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
Muhammad Fauzan Ziqroh
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 
Klasifikasi Jenis Barang Bekas menggunakan Metode Naive Bayes dengan Seleksi Fitur Information Gain (Studi Kasus : Akun Instagram Jual Beli Barang Bekas @infobarkas_Jogja) Muhammad Fauzan Ziqroh; Indriati Indriati; Edy Santoso
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 7 No 1 (2023): Januari 2023
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

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

Abstract

Instagram is one of the most popular social media for marketing. The rapid development of this social media function gave rise to a new term, known as Social Media Influencer. Social media influencers are advertising agencies that reach a certain number of Instagram account followers so that advertisers can place their ads on the account by paying a certain amount of fees. Instagram account of @infobarkas_jogja is one of the social media influencers that provides paid advertising services with advertising content about used goods in Yogyakarta City. However, the management of the account has several obstacles, one of which is the classification or grouping of the goods based on the types of category.The purpose of this research is to create a system that is able to classify types of goods based on their categories. This research uses the Naive Bayes Classifier method using the IG feature selection.The data for this study is in the form of text taken from the caption posts on the Instagram account @infobarkas_jogja with a total of 500 data.With a total of 400 training data and 100 testing data. The classes for the categories in this study were divided into 5 classes, namely property, vehicles, clothing, furniture, and gadgets. The threshold used is 10% ranging from 10% to 90% and produces the highest accuracy of 98% when the threshold is 10%, 40%, 80%, and also 90%. The highest accuracy is also obtained when carrying out classification without resorting to feature selection.
Co-Authors Edy Santoso Indriati Indriati