Enhancing Spatial Information Extraction from Arabic Text: A Hybrid Approach with Ontology and Rule-Based.

The abstract presents a new hybrid approach for automatically extracting spatial information from Arabic text documents in geographic information systems. The main objective is to automate and enhance the performance of GIS systems by making certain tasks explicit and improving the resources for Arabic Natural Language Processing (ANLP). The first step of the study involves the construction of a spatial ontology to index, annotate and extract spatial information from Arabic texts. In the subsequent step, JAPE rules are developed and employed to disambiguate and classify different types of spatial information. The evaluation of the proposed system demonstrates promising performance, with a precision rate of 93.8% and a recall rate of 95.2%. Overall, this hybrid approach presents a significant contribution to automating spatial information extraction from Arabic texts, enhancing GIS systems, and improving ANLP resources. The positive experimental results highlight its potential for various practical applications in geographic information retrieval and natural language processing. [ABSTRACT FROM AUTHOR]