Reasoning in Resource-Constrained Environments: a Matchmaking Engine over Relational Knowledge Bases

We present a framework for logic-based matchmaking on ALN ABoxes stored in a relational database. The proposed approach allows both non-standard reasoning and subsumption check be performed only via standard SQL queries. Main contribution is in the SQL implementation of the following features: (i) compliance with four match classes (i.e., exact, full, partial and potential); (ii) rank computation for each matching outcome and (iii) preferences management in the user query. Performance evaluation carried out on a PostgreSQL 8.4 engine reports reasonable results in terms of scalability and turnaround times for large scale data sets.

• Type: Journal
• Year: 2011
• Where: DLOG
• Authors: Eufemia Tinelli, Francesco M. Donini, Michele Ruta, Eugenio Di Sciascio
• Keywords: Automated Reasoning, DLOG 2011, Query Performance, Sql Implementation, Turnaround Times