Arina, Miska (2024) Perbandingan Metode NaÏve Bayes Dan K-nearest Neighbor (knn) Dalam Analisis Sentimen Terhadap Opini Publik Mengenai Childfree Pada Twitter. Other thesis, Universitas Islam Riau.

Text Miska Arina.pdf - Submitted Version Restricted to Registered users only Download (5MB) | Request a copy

Abstract

The Childfree phenomenon has spread in developed countries such as Japan and Norway. Data from the Central Statistics Agency states that there is a decrease in births in Indonesia which has a possible influence from the Childfree phenomenon. This research aims to explore the extent to which the Childfree polemic has touched the lives of Indonesian people by taking data from one of the social media, namely Twitter. The data will go through the stages of sentiment analysis research to find and compare the accuracy value of the Naïve bayes and K-Nearest Neighbor methods. The results showed less relevant accuracy values for three sentiment classes, therefore experiments with two sentiments classes were carried out to simplify the analysis. The experimental results show that the accuracy value obtained by the Naïve Bayes method is higher using InSet labelling with an accuracy of 77% while the KNN method is higher using VADER labelling with an accuracy of 75%.

Actions (login required)

View Item