Summary: Exploring BERTSUM for Text Summarization

Let's summarize what we have learned so far.

We'll cover the following

Key highlights

Summarized below are the main highlights of what we've learned in this chapter.

  • We started off by understanding what text summarization is. We learned that there are two types of text summarization tasks—one is extractive summarization, and the other is abstractive summarization. In extractive summarization, we create a summary from given text by just extracting only the important sentences. Unlike extractive summarization, in abstractive summarization, we will not create a summary by just extracting important sentences from the given text. Instead, in this type, we create a summary by paraphrasing the given text.

  • We learned how to fine-tune BERT to perform the summarization task. We learned how BERTSUM works and how it is used for summarization tasks. After understanding BERTSUM, we learned how to use BERTSUM with a classifier, with a transformer, and with LSTM for an extractive summarization task.

Get hands-on with 1200+ tech skills courses.