Morphology of composition functions in persian sentences through a newly proposed classified fuzzy method and center of gravity defuzzification method

Abstract

Since data mining plays a key role in a variety of research areas including audio processing, text processing, and machine translation, research efforts in the Persian language has been seriously challenging. Due to difficulties such as clinging letters and absence of vowels in words, research in the Persian language has progressed at a far slower pace than in the English language. In spite of all complexities, it is indispensable to conduct further studies on the Persian morphology. Moreover, one of the widely adopted techniques is statistical, where a major drawback is intensive computational load. For that purpose, this research project intended to propose a “classified, statistical” method. Compared to the solely statistical method, the new one offers the advantage of 80% reduction in computation load at each stage against previous stages. This in turn enhances the essential computation speed. The results indicated that sequencing and classification of stages in Persian proved more successful by about 13% against the combined fuzzy and classified method.