Situation-Aware AI Explainability via Complex Event Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional AI-endowed, process-aware systems (APASs) lack situational awareness, resulting in incomplete explanations of their decision-making processes, which are crucial for regulatory compliance and system optimization but are inadequate due to their inability to account for contextual conditions affecting process outcomes.
Innovation Solution
A method utilizing a complex event processing (CEP) engine to generate enriched event logs that include situational contexts, combined with an AI explainability framework to produce hypothesis-oriented, situationally aware explanations of process outcomes, enabling a comprehensive understanding of decision-making processes by identifying key factors and their strengths.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional AI explainability techniques are used to generate explanations for process outcomes, then the explanation generation process is simple and fast, but the explanations are incomplete and inadequate because they cannot account for situational conditions
Solution Approach 1:
The system segments the explanation generation process into two distinct components: a CEP engine that extracts situational context from event streams, and an AI explainability framework that generates explanations. This segmentation allows each component to specialize - the CEP engine handles complex situational awareness while the AI framework focuses on explanation generation, thereby improving completeness without overwhelming either component.
Solution Approach 2:
The CEP engine acts as an intermediary between the raw event streams and the AI explainability framework. It processes and enriches event data with situational context before passing it to the AI framework, thereby bridging the gap between complex raw data and the need for comprehensive yet manageable explanations.
2Measurement precision
If the system incorporates situational awareness to provide complete explanations, then explanation accuracy improves, but the system complexity increases due to the need for CEP engine and event pattern specifications
Solution Approach 1:
The CEP engine serves multiple functions: it monitors process executions, detects situational contexts, enriches event logs, and provides this enriched information to the AI explainability framework. This multi-functionality reduces the need for separate specialized components, thereby improving accuracy while managing overall system complexity.
Solution Approach 2:
The system performs preliminary action by using the CEP engine to pre-process and enrich event streams with situational context before the AI explainability framework generates explanations. This preliminary enrichment ensures that the AI framework receives comprehensive data, improving explanation accuracy without requiring the framework itself to handle the complexity of situational analysis.
3Reliability
If conventional AI systems process only local process instances, then processing is efficient and fast, but explanations are incorrect or incomplete because they lack contextual awareness
Solution Approach 1:
The CEP engine continuously monitors and processes event streams in real-time, maintaining a continuous awareness of situational contexts. This continuous processing ensures that explanations are always based on the most current and complete information, improving reliability while maintaining processing efficiency through streamlined event-driven architecture.
Solution Approach 2:
The system replaces traditional mechanical processing of isolated process instances with an event-driven architecture where the CEP engine continuously processes event streams. This substitution allows the system to maintain high processing speeds while achieving comprehensive contextual awareness, as the event-driven approach naturally handles parallel processing and real-time data flow.
Data Source
AI summary
Providing a machine-generated explanation of an outcome of a process execution engine includes inputting into a complex event processing (CEP) engine at least one predetermined event pattern specification and a sequence of events, the sequence of events generated, at least in part, by a process execution engine executing a predetermined process in real time. Using the CEP engine, an enriched event log is generated based on the input. The enriched event log includes the sequence of events and additionally includes one or more situational events corresponding to one or more of the sequence of events. The one or more situational events are derived by the CEP engine based on the at least one predetermined event pattern specification. A hypothesis-oriented, situationally aware explanation for the outcome of the predetermined process is determined using an artificial intelligence (AI) explainability framework with input of the enriched event log.