More than Just Another Number: Use of the Smithsonian Trinomial System and the Digital Index of North American Archaeology (DINAA) to Link Open Information about Archaeological Sites Across the Web
Archaeological sites in the United States are often associated with alphanumerical identifiers known as Smithsonian trinomial numbers (STNs). Developed in the mid-Twentieth Century, STNs consist of patterned alphanumeric sequences, potentially recognizable in spreadsheets, archival records, and research literature. The Digital Index of North American Archaeology (DINAA), a linked open data hub for archaeological site information, is attempting "named entity recognition" (a form of text mining) to identify STNs in archaeological literature. The grand challenge undertaken here will ultimately allow a DINAA user to browse, via map-based visualizations, cross-references between DINAA site file information and publications containing related STNs. This initiative includes human-guided cataloging of STNs from recent American Antiquity offerings, and software-enabled text mining of STNs from the entire run of American Antiquity through the JSTOR for Research program. Challenges include: variation in STN formatting, inclusion of STNs in various elements of research articles (abstract, body text, tables, figures, notes, etc.), and publication of competing identification strategies (e.g. state and museum systems). We will present results of this pilot study to inform future linked open data use of STNs in other archaeological literature, governmental records (e.g. Federal Register), museum collections, and archives.
Cite this Record
More than Just Another Number: Use of the Smithsonian Trinomial System and the Digital Index of North American Archaeology (DINAA) to Link Open Information about Archaeological Sites Across the Web. Taylor Wiley, Joshua Wells, Eric Kansa, Patrick Finnegan, R. Carl DeMuth. Presented at The 81st Annual Meeting of the Society for American Archaeology, Vancouver, British Columbia. 2017 ( tDAR id: 432136)
This Resource is Part of the Following Collections
Abstract Id(s): 16859