Summaries and the Process of Summarization from Evaluation of Automatic Text Summarization -a Practical Implementation

Abstract

Text summarization (or rather, automatic text summarization) is the technique where a computer automatically creates an abstract, or summary, of one or more texts. The initial interest in automatic shortening of texts was spawned during the sixties in American research libraries. A large amount of scientific papers and books were to be digitally stored and made searchable. However, the storage capacity was very limited and full papers and books could not be fit into databases those days. Therefore summaries were stored, indexed and made searchable. Sometimes the papers or books already had summaries attached to them, but in cases were no readymade summary was available one had to be created. Thus, the technique has been developed for many years (see Luhn 1958, Edmundson 1969, Salton 1988) and in recent years, with the increased use of the Internet, there have been an awakening interest for summarization techniques. Today the situation is quite the opposite from the situation in the sixties. Today storage is cheap and seemingly limitless. Digitally stored information is available in abundance and in a myriad of forms to an extent as to making it near impossible to manually search, sift and choose which information one should incorporate. This information must instead be filtered and extracted in order to avoiding drowning in it.