Analisis Sentimen Pemindahan Ibu Kota Indonesia Dengan Pembobotan Term BM25 Dan Klasifikasi Neighbor Weighted K-Nearest Neighbor

Marinda Ika Dewi Sakariana (Fakultas Ilmu Komputer, Universitas Brawijaya)
Indriati Indriati (Fakultas Ilmu Komputer, Universitas Brawijaya)
Candra Dewi (Fakultas Ilmu Komputer, Universitas Brawijaya)

Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Vol 4 No 3 (2020): Maret 2020

The relocation of Indonesia capital city is one of the policies that is being intensively discussed at this time. With the policy regarding the relocation of the capital city from Jakarta to Kalimantan, it will certainly cause various reactions or comments from the community that can be found on social media, Twitter. Types of reactions can be divided into positive and negative comments. To find out a comment has a positive or negative value, sentiment analysis is needed as in this study. In this study, there are several steps that must be done to get the final results. These stages are pre-processing data, term weighting and ranking with the BM25 algorithm, and classifying the final results of tweets with Neighbor Weighted K-Nearest Neighbor (NWKNN) algorithm. This study uses 480 training data and 120 test data divided into positive and negative sentiments. The highest accuracy value obtained was 93.33% with a precision value of 92.45%, a recall of 94.67% and an f-measure of 93.55% with a K value of 25, =1,2 and =0,65 also an E value of 4.