Word2vec captures semantic similarity; this is the most important point that we need to keep in mind when we are processing the answer to the preceding question.
If you have an NLP application in which you want to use the distributional semantic, then word2vec is for you! Some NLP applications will use this concept to generate the features and the output vector from the word2vec model, or similarly, vectors will be used as input features for the ML algorithm.
You should know which NLP applications can use word2vec. If you know the list of applications, it becomes easy for you to decide whether you should use it or not. Suppose you can use k-mean clustering for document classification; if you want document classification to carry some of the attributes of semantics, then you can use word2vec as well. If you want to build a question-answer system, then...
