2018 : Word Sense Disambiguation in Bahasa Indonesia Using SVM

Prof. Drs. Ec. Ir. Riyanarto Sarno M.Sc Ph.D


Abstract

Currently, the use of Indonesian language on the internet is growing rapidly and many of the existing sentences contain ambiguous words. In Natural Language Processing the problem to find out the meaning of an ambiguous word is called Word Sense Disambiguation. Word sense disambiguation is a problem about how we know the meaning of an ambiguous word in a given sentence. Many uses if we can solve word sense disambiguation problems such as can be used for text classification, text clustering and for machine translation. In this paper, we propose the use of SVM algorithm with TF-IDF as the feature extraction method and Wikipedia as the training data to solve the WSD problem of Indonesian language. The results of our proposed method reach an accuracy level of 0.877.