Automatic Text Summarization of News Articles in Serbian Language

Text Summarization is a technique of creating short, accurate, and fluent summaries of longer text documents or sets of documents. With growing amount of textual data circulating in the digital space, there is a need to develop machine learning algorithms that can automatically shorten longer texts and deliver accurate summaries. First, the generated summaries should fluently pass the intended messages. Second, the generated summaries should reduce reading time and speed up the process of researching for relevant information. By large majority, most of research in text summarization has been done for English texts. In this paper, we experiment with a variation on a graph-based method called TextRank for extractive text summarization. The dataset used in this study is collected from one of the online news sources in Bosnia and Herzegovina.