Word2Vecにおけるマイナスの概念について Word2Vecsimilarity meow-0.018 100 meowsize ## # sourcecode from gensim.models import Word2Vec sentences = [["cat", "say", "meow"],["dog","say","woof"]] model = word2vec.Word2Vec(sentences, min_count=1,size=100) print(model.wv.most_similar(positive=['cat'])) # output [('woof', 0.12968875467777252), ('say', 0.06406527757644653), ('dog', 0.05123840644955635), ('meow', -0.01819145306944847)] ## Example <

API `most_similar()` 2 2