KLASIFIKASI TEKS MENGGUNAKAN CHI SQUARE FEATURE SELECTION UNTUK MENENTUKAN KOMIK BERDASARKAN PERIODE, MATERI DAN FISIKDENGAN ALGORITMA NAIVEBAYES
Submitted : 2017-10-23, Published : .
Abstract
A comic has its own characteristics compared the other types of books. The difference between comic and other books can be seen from the category o f period, material and physical. Comicand other booksneeded an application o f classification system. Looking for the problem, classification system was made using Chi Square Feature Selection and Naive Bayes algorithm to determine the comic based on the period, material and physical. Delphi programming language and Oracle Database are used to build the Classification System. Chi Square Feature Selection acquired trait a comic is in 0.10347 and which not comic is in 1.9531. Furthermore, data is classified by the Naive Bayes algorithm. From 120 titles o f comic that consists 60 titles o f comic and non comic
used to build classesfor trainand 60 titles o f comic and non comic used to test. The results o f Naive Bayesalgorithm for comic is 96,67%with 3.33% error rate, and non comic is 90% with 10% error rate. The classification to determine comic is good.
Keywords
Full Text:
PDF (Bahasa Indonesia)References
Jogiyanto, Hartono. 2005. Analisis dan Desain Sistem Informasi. Yogyakarta : Penerbit Andi.
Kadir, Abdul. 2009. Dasar Perancangan dan Implementasi Database Relasional. Yogyakarta : Andi Offset.
Kadir, Abdul. 2005. Pemrograman Database dengan Delphi 7 Menggunakan Access dan ADO. Yogyakarta : Andi Offset.
Nugroho, Bunafit. 2005. Database Relational dengan MySQL. Yogyakarta : Andi Offset. 66 Volume 5, Nomor 2, November 2016
Article Metrics
Abstract view: 970 times
Download : 1048 times
This work is licensed under a Creative Commons Attribution 4.0 International License.