TEACHING OLD CALIPERS NEW TRICKS: USING CRANIOMETRICS FOR ANCESTRY ADMIXTURE ESTIMATION VIA FUZZY MATH

<p>Cranial measurements have been a cornerstone of physical anthropology since its formation as a discipline in the early 1900s. However, most other ancestry determination methods come with a significant epistemological issue: they differentiate individuals into discrete categories without accounting for the issue of admixture. Advances in data mining and analysis techniques can now be used to help resolve this issue through soft computing, also known as “fuzzy math”. This type of advanced computational math requires specialized knowledge in computer programming, statistics, and data analysis techniques unless one is using computer programs specially designed to run these analyses.</p> <p>This project compiled a database from multiple open-source craniometrics data and utilized prepared packages within the R statistical environment to find a valid soft computing method for fuzzy ancestry determination that does not require extensive knowledge in computer programming or data mining. Exploration of database demographics notes an excess of White-identified individuals, and when tested, this demographic skew impacts the ability of the given package to return valid results. The package chosen was valid using the compiled database. Exploration of causes for the invalid results, including a significant White skew in the underlying database due to accessibility of metric databases, overfitting, and the inherent issues of admixture on craniometric research, are explored, and future directions discussed.</p>

Name thesis:degree_name

Master of Arts (MA)

Level thesis:degree_level

Masters Thesis

Year dc:date.available

2022

Author dc:creator

• Carnahan, Kristi

Contributors dc:contributor

• Dr. Marie Danforth
• Dr. Bridget Hayden
• Dr. Sharon Young

• craniometrics
• ancestry estimation
• anthropology
• biogeographic estimation
• soft computing in anthropology
• Biological and Physical Anthropology

Repository record dc:identifier

https://aquila.usm.edu/masters_theses/881

OAI identifier oai:identifier

oai:aquila.usm.edu:masters_theses-1944