Word2vec explained.

Original paper T.Mikolov, K.Chen, G.Corrado, J.Dean Efficient Estimation of Word Representations in Vector Space
word_vectors.most_similar(positive=['Paris', 'Spain'], negative=['France'])
[('Madrid', 0.8625079989433289),
('Barcelona', 0.7637037634849548),
('Sevilla', 0.6874054074287415),
('Seville', 0.6747831106185913),
('Malaga', 0.6494932174682617),
('Zaragoza', 0.6459373831748962),
('Valencia', 0.6383104920387268),
('Alicante', 0.6115807890892029),
('Salamanca', 0.6041630506515503),
('Murcia', 0.6019026041030884)]
Source: Word2vec
Thanks to Jiřà Materna for Machine Learning College in Prague. Where you can find much more handful notebooks to understand the basics of machine learning.