Making sense embedding out of word embeddings using graph-based word sense induction - View it on GitHub