In the past couple of years, neural networks have nearly taken over the field of NLP, as they are being used in recent state-of-the-art systems for many tasks. One interesting application is distributional semantics, as they can be used to learn intelligent dense vector representations for words. Marco Baroni, Georgiana Dinu and German Kruszewski presented a paper in ACL 2014 called “Don’t count, predict! A systematic comparison of context-counting vs. context-predicting semantic vectors“, where they compare these new neural-network models with more traditional context vectors on a range of different tasks. Here I will try to give an overview and a summary of their work.

## Distributional hypothesis

The goal is to find how similar two words are to each other semantically. The distributional hypothesis states:

*Words which are similar in meaning occur in similar contexts*

(Rubenstein & Goodenough, 1965).

Therefore, if we want to find a word similar to “magazine”, we can look for words that occur in similar contexts, such as “newspaper”.

I was reading a magazine today |
I was reading a newspaper today |

The magazine published an article |
The newspaper published an article |

He buys this magazine every day |
He buys this newspaper every day |

Also, if we want to find how similar “magazine” and “newspaper” are, we can compare how similar are all the contexts in which they appear. For example, to find the similarity between two words, we can represent the contexts as feature vectors and calculate the cosine similarity between their corresponding vectors.