Conjoint Modeling of Temporal Dependencies in Event Streams
Abstract
Many real world applications depend on modeling the temporal dynamics of streams of diverse events, many of which are rare. We introduce a novel model class, Conjoint Piecewise-Constant Conditional Intensity Models, and a learning algorithm that together yield a data-driven approach to parameter sharing with the aim of better modeling such event streams. We empirically demonstrate that our approach yields more accurate models of two real world data sets: search query logs and data center system logs. 1
Cite
Text
Parikh et al. "Conjoint Modeling of Temporal Dependencies in Event Streams." Conference on Uncertainty in Artificial Intelligence, 2012.Markdown
[Parikh et al. "Conjoint Modeling of Temporal Dependencies in Event Streams." Conference on Uncertainty in Artificial Intelligence, 2012.](https://mlanthology.org/uai/2012/parikh2012uai-conjoint/)BibTeX
@inproceedings{parikh2012uai-conjoint,
title = {{Conjoint Modeling of Temporal Dependencies in Event Streams}},
author = {Parikh, Ankur and Gunawardana, Asela and Meek, Christopher},
booktitle = {Conference on Uncertainty in Artificial Intelligence},
year = {2012},
pages = {65-73},
url = {https://mlanthology.org/uai/2012/parikh2012uai-conjoint/}
}