INNOVATION
Ask me to Share a Cherokee RecipeOur Indigenous Knowledge Graph (IKG) serves as a demo knowledge engine for traditional food recipes and culinary wisdom of indigenous peoples.
The story of our IKG is shared via Sina, our Conversational AI. You can ask Sina about time-tested recipes from ancient cultures that serve as a bridge between the living past and the present. Go ahead, ask Sina to share a Cherokee recipe or one from Samoa. You can interact with Sina on Google Assistant by using the invocation phrase “OK Google: Talk to Sina Storyteller.” With IKG, our goal is to show how creating structure, or knowledge engines, around our stories can foster reasoning and cultural intelligence in machines and conversational AIs. In order to build and support intelligent machines that are more culturally-aware, we need to provide a source of data to develop, train, and test with. |
"We need a global-denominator of perspective that becomes more inclusive and valuable with each and every contribution. The way tribal people think globally is muted, even though our first-voices are indistinguishable from the natural order of the earth.”
Tracy Monteith, Senior Software Engineer at Microsoft, and member of the Eastern Band Cherokee
WHY TELL STORIES AROUND TRADITIONAL RECIPES?
Bringing Cultural Context to Artificial Intelligence
As the storytelling species, human beings transmit knowledge to succeeding generations through stories. A familiar example of this is through the sharing of traditional recipes handed down from generation to generation. These recipes are often accompanied by mythical stories that in many cultures served to preserve important knowledge about the cultivation, harvest and preparation of traditional and indigenous ingredients.
As the storytelling species, human beings transmit knowledge to succeeding generations through stories. A familiar example of this is through the sharing of traditional recipes handed down from generation to generation. These recipes are often accompanied by mythical stories that in many cultures served to preserve important knowledge about the cultivation, harvest and preparation of traditional and indigenous ingredients.
HOW DOES IT WORK?
By using the key elements (ingredients, method, story) we built the foundations of an Indigenous Knowledge Graph (IKG) using food wisdom from Timor-Leste and Native American culture, personally shared by the IKG team. Our aim was to show how creating structure around our stories can foster reasoning and cultural intelligence in machines and conversational AIs.
Click to enlarge
|
Indigenous Knowledge Graph | Neo4j
In our demo, we include the sustainable development goals (SDGs) that are most relevant to each culture and recipe, and these end up acting as an ontology -- a way to sort information into logical hierarchical relationships. Pictured is the IKG created by Wolfgang Victor Yarlott using Neo4j (a graph database platform) which shows the relationship between the SDGs, recipes and ingredients, as well as the corresponding connections between cultural origins and traditional narrative. Combined as a whole these elements are what make up each story. |
The New AI Frontier: Tagging Indigenous Values
READ THE WHITE PAPER
Read the interactive whitepaper on this project on Issuu or download your own PDF: