Article Per Year (5 Year)
p-Index From 2021 - 2026
0.408
P-Index
This Author published in this journals
All Journal JURNAL INSTEK (Informatika Sains dan Teknologi) Jurnal Ketopong Pendidikan
Abraham, Yohanes Adi Saputra
Unknown Affiliation
Computer Science & IT Control & Systems Engineering Education Electrical & Electronics Engineering Engineering Social Sciences

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

Found 2 Documents
Search

 1 
ANALISIS IMPLEMENTASI SUPPORT VECTOR MACHINE DAN RANDOM FOREST UNTUK PREDIKSI KATEGORI INDEKS KUALITAS UDARA JAKARTA Banjarnahor, Evander; Belferik, Ronald; Cendana, Wiputra; Abraham, Yohanes Adi Saputra
Jurnal INSTEK (Informatika Sains dan Teknologi) Vol 10 No 1 (2025): APRIL
Publisher : Department of Informatics Engineering, Faculty of Science and Technology, Universitas Islam Negeri Alauddin, Makassar, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24252/instek.v10i1.56477

Abstract

Kualitas udara yang buruk di Jakarta berdampak signifikan terhadap kesehatan masyarakat dan lingkungan. Oleh karena itu, diperlukan metode prediksi untuk membantu pengambilan kebijakan mitigasi polusi udara. Penelitian ini memprediksi kategori indeks kualitas udara dengan metode Support Vector Machine (SVM) dan Random Forest menggunakan data polutan (PM10, PM2.5, SO₂, CO, O₃, NO₂) dari Kaggle tahun 2021, meliputi PM10, PM2.5, SO2, CO, O3, dan NO2. Analisis korelasi menunjukkan bahwa PM10 dan PM2.5 memiliki hubungan yang sangat kuat (r = 0.96), menandakan keterkaitan erat dalam menentukan tingkat polusi udara. SVM dan Random Forest disimulasikan dengan berbagai rasio pembagian data latih dan uji (10:90, 15:85, 20:80, 25:75, dan 30:70), serta menggunakan stratified k-fold cross-validation untuk meningkatkan validitas hasil dan mengurangi potensi overfitting. Hasil evaluasi menunjukkan bahwa kedua model memberikan performa yang sangat baik dengan akurasi lebih dari 97% pada seluruh skenario pembagian data. Random Forest mencapai akurasi maksimum 100% pada rasio 15:85, sementara SVM mencatatkan akurasi tertinggi 98,9% pada rasio 25:75. Hasil cross-validation menunjukkan bahwa Random Forest mencapai akurasi 100% pada simulasi menggunakan 5-folds, dengan nilai presisi, recall, dan F1-score yang juga 100%. Di sisi lain SVM menunjukkan akurasi sedikit lebih rendah yaitu 97,30% namun lebih konsisten dengan standar deviasi 2,50%.
Survei Penerapan Teori Behaviorisme Skinner Pada Pendidikan Karakter di Sekolah XYZ HI, Dwi Aning Apriandani; Gea, Eklesia Khesywania; Prasetyo, Yogi Widi; Abraham, Yohanes Adi Saputra
Jurnal Ketopong Pendidikan Vol. 5 No. 2: December 2025
Publisher : Magister Teknologi Pendidikan Fakultas Ilmu Pendidikan Universitas Pelita Harapan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.19166/jkp.v5i2.10447

Abstract

Character education is an important aspect in shaping students' behavior and morals. This study aims to evaluate the application of B.F. Skinner's behaviorism theory in character education and analyze its influence on improving students' morals and ethics at XYZ High School. This study used a survey method with a quantitative approach to collect data from 34 students (55.74% of the population of 61 students) at XYZ High School Gading Serpong. The research instrument was a questionnaire with 15 statements measuring five character indicators: religiosity, nationalism, independence, mutual cooperation, and integrity, using a Likert scale of 1-5. Data were analyzed using a validity test (Cronbach's Alpha = 0.82), a normality test (p-value = 0.38), and a regression test. The results showed that the five character indicators had a positive and significant influence (F-count = 51.54 > F-table = 2.934) on students' character formation, with an average character index of 60% (high category). These findings indicate that the application of Skinner's operant conditioning theory through conditioning the school environment and forming repetitive habits has proven effective in improving students' morals and ethics. This study recommends the development of behaviorism-based character education, taking into account the challenges of the digital era and the influence of social media.
Co-Authors Banjarnahor, Evander Belferik, Ronald Cendana, Wiputra Gea, Eklesia Khesywania HI, Dwi Aning Apriandani Prasetyo, Yogi Widi