@revista_internacional{891, keywords = {FDOSM, Likert scale, multicriteria decision-making, Pythagorean fuzzy sets, Sign language recognition systems}, author = {Sarah Qahtan and Hassan A. Alsattar and Aws Alaa Zaidan and Muhammet Deveci and Dragan Pamucar and Luis Martínez}, title = {A comparative study of evaluating and benchmarking sign language recognition system-based wearable sensory devices using a single fuzzy set}, abstract = {Recent research has focused on developing real-time sign language recognition systems (SLRSs) based on gesture recognition to classify hand motions into their equivalent meaning in spoken language, but no comprehensive system with all desirable features has been presented. The existence of different systems has hindered the process of selecting the most preferred system. Therefore, many researchers have compared and evaluated several recognition systems to identify the best one using multicriteria decision-making methods. These studies extended the fuzzy decision by opinion score method (FDOSM) using a single Likert scale under the Pythagorean fuzzy set (PFS) or one of its extensions. However, no comparative study has examined the influence of using multiple Likert scales with a single fuzzy set. Furthermore, the effect of employing multiple Likert scales on benchmarking results is a challenging task. Therefore, this paper examines the three Likert scales (five-, seven- and ten-point) under the same fuzzy environment. This paper extends FDOSM into PFSs based on the power Bonferroni mean (PBM) operator (named PFDOSM-PBM) to benchmark the real-time SLRS. The decision matrix is constructed based on 30 real-time SLRS-based wearable sensory devices and the 11 evaluation criteria. The results reveal that the five-point Likert scale is superior to other scales (i.e., seven- and ten-point) as it is flexible, easy to use and generates more accurate findings on the basis of uncertainty compared to other scales. Systematic ranking and comparative analysis are conducted to validate and evaluate the proposed method.}, year = {2023}, journal = {Knowledge-Based Systems}, volume = {269}, pages = {110519}, issn = {0950-7051}, url = {https://www.sciencedirect.com/science/article/pii/S0950705123002691}, doi = {https://doi.org/10.1016/j.knosys.2023.110519}, }