I participated in DBpedia’s Autumn 2020 Hackathon. I joined as part of the Diffbot track, which granted access to Diffbot’s knowledge graph and its natural language api.

For my submission, I scrapped together an app called KG Shiritori. This is a play on an age-old Japanese word game, where players take turns saying words that match the last letter of the previous word (e.g., dog->gate->egg->…).

To spin this game to work on the kind of information available in a knowledge graph, in KG Shiritori you make connections between entities using common facts about them. For example, “Barack Obama” and “George W. Bush” can be connected by the fact that they had the employment title “president”.

Diffbot’s knowledge graph was used to identify entities as well as a large collection of “facts” about them. I also used Diffbot’s natural language API to parse input sentences to allow users to play the game.

My implementation is open source, but it requires an access token to Diffbot’s APIs so it unfortunately isn’t available to try for yourself. You can watch a sloppy demo video of me walking through the app here.