A Recipe for Success in Flavor Graph, the Neo4j Heroku Challenge Winner
The content of this article was originally written by Andreas Kollegger on the Neo4j blog.
Flavorwocky, an amusing
name for a clever idea that highlights the connectedness of everyday
life. With a little bit of coding, the simple idea about related flavors
became Luanne Misquitta's winning entry for the Neo4j Heroku Challenge. What was her recipe for success?
It Starts With an Idea
"The idea came to me while I was doing nothing in particular," Luanne explains in her blog,
"to build a simple app that would help one find ingredients whose
flavors complement one another." Foodies everywhere will recognize the
classic "pairing of flavors" whether wine with an entree, or
cross-cultural fusion recipes. Luanne focused on the generalized notion
for ingredients, because "knowing which ingredients have flavor
affinities can produce some amazing new dishes."
Why Flavorwocky? Luanne had asked her husband, "what should I call the app?", and he suggested "jabberwocky." Appreciating the playfulness, she adapted it to the domain, exclaiming "behold — Flavorwocky."

To the Whiteboard!
Luanne elaborated on the idea, realizing that "a graph is a good fit for
this domain, so, using Neo4j, we model the way ingredients in a great
dish pair together based on their flavor affinity." Sketching out the
domain was easy, translating directly into the graph. There are
high-level Categories for groups like vegetables, spices, and dairy.
Ingredients are related to the category first, then to each through a
weighted PAIRS_WITH relationship. For convenience (anticipating the fun
visualization) the Categories also have a color property.
The result, is a "version of this app [which] captures flavor pairings and how well ingredients pair together, and then allows one to search by ingredient and view complementary combinations. Needless to say, this just scratches the surface of the possible features for an app like this."






