پیوندهای مفید
اخبار و اعلانات
آمار
| تعداد نشریات | 44 |
| تعداد شماره‌ها | 1,919 |
| تعداد مقالات | 15,607 |
| تعداد مشاهده مقاله | 47,422,820 |
| تعداد دریافت فایل اصل مقاله | 19,103,641 |
Comparative analysis of size removal methods in traditional morphometrics: A case study with Caspian bighead goby (Ponticola gorlap, Gobiidae, Iljin, 1949) | ||
| Taxonomy and Biosystematics | ||
| مقالات آماده انتشار، پذیرفته شده، انتشار آنلاین از تاریخ 20 آبان 1405 | ||
| نوع مقاله: Original Article | ||
| شناسه دیجیتال (DOI): 10.22108/tbj.2026.149515.1341 | ||
| نویسندگان | ||
| Ehsan Shatoei Gharenjeh؛ Soheil Eagderi* ؛ Hadi Poorbagher | ||
| Department of Fisheries, Faculty of Natural Resources, University College of Agriculture and Natural Resources, University of Tehran, Karaj, Iran. | ||
| چکیده | ||
| Size removal is a critical step in morphometric analysis, as it isolates shape variation from size-related differences among specimens. However, the choice of size-removal method can substantially influence downstream analyses. In this study, we compared the performance of three size-removal methods - Isometric Burnaby's method, Allometric Burnaby's method, and the Allometric versus standard method - using a Random Forest classifier to evaluate residual size-class predictability after size correction. The methods were applied to a dataset of 50 specimens of the Caspian bighead goby, Ponticola gorlap, collected from five river stations in Guilan Province, northern Iran (n = 10 per station): the Loundvil, Pishkhan, Sefidrud, Shalmanrud, and Kelachay Rivers. Model performance was assessed using accuracy and Cohen's kappa on both training and testing datasets. For the testing datasets, sensitivity, specificity, and balanced accuracy were also considered. Isometric Burnaby's method yielded the lowest mean kappa (0.160) and balanced accuracy (0.549), compared with 0.2336 and 0.584 for Allometric Burnaby's method and 0.222 and 0.601 for the Allometric-versus-standard method, respectively. These results indicate that Isometric Burnaby's method retained less residual size-related information in the present dataset. Because the differences among methods were evaluated descriptively rather than through formal statistical comparison, the findings should not be interpreted as evidence of statistically significant superiority. The results support the potential use of Random Forest-based size-class predictability as a criterion for evaluating size-removal methods in traditional morphometrics. Validation across independent datasets and alternative machine-learning algorithms is needed to assess the framework's consistency and generalizability. | ||
| کلیدواژه‌ها | ||
| Caspian؛ Gobiidae؛ Morphology؛ Random Forest algorithm؛ Shape | ||
|
آمار تعداد مشاهده مقاله: 1 |
||