Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Galuh Fadillah Grandis
Fakultas Ilmu Komputer, Universitas Brawijaya
Computer Science & IT Control & Systems Engineering Education Electrical & Electronics Engineering Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Seleksi Fitur Gain Ratio pada Analisis Sentimen Kebijakan Pemerintah Mengenai Pembelajaran Jarak Jauh dengan K-Nearest Neighbor Galuh Fadillah Grandis; Yuita Arum Sari; Indriati Indriati
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 5 No 8 (2021): Agustus 2021
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

The Ministry of Education and Culture (Kemendikbud) released Circular Number 15 of 2020 concerning Guidelines for Implementing Learning from Home in an Emergency for the Spread of COVID-19, which is known as Distance Learning (PJJ). This Circular was issued during the Covid-19 emergency to protect students' rights to receive educational services while simultaneously preventing the spread of Covid-19. Following the publication of this policy, numerous types of responses or opinions from the general public began to be expressed on social media, especially Twitter. This can take the shape of a good or negative opinion, thus a sentiment analysis is required to determine whether this policy has gotten a lot of favorable or negative feedback. Sentiment analysis is a method for determining the sentiments that present in each viewpoint. The K-Nearest Neighbor (KNN) classification approach is used for sentiment analysis, and it seeks to find the outcomes or values of the closest documents. In addition, the increase ratio will be used to remove irrelevant terms via feature selection. As a result, the gain ratio with the highest f-measure value, namely 0.74 at k = 11 with testing on the second fold and k = 90 with testing on the first fold, is used. In comparison to using the information gain, the outcome of employing the gain ratio for each fold has a steady f-measure value.
Co-Authors Indriati Indriati Yuita Arum Sari