There are broadly two types of summarization :
In the extractive case, summarization techniques select relevant phrases from the input document and concatenate them to form sentences.
⇒ These techniques are very popular in the industry as they are very easily implemented and reasonably accurate on short text given their use of existing natural
language phrases. Additionaly, since they are unsupervised techniques, they are also ridiculously fast compared to their abstractive counterpart.
On the other side, abstractive summarization techniques generates new phrases through rephrasing or using words that were not in the original text.
⇒ For this reason, the abstractive approach is much harder given that to obtain a perfect abstractive summary, the model has to first truly understand the document
and then try to express that understanding in short possibly using new words and phrases.
