17 Facts About Automatic summarization

Automatic summarization is the process of shortening a set of data computationally, to create a subset that represents the most important or relevant information within the original content.

Text summarization finds the most informative sentences in a document; various methods of image summarization are the subject of ongoing research, with some looking to display the most representative images from a given collection or generating a video; video summarization extracts the most important frames from the video content.

Abstractive Automatic summarization methods generate new text that did not exist in the original text.

The second is query relevant Automatic summarization, sometimes called query-based Automatic summarization, which summarizes objects specific to a query.

An example of a summarization problem is document summarization, which attempts to automatically produce an abstract from a given document.

Related searches

Image collection summarization is another application example of automatic summarization.

Video summarization is a related domain, where the system automatically creates a trailer of a long video.

At a very high level, Automatic summarization algorithms try to find subsets of objects, which cover information of the entire set.

Some techniques and algorithms which naturally model Automatic summarization problems are TextRank and PageRank, Submodular set function, Determinantal point process, maximal marginal relevance etc.

The main difficulty in supervised extractive Automatic summarization is that the known summaries must be manually created by extracting sentences so the sentences in an original training document can be labeled as "in summary" or "not in summary".

Unsupervised approach to Automatic summarization is quite similar in spirit to unsupervised keyphrase extraction and gets around the issue of costly training data.

Some unsupervised Automatic summarization approaches are based on finding a "centroid" sentence, which is the mean word vector of all the sentences in the document.

In such a way, multi-document Automatic summarization systems are complementing the news aggregators performing the next step down the road of coping with information overload.

Multi-document Automatic summarization creates information reports that are both concise and comprehensive.

State of the art results for multi-document Automatic summarization are obtained using mixtures of submodular functions.

New method for multi-lingual multi-document Automatic summarization that avoids redundancy works by simplifying and generating ideograms that represent the meaning of each sentence in each document and then evaluates similarity "qualitatively" by comparing the shape and position of said ideograms has recently been developed.

Pattern-based Automatic summarization was the most powerful option for multi-document Automatic summarization found by 2016.