domiKnow game data
==================

The files in this folder (vocab.txt, lexicon.txt, phrases.txt, edges.bin,
edge_rels.bin, phrase_edges.bin, phrase_edge_rels.bin, vectors.bin,
families.bin, solver_veto.bin, puzzles.json, meta.json) are adapted from the
datasets below and are licensed under the Creative Commons
Attribution-ShareAlike 4.0 International licence (CC BY-SA 4.0):
https://creativecommons.org/licenses/by-sa/4.0/

Changes made: filtered to English words and short concepts that can be played;
ConceptNet edges kept between those words only (weight 1.0 and up), with the
most specific relation per pair; Numberbatch vectors smoothed with ConceptNet
neighbours, reduced to 96 dimensions (PCA) and quantised to int8; word
frequencies rounded; puzzles and solutions generated from all of the above.

Sources
-------

ConceptNet 5.7 (https://conceptnet.io), CC BY-SA 4.0.
  Robyn Speer, Joshua Chin and Catherine Havasi. "ConceptNet 5.5: An Open
  Multilingual Graph of General Knowledge." AAAI 2017.
  ConceptNet includes material from Wiktionary, WordNet, Open Mind Common
  Sense, OpenCyc, DBPedia and other sources:
  https://github.com/commonsense/conceptnet5/wiki/Copying-and-sharing-ConceptNet

ConceptNet Numberbatch 19.08
  (https://github.com/commonsense/conceptnet-numberbatch), CC BY-SA 4.0.
  Same authors and citation as ConceptNet.

wordfreq (https://github.com/rspeer/wordfreq), by Robyn Speer: word
  frequencies. Its data is CC BY-SA 4.0.

ENABLE word list (public domain).

The game's code is not covered by this licence.