Discovering Causal Chains by Integrating Plan Recognition and Sequential Pattern Mining

In this paper we define the notion of causal chains. Causal chains are a particular kind of sequential patterns that reflect causality relations according to background knowledge. We also present an algorithm for mining causal chains from a collection of action traces. We run this algorithm on a realworld domain and observe that causal chains can be computed efficiently by quickly identifying inter-related actions.

• Type: Conference
• Year: 2004
• Where: FLAIRS
• Authors: Shreeram Sahasrabudhe, Héctor Muñoz-Avila
• Keywords: Artificial Intelligence, Causal Chains, FLAIRS 2004, Mining Causal Chains, We Run This