Littera Deusto

Modern Languages, Basque Studies and Humanities

Questionnaire #2: Research Topics. Automatic Summarization.

abril 22nd, 2009 · No hay Comentarios

Automatic SummarizationAutomatic Summarization is a technique which has a huge importance nowadays, but what is it exactly?
It is a process where you can summarize any type of text by just using a computer program. Your own computer extracts the content from the text, and presents the most important points of it, so you don’t have to do anything for yourself.

We have to remember that summaries are really important in our everyday life, and we don’t even notice it. We just have to think about the newspaper headlines, or about articles about speeches. They are all summarized so we don’t have so much information to read. Now, it’s possible to summarize all these things just by a program, and, contrary to what people may think, it does make sense.

 

You may also think that it’s difficult to find a program for this. But it’s really not. Let’s just think about Microsoft Word, a program which everyone with a computer has used at least once in their lives. Did you know that we can’t find a summarizer in it? Just press “Tools > Summary” and it will make a summary for you (”Herramientas > Autorresumen” in Spanish).

But, how does a program summarize a text? Well, there are two different methods for this:

  • Extraction: It uses keywords and key phrases to do the summary. We could consider it as the low-cost method because it doesn’t really summarize the way a person would, but it’s efficient anyway.
  • Abstraction: More similar to what a person would write. The objective is to make a text that makes sense, maintaining as much as possible the essence of the original text, so it’s usual to see paraphrased sentences in the summary.

All this methods, however, are still being developed so we will still have to wait a bit to see how good this programs can become.

 

Resources:

Etiquetas:

  • Etiquetas