For word representation baseline people use bag-of-words or word embedding. Here, I want to understand all approaches that can be used for word representations. For example: -Bag-of-words (tfidf, n-gram,etc.) -word embedding (GloVe, word2Vec, skip gram, etc.) etc. Is there any source/article that explains all of the word representation group? Does anyone have a good understanding and explain it here? thanks.
1 Answer
'All' is a very strong quantifier :) very tough requirement to find 'one article to rule them all (word representations)'.
And representation of word can vary depending on the task to be solved.
E.g. a word can be represented also as its morhological tags - part of speech, atributes like (grammatical) number etc. - for the task of parsing.
You may sometimes skip meaning of word and jump directly to meaning of whole text (like in doc2vec).