Article Per Year (5 Year)
p-Index From 2021 - 2026
0.444
P-Index
This Author published in this journals
All Journal Techno.Com: Jurnal Teknologi Informasi Jurnal Penelitian Rumpun Ilmu Teknik
Pratama, Bima Gerry
Unknown Affiliation
Computer Science & IT Engineering

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

Found 2 Documents
Search

 1 
Klasifikasi Sinyal EEG Subband Beta untuk Identifikasi Persepsi Rasa Manis dan Asam Menggunakan Algoritma Machine Learning Lejap, Marianus Yakobus Lili; Tena, Silvester; Pratama, Bima Gerry
Techno.Com Vol. 24 No. 4 (2025): November 2025
Publisher : LPPM Universitas Dian Nuswantoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62411/tc.v24i4.15011

Abstract

Aktivitas gelombang otak (EEG) dapat digunakan untuk mengenali respons manusia terhadap stimulus sensorik, termasuk persepsi rasa. Penelitian ini bertujuan untuk mengklasifikasikan aktivitas otak terhadap dua jenis stimulus rasa, yaitu rasa manis (Susu) dan rasa asam (Lemon), menggunakan sinyal EEG pada subband Beta (12–25 Hz) dengan pendekatan machine learning. Penelitian ini merupakan pengembangan dari studi sebelumnya yang hanya menampilkan visualisasi topografi otak (brain heatmap), dengan menambahkan analisis klasifikasi otomatis berbasis kecerdasan buatan. Data EEG direkam dari empat kanal utama, yaitu T3, T4, CP1, dan CP2, kemudian diekstraksi menggunakan dua fitur utama: Mean Absolute Value (MAV) dan Variance (VAR). Total data yang digunakan sebanyak 10.644 potong data (3.550 Susu dan 7.094 Lemon). Tiga algoritma machine learning digunakan untuk membandingkan performa klasifikasi, yaitu Support Vector Machine (SVM), K-Nearest Neighbor (KNN), dan Decision Tree (DT). Hasil pengujian menunjukkan bahwa Decision Tree menghasilkan performa terbaik dengan akurasi 84,0%, F1-score 0,727, dan ROC AUC 0,789, diikuti oleh KNN dengan akurasi 76,6%. Model SVM linear menunjukkan performa terendah akibat ketidakseimbangan data dan distribusi non-linear. Hasil ini membuktikan bahwa fitur EEG pada subband Beta dapat digunakan untuk membedakan stimulus rasa manis dan asam secara objektif. Penelitian ini memberikan kontribusi terhadap pengembangan sistem EEG-based Taste Recognition dan membuka peluang penerapan dalam bidang neurogastronomi serta Brain–Computer Interface (BCI). Kata kunci: EEG, subband Beta, klasifikasi rasa, machine learning, Decision Tree.
Pembelajaran Bertahap Logika dan Pemrograman Visual Menuju Implementasi Smart Watering System Berbasis ESP32 dalam Program PkM di Salah Satu SMA Swasta di Kota Jambi Violla Gunova; Bima gerry Pratama; Ahmad Abuzar Al-Hamdani; Shelly Angella
Jurnal Penelitian Rumpun Ilmu Teknik Vol. 5 No. 2 (2026): Jurnal Penelitian Rumpun Ilmu Teknik
Publisher : Lembaga Pengembangan Kinerja Dosen

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55606/juprit.v5i2.6824

Abstract

Microcontroller-based system training requires participants to understand fundamental logic and programming concepts before directly implementing hardware-based applications. This study analyzes the improvement of participants' basic competencies through the initial stage of an ESP32-based Smart Watering System training program designed as a progressive learning sequence from logical concepts to visual programming. The initial training stage was conducted for six days and covered mathematical logic and logic gates, visual programming using Scratch, and mobile application development using MIT App Inventor. Prior to the training, participants completed a 20-item pre-test to measure their initial competence, while a post-test was administered after completing the entire initial training stage. A total of 34 participants were registered at the beginning of the program, of whom 21 completed all training activities and participated in the post-test. The results showed that the participants' mean score increased from 68.33 in the pre-test to 86.67 in the post-test, representing an improvement of 18.33 points. A paired-samples t-test indicated a significant difference between pre-test and post-test scores with t(20)=4.36 and p<0.001. Cohen's dz effect size of 0.95 indicated a large magnitude of change. These results suggest that progressive learning through mathematical logic, logic gates, Scratch, and MIT App Inventor can support the development of participants' fundamental competencies before proceeding to C programming, Arduino, and the implementation of an ESP32-based Smart Watering System.
Co-Authors Ahmad Abuzar Al-Hamdani Lejap, Marianus Yakobus Lili Shelly Angella Silvester Tena Violla Gunova