A popular thing done these days is surfing through the internet for a particular topic. Retrieving information on anything and everything is just right there. The information gotten isn’t always what is required. The next thing we do is glance through the small paragraphs below each link to find out what the webpage is all about. These paragraph summarize the article. The internet consists of various news, articles, research, webpages and blogs and is impossible to have a summary of each article manually created. The net is loaded with so much new data every minute. A typical example is forming a summary of an article though, there are numerous cases of such summary which we may also need.
Search engines such as yahoo, google and Bing utilize automatic text summarizing tools in summarizing all long documents. A summarizer can be defined as an algorithm which forms sentences from a text article, selects what is considered important, and brings them back in a form that is shorter, structured and readable and automatic text summarization is encompassed in the field of natural language processing, in which computers can process and get meaning of human language.
Tools used for Automatic summarization have two major ways of summarizing text documents; they included:
- Attractive method
- Abstractive method
Parts of text summarization is divided based on its type of input that is, either a single or multi documents, type of output such abstractive or extractive and purpose such as domain-specific, queen-based or generic.
In extractive text summarization, sentences and phrases are college from the initial document to form the summary. It utilizes various techniques ranging from placing the importance of phrases in order to pick only the important ones to the meaning of the source.
In Abstractive text summarization new sentences and phrases are formed to understand the meaning of the source document. This technique is more difficult and gives results which are more difficult and it gives results which are more realistic because it is the method used by humans. Its mechanism is by selection and compression contents extract from the source documents but may have additional words not contained in the original document.
The abstractive techniques are seen to have a more general solution to the problem but the Extractive technique are often used and are more successful due to its availability and easier approach.