Searching through the internet for a particular topic is one of the most common thing we do now a days. Getting information on anything and everything is just a click away. But the information that is retrieved by the search engine is not always exactly what we need. Then we go through those small paragraphs given under each link to know what the webpage is actually about. These tiny text paragraphs are summaries of the actual article. Internet is comprised of unlimited number of webpages, articles, news, researches, blogs and other information and it is definitely not possible to manually create the summary of each article. Every minute, internet is loaded with so much of new information. One of the most common example is to create a concise summary of a long news article, but there are many more cases of such text summaries that we may come across every day and we may need too.
Search engines like Google, Yahoo and Bing use automatic text summarizing tools to create summaries for all the long text documents. Basically, a summarizer is an algorithm that extracts sentences from a text document, determines which are most important, and returns these sentences in a readable and structured way but in a shorter text and automatic text summarization is part of the field of natural language processing, in which computers can analyze, understand, and derive meaning from human language.
Automatic Summarization tools have two main approaches to summarizing text documents; which are:
- Extractive Method
2. Abstractive Method
The dimensions of text summarization is categorized based on its input type like if it is a single or multi document, purpose like generic, domain specific, or query-based and output type means extractive or abstractive.
Extractive text summarization selects phrases and sentences from the original source document to create the new summary. It involves varied techniques from ranking the relevance of phrases in order to choose only those most relevant to the meaning of the source.
Abstractive text summarization generates entirely new phrases and sentences to capture the meaning of the source document. It is a more challenging method of summarizing and provides more realistic results because it is basically the approach ultimately used by humans. This method operates by selecting and compressing content from the source document but may contain words which are not present in the original document.
Though Extractive summarization methods are more successful and commonly used due to its easier approach and availability but abstractive methods are considered having more general solutions to the abstraction problem.