Article Per Year (5 Year)
p-Index From 2021 - 2026
0.778
P-Index
This Author published in this journals
All Journal JUSTIN (Jurnal Sistem dan Teknologi Informasi) Journal of Data Science Methods and Applications
Triyasri, Novita
Unknown Affiliation
Biochemistry, Genetics & Molecular Biology Computer Science & IT Engineering Library & Information Science

Published : 4 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 4 Documents
Search

 1 
Prediksi Keberhasilan Akademik Menggunakan Metode Regressi Logistik Dan Support Vector Machine Triyasri, Novita; Safitri, Egi; Kurniawan, Hendra; Saputra, M Hardi; Syidada, Amran Rahman; Pratama, Raynaldo Syah
JUSTIN (Jurnal Sistem dan Teknologi Informasi) Vol 13, No 4 (2025)
Publisher : Jurusan Informatika Universitas Tanjungpura

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26418/justin.v13i4.89731

Abstract

Penelitian ini bertujuan untuk memprediksi keberhasilan akademik dengan menggunakan dua metode yaitu regresi logistik dan support vector machine (SVM). Keberhasilan akademik seringkali dipengaruhi oleh banyak faktor, antara lain motivasi siswa, keterampilan belajar, dan kondisi sosial ekonomi. Oleh karena itu penting untuk mengidentifikasi variabel-variabel yang mempengaruhi keberhasilan akademik dan menggunakan teknik analisis yang tepat untuk menghasilkan prediksi yang akurat. Data yang digunakan di Penelitian ini mencakup variabel-variabel seperti nilai ujian, motivasi belajar dan tingkat kehadiran siswa. Metode regresi logistik digunakan untuk menganalisis hubungan antara variabel independen dan variabel dependen (hasil akademik), sedangkan SVM digunakan untuk mengklasifikasikan siswa berdasarkan prestasi akademiknya. Hasil penelitian menunjukkan bahwa kedua metode tersebut memberikan tingkat akurasi yang signifikan dalam memprediksi keberhasilan akademik siswa. Namun Regresi logistik menghasilkan model yang lebih sederhana, SVM menunjukkan keunggulan dalam hal akurasi dan kemampuan mengklasifikasikan siswa dengan prestasi akademik lebih tinggi. Penelitian ini memberikan informasi berharga bagi para pendidik dan manajer pendidikan untuk mengidentifikasi siswa yang memerlukan perhatian lebih dalam pembelajaran dan merancang intervensi yang lebih efektif untuk meningkatkan hasil akademik siswa.
Analisis Klastering dari Data Behavior Online Gaming Menggunakan Algoritma K-Means Salsabila Shahibah; Triyasri, Novita; Inanti, Adisty Anggi; Rachel, Jovita; Pratama, Niko Diki; Andriansyah, Andriansyah
Journal of Data Science Methods and Applications Vol. 1 No. 1 (2025)
Publisher : Program Studi Sains Data - Institut Informatika dan Bisnis Darmajaya

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

Abstract

Games that require an internet connection are called online games. Just like offline games, online games also have many genres. Some of them are Action, advanture, sports, RPG, and simulation. The emergence of various types of online games provides many choices to eliminate boredom in filling free time. In addition, there are also various levels such as easy, medium, and hard. The levels in this game also affect the habits of players in playing games. This study aims to find the optimal cluster in the dataset using clustering analysis using the K-Means algorithm on the RapidMiner application. The results of this study show that cluster 1 at k=3 from (k=2-7) is the best cluster compared to other clusters
Analisis Klasifikasi Multikelas Obesitas Menggunakan Algoritma Decision Tree Classifier, Random Forest Classifier, dan Support Vector Classifier (SVC) Septa, Oon; Triyasri, Novita; Permata, Maharani Aulia; Salsabila, Aghitsna; Firdani, Fahri
Journal of Data Science Methods and Applications Vol. 2 No. 1 (2026)
Publisher : Program Studi Sains Data - Institut Informatika dan Bisnis Darmajaya

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

Abstract

Obesity remains a significant global health challenge, making early classification and detection essential to minimize the risk of more serious degenerative diseases. This study compares three machine learning algorithms—Random Forest, Decision Tree, and Support Vector Classifier (SVC) to determine which model is most effective in predicting weight status categories and obesity risks. The dataset used includes physical features such as age, height, weight, and Body Mass Index (BMI). The research process encompasses data preprocessing stages, feature correlation analysis, data splitting, model training, and evaluation using various performance metrics.The research results indicate that Random Forest demonstrates the highest discriminative ability with an AUC of 0.99, showing perfect accuracy in distinguishing between obesity categories. Decision Tree provides identical results in terms of accuracy at 95.45%, but with a slightly lower AUC value of 0.96. Meanwhile, SVC yields competitive results with an accuracy of 81.82%, although its performance remains below the two tree-based models. Overall, this study demonstrates that ensemble methods such as Random Forest hold great potential for use as decision support systems in detecting and classifying obesity status more accurately and reliably.
Prediksi Tingkat Pengetahuan Mahasiswa Menggunakan Logistic Regression dan Random Forest Triyasri, Novita; Andini, Rekha Apriliana; Chandra, Aurea Ivana; Saprianti, Assyifa; Yusiandra, Erick
Journal of Data Science Methods and Applications Vol. 2 No. 1 (2026)
Publisher : Program Studi Sains Data - Institut Informatika dan Bisnis Darmajaya

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

Abstract

particularly for predicting students’ knowledge levels more accurately. This study aims to compare the performance of Logistic Regression and Random Forest algorithms in predicting students’ knowledge levels using the User Knowledge Modeling dataset. The dataset consists of 258 instances with five numerical attributes, namely STG, SCG, STR, LPR, and PEG, and one target variable UNS representing students’ knowledge levels. The reserch stages include data selection, preprocessing, data normalization, train-test splitting, and handling class imbalance using the SMOTE method. Model performance is evaluated using accuracy, precision, recall, and F1-score metrics. The results show that Logistic Regression outperforms Random Forest, achieving higher accuracy and F1-score values. These findings indicate that the relationships among variables in the dataset tend to be linear. Therefore, Logistic Regression is considered more suitable for predicting students’ knowledge levels in this study.
Co-Authors Andini, Rekha Apriliana Andriansyah Andriansyah Chandra, Aurea Ivana Firdani, Fahri Hendra Kurniawan Inanti, Adisty Anggi Permata, Maharani Aulia Pratama, Niko Diki Pratama, Raynaldo Syah Rachel, Jovita Safitri, Egi Salsabila Shahibah Salsabila, Aghitsna Saprianti, Assyifa Saputra, M Hardi Septa, Oon Syidada, Amran Rahman Yusiandra, Erick