Reactive hybrid knowledge bases

• Author(s): Cruz, Nuno Alexandre Neves
• Date: 2014
• Persistent ID: http://hdl.handle.net/10362/14058
• Origin: Repositório Institucional da UNL
• Subject(s): Rules; Ontologies; Action rules; Semantic web
• Document Type: Master thesis
• Language: English
• Advisor(s): Damásio, Carlos; Knorr, Matthias
• Contributor(s): RUN

Description

Hybrid knowledge bases are knowledge bases that combine ontologies with non-monotonic rules, allowing to join the best of both open world ontologies and close world rules. Ontologies shape a good mechanism to share knowledge on theWeb that can be understood by both humans and machines, on the other hand rules can be used, e.g., to encode legal laws or to do a mapping between sources of information. Taking into account the dynamics present today on the Web, it is important for these hybrid knowledge bases to capture all these dynamics and thus adapt themselves. To achieve that, it is necessary to create mechanisms capable of monitoring the information flow present on theWeb. Up to today, there are no such mechanisms that allow for monitoring events and performing modifications of hybrid knowledge bases autonomously. The goal of this thesis is then to create a system that combine these hybrid knowledge bases with reactive rules, aiming to monitor events and perform actions over a knowledge base. To achieve this goal, a reactive system for the SemanticWeb is be developed in a logic-programming based approach accompanied with a language for heterogeneous rule base evolution having as its basis RIF Production Rule Dialect, which is a standard for exchanging rules over theWeb.