PERBANDINGAN METODE K-NEAREST NEIGHBOR DAN METODE NAIVE BAYES UNTUK KLASIFIKASI KOMENTAR PRODUK ONLINE SHOP BERDASARKAN ULASAN PEMBELI
Total View This Week0

Institusion

Universitas Sriwijaya

Author

ADI NUGROHO (STUDENT ID : 09121002003)
Rusdi Efendi (LECTURER ID : 8826630017)
Kanda Januar Miraswan (LECTURER ID : 0009019002)

Subject

R858-859.7 Computer applications to medicine. Medical informatics 

Datestamp

2019-10-03 07:45:45 

Abstract :
Classification is the process of predicting data class or object which class label is unknown to the existing label or class. The classification methods including K- Nearest Neighbor and Naive Bayes. Both methods are compared for the comments classification based on the words contained in the comments that based on buyer reviews on online shop product. The comparison result shows that Naïve Bayes obtained better recall accuration rate than K-Nearest Neighbor in negative class using T-Test with value of 0.78, while recall value of K-Nearest Neighbor is 0.61. 

File :
RAMA_55201_09121002003_8826630017_8826630017_01_front_ref.pdf
RAMA_55201_09121002003_8826630017_8826630017_02.pdf
RAMA_55201_09121002003_8826630017_8826630017_03.pdf
RAMA_55201_09121002003_8826630017_8826630017_04.pdf
RAMA_55201_09121002003_8826630017_8826630017_05.pdf
RAMA_55201_09121002003_8826630017_8826630017_06.pdf
RAMA_55201_09121002003_8826630017_8826630017_07_ref.pdf