The energetic growth of the Internet produces a remarkable amount of online information and documents. This huge availability of documents has desired tremendous research in text generation. Automatic text summarization has recently gained more attention and has been exploited in various domains.
The significant advantages of text summarization are time-saving, scalable, and widely applied. Text summarization is a significant natural language processing task.
Some text summarization tasks are Tourism text summarization, Legal Document text summarization, Story text summarization, Medical Documents text summarization, Sentiment Summarization, Email Summarization, Tweet Summarization, Books Summarization Novel Summarization, Biomedical Documents text summarization, and News summarization.
Deep learning models are employed for automatic text summarization, and some of the deep neural networks are Graph Neural Networks, Convolutional Neural Networks, and Recurrent Neural Networks. Test summarization applications are Entity timelines, Storylines of events, Sentence compression, Event understanding, Summarization of user-generated content, and Clinical text summarization.
Various surveys and reviews have been investigated, describing text summarization approaches, deep learning architectures, future developments, different applications, benchmark datasets, and evaluation methods.