Article Per Year (5 Year)
p-Index From 2021 - 2026
0.61
P-Index
This Author published in this journals
All Journal DEVICE : JOURNAL OF INFORMATION SYSTEM, COMPUTER SCIENCE AND INFORMATION TECHNOLOGY PERSEPTIF: Jurnal Ilmu Sosial dan Humaniora
Siti Sarah
Universitas Islam Negeri Sumatera Utara
Religion Arts Humanities Computer Science & IT Economics, Econometrics & Finance Education Environmental Science Languange, Linguistic, Communication & Media Law, Crime, Criminology & Criminal Justice Mathematics Social Sciences Other

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
Strategi Pemberitaan Diskominfo Medan Terhadap Kepercayaan dan Literasi Informasi Masyarakat di Kota Medan Siti Sarah; Muhammad Alfikri
PERSEPTIF: Jurnal Ilmu Sosial dan Humaniora Vol. 4 No. 2 (2026): PERSEPTIF: Jurnal Ilmu Sosial dan Humaniora
Publisher : Lembaga Penelitian dan Pendidikan (LPP) Kalibra

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.70716/perseptif.v4i2.488

Abstract

This study aims to analyze the news reporting strategy implemented by the Medan City Communication and Informatics Office (Diskominfo) as a form of public communication practice and its impact on public trust and information literacy among urban communities. The study employed a descriptive qualitative approach. Data were collected through in-depth interviews, participant observation, and documentation. Informants were selected using purposive sampling, consisting of internal staff of the Medan City Diskominfo and members of the public as information recipients. The findings reveal that Diskominfo’s reporting strategy primarily focuses on the utilization of digital media platforms, particularly Instagram and TikTok, with a two-way communication pattern established through active social media interaction. From the public perspective, government information is perceived as accessible and relatively credible, although audiences still tend to independently verify the information they receive. The study also found that this reporting strategy contributes to improving public information literacy, despite uneven public abilities in filtering and evaluating information. The novelty of this research lies in its emphasis on the relationship between government digital reporting strategies, public trust, and information literacy within the context of urban digital communication. This study contributes to the development of public communication studies by demonstrating that government reporting functions not only as an information delivery mechanism but also as a strategic instrument for strengthening public trust and digital information literacy simultaneously.
Analisis Sentimen Program Makanan Bergizi Gratis Menggunakan Naïve Bayes dan Random Forest Berbasis CRISP-DM Ragilia Putri Dinanti; Siti Sarah; Fiqri Dwi Al Hafiz; Aidil Alfarizi
DEVICE : JOURNAL OF INFORMATION SYSTEM, COMPUTER SCIENCE AND INFORMATION TECHNOLOGY Vol 7, No 1: JUNI 2026
Publisher : Universitas Dharmawangsa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.46576/device.v7i1.8766

Abstract

Program Makanan Bergizi Gratis (MBG) merupakan kebijakan strategis pemerintah Indonesia untuk menangani masalah kekurangan gizi dan menekan angka stunting yang mencapai 14% pada tahun 2024 demi mencapai visi Indonesia Emas 2045. Kebijakan ini memicu diskusi publik yang masif di media sosial YouTube, yang menghadirkan tantangan berupa volume data besar dan keberagaman opini. Penelitian ini bertujuan untuk menganalisis sentimen publik terhadap program MBG menggunakan metodologi Cross-Industry Standard Process for Data Mining (CRISP-DM). Dataset yang digunakan bersumber dari Kaggle yang terdiri dari 6.419 komentar YouTube. Data diproses melalui tahapan preprocessing teks dan representasi fitur TF-IDF, kemudian diklasifikasikan menggunakan algoritma Naïve Bayes dan Random Forest. Hasil penelitian menunjukkan bahwa algoritma Random Forest menghasilkan performa yang lebih unggul dengan akurasi sebesar 77,43%, sementara Naïve Bayes mencapai 65,64%. Berdasarkan distribusi data, sentimen Netral mendominasi sebesar 56,40%, diikuti oleh sentimen Positif (24,13%) dan Negatif (19,47%). Dominasi sentimen netral ini mengindikasikan bahwa masyarakat masih bersikap wait-and-see terhadap efektivitas implementasi program MBG di lapangan. Penelitian ini menyimpulkan bahwa pendekatan ensemble learning pada Random Forest lebih efektif dalam menangkap pola bahasa alami yang kompleks dibandingkan metode berbasis probabilitas murni.
Co-Authors Aidil Alfarizi Fiqri Dwi Al Hafiz Muhammad Alfikri Munajmi Ajri Nursapia Harahap Nursiah Nursiah Puspita Simbolon Putri Sukma Nabilla Ragilia Putri Dinanti Saif Eldi Hakim Sintiya Kartika