Penentuan Nomor Panggil Otomatis Berbasis DDC pada Dokumen Perpustakaan Menggunakan Klasifikasi Naive Bayes

Abstract

This research discusses automatic call number classification for library document with Dewey decimal classification (DDC) scheme. The goal of this research is to determine the call number of library document automatically based on DDC using naive Bayes classification. This research uses control vocabulary to determine DDC class of each document. Documents used in this are 190 documents with 114 training documents and 76 testing documents. This research uses information gain for feature selection. Result of the feature selection will be used as a reference in modelling classification. In modelling classification, a similarity calculation between results of token feature selection processing and control vocabulary using Natural Language ToolKit (NLTK) is performed. The accuracy obtained in the application of the naive Bayes method with the use of stemming is 45.26% and the non-stemming is 44.21%.