PEMILAHAN ARTIKEL BERITA DENGAN TEXT MINING

Abstract

With the rapid development of Internet, more and more also emerging sites or blogs that provide a wide range of online news articles. An article, before it can be published, originally sent by the reporter to editor to be sorted. Sorting type of news is relatively easily done by humans, but if the case was brought to a level of segregation in automation with computers will bring its own problems, although for a shorter story. Text mining is one way that is expected to solve the above problems. With text mining, can be searched words that can represent the content of news articles, then its category is determined based on the frequency of words contained in it. Stage by the author on the study are: (i) development of a database for the keyword vector, (ii) sorting of news sources based on the database of step (i). This paper is expected to help the electronic editorial system to be able to sort or find out the category of a news article without the need of an editor that saves time and cost of doing business on the model of an electronic news service on-line internet based.