Klasifikasi Berita Menggunakan Metode KNearest Neighbor (K-NN)

Pencarian berdasarkan :

SEMUA Pengarang Subjek ISBN/ISSN Pencarian Spesifik

Pencarian terakhir:

Tugas Akhir Informatika

Klasifikasi Berita Menggunakan Metode KNearest Neighbor (K-NN)

Baiq Pratiwi Komalasari - Nama Orang;

Abstract The number of online news is increasing day by day. It is not easy for the news editor to categorize the news article manually. Therefore, an automatic news category classifier is needed. Classification is a part of data mining that is used to determine the class of unclassified object. One of the classification methods is K-Nearest Neighbor (K-NN), which is a method to calculate the distance between two objects and put the objects whose closest distance into one class. In this research, two distance calculation methods, Cosine Similarity and istance are compared in classification using K-NN. My experimental results show that Cosine similarity outperforms the istance. With the precision and recall for Cosine Similarity are 0.97 and 0.97 respectively. For the istance, it reaches 0.81 and 0.41 respectively.

Ketersediaan

Tidak ada salinan data

Informasi Detail

Judul Seri: -
No. Panggil: -
Penerbit: UNRAM : Fakultas Teknik Unram., 2018
Deskripsi Fisik: -
Bahasa: Indonesia
ISBN/ISSN: -
Klasifikasi: NONE
Tipe Isi: other
Tipe Media: other
Tipe Pembawa: -
Edisi: Edisi 1 Jilid 1
Subjek: Nearest Neighbor
Document classification
k-NN, VSM
Cosine Similarity
istance
Info Detail Spesifik: -
Pernyataan Tanggungjawab: Pratiwi Komalasari, Baiq

Versi lain/terkait

Tidak tersedia versi lain

Lampiran Berkas

Klasifikasi Berita Menggunakan Metode KNearest Neighbor (K-NN)

Komentar

Anda harus login sebelum memberikan komentar