Meta-heuristics for generation of linguistic descriptions of weather data: Experimental comparison of two approaches

In this paper we experimentally assess, from both algorithmic and pragmatic perspectives, the adequacy of linguistic descriptions real data generated by two metaheuristics: simulated annealing genetic algorithm meta-heuristics. The type consider are fuzzy quantified statements (both Zadeh's type-1 type-2) involving three well-known quantification models (Zadeh's scalar Delgado's GD). We conducted an empirical validation using observation prediction meteorological data, where automatic (metrics-based) manual (human experts-based) assessment on was assessed. Results indicate that, overall, approach performs better than in terms quality obtained time execution. Significance outperforming depends model selected. Tests statistical significance point out that for no significant differences exist between meta-heuristics case. For type-2 descriptions, GD types data. (observation or prediction). Globally, over i) is 4 12 scenarios considered (all them type-2), ii) not other 8 type-2). Also human expert conducted, showing behave similarly while algorithms produce more suitable descriptions.