A Characterwise Windowed Approach to Hebrew Morphological Segmentation
This paper presents a novel approach to the segmentation of orthographic word forms in contemporary Hebrew, focusing purely on splitting without carrying out morphological analysis or disambiguation. Casting the analysis task as character-wise binary classification and using adjacent character and word-based lexicon-lookup features, this approach achieves over 98 the benchmark SPMRL shared task data for Hebrew, and 97 of domain Wikipedia dataset, an improvement of 4 of the art performance.
READ FULL TEXT