Modul Word Graph Kata Kerja pada KG_EDITOR Berbasis Desktop

Abstract

Knowledge Graph (KG) represents result of text-semantic analysis as a graph. KG_EDITOR is an application to generate KG. It is a desktop-based application developed using Java programming language. The purpose of this research is to create a module to generate a word graph of Indonesian verb using KG method based on user‟s input. Based on the recent research, there are ten types of word graph for Indonesian verb along with 31 formation patterns due to affixation process. KG generation process is started by input pre-processing. Afterwards the input is looked up in Kamus Besar Bahasa Indonesia (KBBI). The word will be stemmed, if it is not found in KBBI. Then, data will be generated in the form of basic word and affix. Those data are used to determine the word graph pattern to generate. In this research, the module can recognize 10 types of word graph of verb with more than 98% accuracy