Multi -Perspectives evaluation and benchmarking of real-time sign language recognition systems based on fuzzy multi-criteria decision analysis
<p>The real-time Sign Language Recognition Systems (SLRSs) have been developed</p><p>recently to assist the deaf and dumb community in translating hand gestures to their</p><p>spoken language equivalents. However, the multidimensi...
Saved in:
| Main Author: | |
|---|---|
| Format: | thesis |
| Language: | eng |
| Published: |
2022
|
| Subjects: | |
| Online Access: | https://ir.upsi.edu.my/detailsg.php?det=9796 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | <p>The real-time Sign Language Recognition Systems (SLRSs) have been developed</p><p>recently to assist the deaf and dumb community in translating hand gestures to their</p><p>spoken language equivalents. However, the multidimensional evaluation and</p><p>benchmarking of these systems considered a Multi-attribute Decision-Making</p><p>(MADM) problem due to the presence of several issues, including multiple evaluation</p><p>criteria, multi ortance, and criteria confliction. In this study, a new extension of the</p><p>Fuzzy Decision by Opinion Score Method (FDOSM) for evaluating and benchmarking</p><p>SLRSs is developed under a Pythagorean Fuzzy Set (PFS). Fundamentally, the</p><p>methodology divided into 4 phases. The first phase is the preliminary study, while the</p><p>construction of the decision matrix is the second phase, then the third phase is the</p><p>formulation of the proposed methods, and the fourth phase is the results evaluation.</p><p>Results indicate the following: (1) individual benchmarking results of real-time SLRS</p><p>showed high variation based on the preference of each Decision Maker (DM). (2) The</p><p>group benchmarking results for Pythagorean Fuzzy Decision by Opinion Score Method</p><p>- Interactive Hybrid Arithmetic Mean PFDOSM-IHAM indicate that the 29th real-time</p><p>SLRS was the best, whereas the worst real-time SLRS was attributed to SLRS (6th).</p><p>While the results of group benchmarking for Interval-Valued Pythagorean Fuzzy</p><p>Decision by Opinion Score Method IVP-FDOSM reveal that the 10th real-time SLRS</p><p>was the optimal one and the 6th was the worst. In addition, the rates of ranking match</p><p>between the group benchmarking and each DM captured and discussed from analytical</p><p>perspective. (3) for the results evaluation, two MADM assessments, namely, systematic</p><p>ranking and comparative analysis are used to validate the robustness of the proposed</p><p>MADM methods. The research contributed to the deaf mute community by providing</p><p>the suitable SLRS selection bases on their life needs, benefiting the SLRS industrial</p><p>field, and the special education centers.</p> |
|---|