Context-Based Arabic Text Processing System

In the last decades, the rapid growth of information increased the need for summarizing systems capable of processing large amounts of electronic texts. Implementing a system that allows a user to search a collection of documents, process them and extract relevant information is a non-trivial task especially if the system deals with dynamic and heterogeneous texts on the Web. This paper describes a system for summarizing domain-free texts written in Arabic. The idea is aiming at getting an idea of the text content without understanding its meaning. We are interested in discursive markers, which can give an idea of the sentences relevance