Towards the Explanation of Workflows James R. Michaelis, Li Ding, and Deborah L. McGuinness Tetherless World Constellation, Rensselaer Polytechnic Institute, 110 8 th Street, Troy, NY 12180, USA {michaj6, dingl, dlm}@cs.rpi.edu Abstract. Across many fields involving complex computing, software systems are being augmented with workflow logging functionality. The log data can be effectively organized using declarative structured languages such as OWL; however, such declarative encodings alone are not enough to facilitate understandable workflow systems with high quality explanation. In this paper, we present our approach for visually explaining OWL-encoded workflow logs for complex systems, which includes the following steps: (i) identifying and normalizing provenance in workflow logs using the provenance interlingua PML2, (ii) using this provenance information, as well as supplemental log data, building an abstracted workflow representation known as a RITE network (capable of storing workflow state Relationships, Identifiers, Types, and Explanations), and (iii) visualizing the workflow log by displaying its provenance information as a directed acyclic graph and presenting supplemental explanations for individual workflow states and relationships. To demonstrate these techniques, we describe the design of a workflow explainer for the Generalized Integrated Learning Architecture (GILA) – a multi-agent platform designed to use multiple learners to solve problems such as resolving airspace allocation conflicts. We also comment on how our approach can be generalized to explain other complex workflow systems. 1 Introduction Increasingly, application developers are making an effort to audit system activities using intensive logging, so as to provide increased insight into the underlying workflows of their systems. In addition to recording system activities, such workflow logs can explicitly capture both system operating contexts and capabilities. Inherently, this data can become very complex as system complexity increases, resulting in diminished intuitiveness for human auditors. One coping strategy is to facilitate the development of visualization-oriented interfaces aimed at helping auditors explore, understand and validate these complex workflow logs. While most workflow logs are inherently well structured, this won’t automatically make it easy for auditors to explore and interpret them. A small piece of structured domain data may not be hard to understand. However, a huge amount of such data could easily overwhelm a typical human’s perception capacity. Moreover, the domain-specific structure of workflow data typically requires auditors to know a non- trivial amount of domain knowledge.