Claim Missing Document
Check
Articles

Found 3 Documents
Search

Implementation of AI in Student Health Risk Analysis: Case Study Using Random Forest and SVM Algorithms Daulay, Parhan; Ridho, Muhammad; Hidayat, Muhammad Ferdiansyah; Dewi, Sri; Ramadhani, Fanny
QISTINA: Jurnal Multidisiplin Indonesia Vol 3, No 2 (2024): December 2024
Publisher : CV. Rayyan Dwi Bharata

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57235/qistina.v3i2.4160

Abstract

The health of students on campus is often threatened by stress and unhealthy living habits, such as poor sleep patterns or lack of physical activity. Students' health is important to their achievement and well-being. By using AI and machine learning algorithms such as Random Forest and Support Vector Machine (SVM), we can analyze health data more quickly and accurately. This helps detect problems early and provide timely assistance. Early identification of students who have the potential to experience health problems, so that timely intervention can be carried out. This research uses quantitative methods by collecting from various sources related to the use of AI and machine learning algorithms in analyzing student health at Medan State University. Data was collected through a survey at Medan State University, covering daily habits such as sleep patterns, physical activity and stress levels. With the results of this analysis, campuses can design health policies that are more effective and appropriate to student needs. It is hoped that this research can contribute to improving the health and well-being of students in the campus environment, as well as preventing more serious health problems in the future.
Implementation of the Caesar Chiper Method in Cryptography for Message Data Security Hidayat, Muhammad Ferdiansyah; Ridho, Muhammad; Indra, Zulfahmi
QISTINA: Jurnal Multidisiplin Indonesia Vol 3, No 2 (2024): December 2024
Publisher : CV. Rayyan Dwi Bharata

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57235/qistina.v3i2.4027

Abstract

Data security in the digital era is a very important aspect, especially in maintaining the confidentiality and integrity of information transmitted over the network. One of the classic cryptographic methods that is often used to secure messages is the Caesar Cipher. This method uses a simple substitution technique, where each letter in the message is shifted a certain number of steps in the alphabet. This research discusses the implementation of the Caesar Cipher method as a basic solution for message data security. Although this algorithm is relatively easy to implement and understand, the Caesar Cipher has weaknesses, mainly due to its fixed shifting pattern making it vulnerable to brute force attacks or frequency analysis. Through testing and implementation, this research shows how Caesar Cipher works in the encryption and decryption process, and evaluates its effectiveness in protecting messages from unauthorized access. The results show that this method, although less powerful than modern algorithms, can still be used for certain scenarios that do not require a high level of security. On the other hand, the weaknesses underscore the importance of developing and using more complex cryptographic methods to protect more sensitive information in today's digital era.
Perancangan Sistem Pendukung Keputusan Rekomendasi Produk Skincare untuk Kulit Berminyak, Berjerawat, Normal, dan Kering Menggunakan Metode Weighted Product Putri, Rezkya Nadilla; Sitepu, Keysa Shifa Adwitia; Hidayat, Muhammad Ferdiansyah; Niska, Debi Yandra
TEKNIKA Vol. 19 No. 2 (2025): Teknika Mei 2025
Publisher : Politeknik Negeri Sriwijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.5281/zenodo.15438058

Abstract

Permasalahan dalam pemilihan produk skincare yang sesuai dengan jenis kulit masih sering dialami oleh masyarakat, terutama karena kurangnya pemahaman terhadap karakteristik kulit dan kandungan bahan aktif yang tepat. Hal ini dapat menyebabkan penggunaan produk yang tidak sesuai serta memicu permasalahan kulit baru. Penelitian ini bertujuan untuk merancang sistem pendukung keputusan (SPK) berbasis metode Weighted Product (WP) guna memberikan rekomendasi produk skincare sesuai dengan jenis kulit pengguna, yaitu berminyak, berjerawat, normal, dan kering. Data diperoleh melalui konsultasi dengan dokter spesialis kulit serta referensi produk dari platform kesehatan. Sistem dibangun dengan mempertimbangkan lima kriteria utama dan sepuluh alternatif produk pada tiap kategori skincare. Hasil perhitungan menunjukkan bahwa untuk kategori sunscreen kulit berminyak, produk dengan skor tertinggi adalah Runaskin Unscented Centella Sunscreen dan Somethinc Holyshield! UV Watery Sunscreen Gel dengan nilai vektor V 0.2448, diikuti oleh Emina Sun Battle SPF 45 PA+++ (0,0975), dan yang terendah adalah Carasun Solar Smart UV Protector (0,0123). Penerapan metode WP dalam sistem ini terbukti mampu memberikan rekomendasi yang objektif dan akurat, serta menjadi solusi efektif dalam menentukan produk skincare yang tepat sesuai jenis kulit.