A Hybrid Diagnosis Approach Combining Black-Box and White-Box Reasoning
- Cite this paper as:
- Chen M., Yu S., Franz N., Bowers S., Ludäscher B. (2014) A Hybrid Diagnosis Approach Combining Black-Box and White-Box Reasoning. In: Bikakis A., Fodor P., Roman D. (eds) Rules on the Web. From Theory to Applications. RuleML 2014. Lecture Notes in Computer Science, vol 8620. Springer, Cham
We study model-based diagnosis and propose a new approach of hybrid diagnosis combining black-box and white-box reasoning. We implemented and compared different diagnosis approaches including the standard hitting set algorithm and new approaches using answer set programming engines (DLV, Potassco) in the application of Euler/X toolkit, a logic-based toolkit for alignment of multiple biological taxonomies. Our benchmarks show that the new hybrid diagnosis approach runs about twice fast as the black-box diagnosis approach of the hitting set algorithm.
Unable to display preview. Download preview PDF.