Soft Set Decision/Forecasting System Based on Hybrid Parameter Reduction Algorithm

Abstract

Existing classification techniques, which are previously proposed for eliminating data inconsistency, could not achieve an efficient parameter reduction in soft set theory as it affects the obtained decisions. Additionally, data decomposition based on previous algorithms could not achieve better parameter reduction with available domain space. Meanwhile, the computational cost made during the combination generation of datasets can cause machine infinite state as Nondeterministic Polynomial time (NP). Although the decomposition scenario in the previous algorithms detects the reduction, it could not obtain the optimal decision. The contributions of this study are mainly focused on minimizing choices costs through adjusting the original classifications by decision partition order and enhancing the probability of search domain by a developed HPC algorithm. The results show that the decision partition order technique performs better in parameter reduction up to 50%, while other algorithms could not obtain any reduction in some scenarios.