes. , 2005 .

[7]  Keiji Kanazawa,et al.  A model for reasoning about persistence and causation , 1989 .

[8]  Daphne Koller,et al.  Continuous Time Bayesian Networks , 2012, UAI.

[9]  Sean P. O'Brien,et al.  Crisis Early Warning and Decision Support: Contemporary Approaches and Thoughts on Future Research , 2010 .

[10]  Antonio Gomariz,et al.  SPMF: a Java open-source pattern mining library , 2014, J. Mach. Learn. Res..

[11]  Tian Gao,et al.  Parallel Bayesian Network Structure Learning , 2018, ICML.

[12]  Michael Eichler,et al.  Graphical Models in Time Series Analysis , 1999 .

[13]  Gavin Smith,et al.  Event Series Prediction via Non-Homogeneous Poisson Process Modelling , 2016, 2016 IEEE 16th International Conference on Data Mining (ICDM).

[14]  Daphne Koller,et al.  Ordering-Based Search: A Simple and Effective Algorithm for Learning Bayesian Networks , 2005, UAI.

[15]  Richard Mortier,et al.  CT-NOR: Representing and Reasoning About Events in Continuous Time , 2008, UAI.

[16]  Michael I. Jordan,et al.  Modeling Events with Cascades of Poisson Processes , 2010, UAI.

[17]  Stuart J. Russell,et al.  Dynamic bayesian networks: representation, inference and learning , 2002 .

[18]  Qiang Ji,et al.  Efficient Structure Learning of Bayesian Networks using Constraints , 2011, J. Mach. Learn. Res..

[19]  David Page,et al.  Forest-Based Point Process for Event Prediction from Electronic Health Records , 2013, ECML/PKDD.

[20]  Christopher Meek,et al.  Universal Models of Multivariate Temporal Point Processes , 2016, AISTATS.

[21]  Ankur Parikh,et al.  Conjoint Modeling of Temporal Dependencies in Event Streams , 2012, BMA.

[22]  Vanessa Didelez,et al.  Graphical models for marked point processes based on local independence , 2007, 0710.5874.