چکیده (انگلیسی):

The areas of information retrieval (IR) and information extraction (IE) are the subject of active research for several years in the community of Artificial Intelligence and Text Mining. With the appearance of large textual corpora in the recent years, we felt the need to integrate modules for information extraction in the existing information retrieval systems. The processing of large textual corpora leads needs that are situated at the border of information extraction and information retrieval areas. Our work in this paper, focus on the extraction of the surface information, i.e. information that not requires complex linguistic processing to be categorized. The goal is to detect and extract passages or sequences of words containing relevant information from the prophetic narrations texts. We propose Finite state transducers-based system that solves successively the problem of texts comprehension. Experimental evaluation results demonstrated that our approach is feasible. Our system achieved encouraging precision and recall rates, the overall precision and recall are 71% and 39% respectively.