Evolution of Word to Vector

Word embeddings are a type of word representation that allows words with similar meanings to have a similar representation. A word is characterized by the company it keeps -- J.R.Firth (1957) All the NLP applications we build today have a single purpose and that is to make the computers understand human language but the biggest challenge to do that makes the machines understand how we understand human language in the form of reading, writing, or speaking. To start with we first train our machine learning or deep learning algorithms to understand textual data. As machines do not understand the text we need to make the input to a machine-readable format. For example, Imagine I'm trying to describe my dog -- With all of this dog's features and precise description, anyone could draw it, even though we have never seen it.