Ethnographic Semantic Data Modeling (EKSDM) is a methodology that combines the practice of ethnography with semantic data modeling techniques to create structured, queryable representations of qualitative research. The approach merges ethnographic practices with domain analysis, concept mapping, social network analysis, text analysis, ontology engineering, and knowledge management to build systems of analysis that embody greater context and explainability than conventional databases or spreadsheets can provide.
How it works
EKSDM follows a structured process: deconstruct ethnographic data, identify domains, develop a glossary of terms, define relationships between concepts, and add structured data and properties to a graph database. The result is a knowledge graph that captures the unstructured and implicit knowledge typically buried within qualitative research, making it queryable, shareable, and extensible. Each step translates the richness of ethnographic fieldwork into a formal representation without reducing it to simple categories or keyword tags.
Relationship to the Anthropological Knowledge Graph
EKSDM is the construction methodology for the Anthropological Knowledge Graph (AKG). Where the AKG is the product, a structured graph of ethnographic knowledge, EKSDM is the process for building it. The methodology requires skills from both anthropology and information science, combining ethnographic sensibility with the technical architecture of semantic data systems.
Origin
Matt Artz introduced EKSDM in the UNESCO paper From Machine Learning to Machine Knowing (2023). The methodology is a core component of his work in AI Anthropology and provides the bridge between ethnographic practice and the structured knowledge systems that enable machine knowing.