Micro Decision Support Systems
This topic deals with the definition and realization of a new generation of crisp and fuzzy rule engines characterized by a light-weight, object-oriented and update-versatile implementation suitable for resource-limited mass-market mobile devices so that they can be used to build micro crisp or fuzzy DSSs with the aim of facing a set of new challenging scenarios, where information must be used anywhere for supporting the decision-making tasks seamlessly and ubiquitously.
Mobile Fuzzy Inference Engines
We developed a fuzzy inference engine for mobile devices aimed at easing the design process of Fuzzy DSSs by providing the user with a wide and self-contained range of fuzzy connectives, linguistic hedges, membership functions, implication, aggregation and defuzzification methods. The fuzzy inference engine allows to define single-input-single-output (SISO) systems, multi-input-single-output (MISO) systems, and multi-input-multi-output (MIMO) systems and uses a First Infer Then Aggregate (FITA) approach for the inferential process.
Mobile Crisp Rule Engines
We defined and realized an innovative crisp rule engine for mobile devices based on a forward chaining technique that uses a lazy evaluation strategy to reduce both space and time complexity of inference process with the goal of coping the resources constraints of mobile devices. The idea of lazy evaluation is to compute only one rule activation in each cycle based on the observation that only one rule is fired anyway.