Szeniczey, Tamás (2020) A nazomaláris szög és szimotikus húr alkalmazása metrikus alapú taxonómiai osztályozási feladatokban = The Utility of Nasomalar Angle and Simotic Chord in Ancestry Estimation from Craniometric Data. ANTHROPOLOGIAI KÖZLEMÉNYEK, 61. pp. 53-62. ISSN 0003-5440 (nyomtatott); 2560-0753 (online)
|
Text
Szeniczey_AK2020.pdf Download (1MB) | Preview |
Abstract
Ancestry can be estimated in a probabilistic framework using facial morphological signatures of the different population histories. While the assessment of qualitative traits requires more experience, the measurement of the most distinctive quantitative characters usually demands anthropological tools seldom available. However, some of the facial measurements related to ancestry can be easily recorded with a caliper. The study aims to present the efficiency of simotic chord (SC) and nasomalar angle (M77) in ancestry estimation by applying supervised and unsupervised algorithms to classify skulls with European and Asian ancestry. Linear discriminant analysis and Gaussian Mixture models were applied to a subset of the Howells’ craniometric dataset comprised of individuals with European and Asian ancestry. Prediction of ancestry was carried out on a set of craniometric traits describing height, length and width parameters of the crania and then repeated on this set supplemented with M77 and SC measurements. An increased percentage of true positive prediction of ancestry was achieved by involving the M77 and SC measurements.
Item Type: | Article |
---|---|
Uncontrolled Keywords: | Ancestry estimation; Nasomalar angle; Simotic chord; Gaussian mixture model; Linear discriminant analysis |
Subjects: | G Geography. Anthropology. Recreation / földrajz, antropológia, kikapcsolódás > GN Anthropology / embertan, fizikai antropológia |
SWORD Depositor: | MTMT SWORD |
Depositing User: | MTMT SWORD |
Date Deposited: | 04 Dec 2020 09:18 |
Last Modified: | 31 Mar 2023 07:18 |
URI: | http://real.mtak.hu/id/eprint/117857 |
Actions (login required)
![]() |
Edit Item |