Using distibuctional semantics (word2vec family algorithms and the CADE framework) to learn word embeddings from the Italian literary corpuses we generated.

Wish you could follow a new sport but don't know enough about it? This application allows you to expand your knowledge by using analogies with sports you are familiar with. These analogies are calculated by aligning word embeddings spaces obtained from different text corpora (one for each sport).

(可以借鉴 user keras) Collaborative Denoising Auto-Encoder (CDAE)