Kaggle Chief Data Scientist, Word2Vec “is a really, really, really big deal”

Agreed. I once used a rich vector space model to answer natural language questions related to geography with a similar approach.

Word2Vec is part of a Kaggle competition to classify sentiment of movie reviews from rotten tomatoes. There seems to be a better method than word2vec from berkeley lab. I found that interesting considering the impact Jeremy Howard said word2vec could have.