Perbandingan Penggunaan Metode Part-of-speech pada Bahasa Indonesia Menggunakan Metode N-Gram, TnT, dan Classifier Based

Abstrak

The lack of a corpus in Indonesian makes text processing in Indonesian have little interest in processing the text, therefore the aim of doing POS-Tagging is to do word labeling that takes into account the essence of the sentence based on the words contained in the sentence into a per-word tagset , the tagging process uses the N-gram, TnT, and Classifier Based methods. The three methods are used because the three methods have similar work methods and work by looking at the traces of learning and working on the previous method. The method that produces the optimal tagger of the three methods is the Classifier Based method because this method is a combination of the three methods, so that the tagger results issued are more optimal than the other two methods.