A Model for Reasoning with Uncertain Rules in Event Composition Systems

TL;DR

This paper introduces a formal probabilistic framework for reasoning about uncertain event inference in active systems, enabling automatic decision-making based on complex, uncertain event patterns.

Contribution

It presents the first formal model combining event composition with probabilistic reasoning using Bayesian networks for active systems.

Findings

Defines a probabilistic event inference framework

Constructs Bayesian networks for probability calculation

Enables active systems to handle uncertainty in event detection

Abstract

In recent years, there has been an increased need for the use of active systems - systems required to act automatically based on events, or changes in the environment. Such systems span many areas, from active databases to applications that drive the core business processes of today's enterprises. However, in many cases, the events to which the system must respond are not generated by monitoring tools, but must be inferred from other events based on complex temporal predicates. In addition, in many applications, such inference is inherently uncertain. In this paper, we introduce a formal framework for knowledge representation and reasoning enabling such event inference. Based on probability theory, we define the representation of the associated uncertainty. In addition, we formally define the probability space, and show how the relevant probabilities can be calculated by dynamically constructing a Bayesian network. To the best of our knowledge, this is the first work that enables taking such uncertainty into account in the context of active systems. herefore, our contribution is twofold: We formally define the representation and semantics of event composition for probabilistic settings, and show how to apply these extensions to the quantification of the occurrence probability of events. These results enable any active system to handle such uncertainty.