ERRORS BY AUTO-MORPHOLOGICAL ANALYSIS IN A CHILDREN STORY CORPUS: AN EVALUATION OF MORPHIND PROGRAM

Abstract

Indonesian Morphological Tool, Morphind, is meant to make a proper morphological analysis before doing further automatic language processing.Morphind is applied to enrich raw Indonesian text with morphological information, the preprocessing stage of an Indonesian corpus. In this study, the data is obtained from children's stories in the website ceritaanak.org by taking 500 types of total 2101 types. The purpose of this study is to identify and classify the types of errors present in data processing using morphind program. In the analalysis I uses the method Introspective and Dictionary Indonesian (KBBI) to validate the analysis. The findings of this research suggest that there are still many aspects that can be improved about morphind. Recommendations are fixing the data base especially for OOV (out of vocabulary) and dictionary accuracy, improving the display for the Allomorph, and improving the algorithm for morpheme segmentation.