Yapay Zeka Hızlı Bakış & Çıkarımlar
AI InsightFinally, we discuss the need for further research.
📌 3 Maddede Bu Makale
Abstract
The uni-int decision-making method constructed by and-product/or-product was defined and applied to a decision-making problem by Çağman and Enginoğlu. The method has a potential for applications in several areas such as machine learning and image processing. Recently, this method has been configured by Enginoğlu and Memiş via fuzzy parameterized fuzzy soft matrices (fpfs-matrices), faithfully to the original, because a more general form is needed for the method in the event that the parameters or objects have uncertainties. However, in the case that a large amount of data is processed, the method has a disadvantage regarding time and complexity. To deal with this problem and to be able to use this configured method denoted by CE10 effectively, we suggest two algorithms in this paper, i.e. EMO18a and EMO18o, and prove that CE10 constructed by and-product (CE10a) and constructed by or-product (CE10o) are special cases of EMO18a and EMO18o, respectively, if first rows of the fpfs-matrices are binary. We then compare the running times of these algorithms. The results show that EMO18a and EMO18o outperform CE10a and CE10o, respectively. Particularly in problems containing a large amount of parameters, EMO18a and EMO18o offer up to 99.9966% and 99.9965% of time advantage, respectively. Afterwards, we apply EMO18o to a performance-based value assignment to the methods used in noise removal, so that we can order them in terms of performance. Finally, we discuss the need for further research.