Article Per Year (5 Year)
p-Index From 2020 - 2025
0.61
P-Index
This Author published in this journals
All Journal Mechatronics, Electrical Power, and Vehicular Technology Jurnal Teknologi Informasi dan Ilmu Komputer Journal of Information Technology and Computer Science Jurnal Elektronika dan Telekomunikasi
Kusumo, R. Budiarianto Suryo
Unknown Affiliation
Computer Science & IT Control & Systems Engineering Electrical & Electronics Engineering Engineering

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

Found 4 Documents
Search

 1 
DEEP CNNBASED DETECTION FOR TEA CLONE IDENTIFICATION Ramdan, Ade; Suryawati, Endang; Kusumo, R. Budiarianto Suryo; Pardede, Hilman F.; Mahendra, Oka; Dahlan, Rico; Fauziah, Fani; Syahrian, Heri
Jurnal Elektronika dan Telekomunikasi Vol 19, No 2 (2019)
Publisher : Indonesian Institute of Sciences

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14203/jet.v19.45-50

Abstract

One factor affecting the quality of tea is the selection of plant material that would be planted on the field. Clonal selection is a common way to produce tea with better quality. However, as a natural cross pollination species, tea often consists of various clones or progenies of cross-pollinated process. This commonly occurs on plantations owned by smallholder farmers. To produce a consistent quality tea, the clones or progenies need to be identified. Usually, human experts distinguish the plants from leaves by visual inspection on the physical attributes of the leaves, such as the textures, the bone structures, and the colors. It is very difficult for non-experts or common farmers to do such identifications. In this, we propose a deep learning-based identification of tea clones. We apply deep convolutional neural network (CNN) to identify 3 types of tea clones of Gambung series, a series of tea clones developed at Research Institute of Tea and Cinchona. Our study indicates that the performance of the CNN systems are affected by the depth of the convolutional layers. VGGNet, a popular CNN architectures with 16 layers, achieves better accuracy compared to AlexNet, a CNN with 6 layers.
Robust remaining useful life prediction of lithium-ion battery with convolutional denoising autoencoder Yuliani, Asri Rizki; Pardede, Hilman Ferdinandus; Ramdan, Ade; Zilvan, Vicky; Yuwana, Raden Sandra; Amri, M Faizal; Kusumo, R. Budiarianto Suryo; Pramanik, Subrata
Journal of Mechatronics, Electrical Power, and Vehicular Technology Vol 15, No 1 (2024)
Publisher : National Research and Innovation Agency

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55981/j.mev.2024.905

Abstract

Using lithium-ion (Li-ion) batteries exceeding their useful lifetime may be dangerous for users, and hence, developing an accurate prediction system for batteries that remain useful for life is necessary. Many deep learning models, such as gated recurrent units and long short-term memory (LSTM), have been proposed for that purpose and have shown good results. However, their performance when dealing with noisy data degrades significantly. This may hamper their implementations for the real world since battery data are prone to noise. In this paper, we develop a robust prediction model in a noisy environment for predicting the remaining useful life (RUL) of Li-ion batteries. We propose a denoising autoencoder (DAE) utilized to remove noise from the data. The DAE is built with convolutional layers instead of traditional feed-forward networks here. We combine DAE with LSTM as the predictor. The proposed framework is evaluated using artificially corrupted battery data provided by National Aeronautics and Space Administration (NASA). The results reveal that our proposed method improves robustness when data contain various types of noise. A comparative study using the traditional approach has also been conducted. Our evaluation shows that convolutional layers are more effective than the traditional approach and that the original composition of the DAE was built using traditional feed-forward networks. DAE with convolutional layers has the best average performance with MSE of 0.61 and is the most consistent model.
Analysis of Entrance Test Results Effect on Student's Performance using Multiple Linear Regression Gultom, Dito William Hamonangan; Supianto, Ahmad Afif; Bachtiar, Fitra Abdurrachman; Krisnandi, Dikdik; Kusumo, R. Budiarianto Suryo; Heryana, Ana
Journal of Information Technology and Computer Science Vol. 9 No. 3: December 2024
Publisher : Faculty of Computer Science (FILKOM) Brawijaya University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25126/jitecs.93343

Abstract

The entrance selection test is the starting gate to find out the ability of prospective students. Student performance, measured by the Academic Achievement Index and the length of study, are two factors that measure the quality of a department or faculty. Therefore, it is necessary to analyze the effect of the entrance test results on student performance. The Faculty of Computer Science at Brawijaya University has various types of tests that prospective students must take in order to be accepted into the Computer Science Master's Program. Broadly speaking, these types of tests consist of Interview Tests, Academic Potential Tests, TOEFL Tests, Field Ability Tests, Psychological Tests, and S1 GPA. Regression analysis using the Multiple Linear Regression method is applied in the first 4 semesters of lectures. Tests conducted on the regression model resulted in a Mean Square Error value of 0.0321 in the first semester, 0.0273 in the second semester, 0.015 in the third semester, and 0.031 in the fourth semester, and 1.5301 for the graduation semester. While the K- Fold Cross Validation score resulted in a score of -0.044 in the 1st semester, 0.2838 in the second semester, 0.9037 in the third semester, and 0.9011 in the fourth semester, and -0.2786 in the graduation semester. In addition, dashboard visualization gets an average score of 68.33, which means it can be accepted and can be used by the Academic Team for the Master's Program in Computer Science at Brawijaya University.
Peningkatan Performa Pengelompokan Siswa Berdasarkan Aktivitas Belajar pada Media Pembelajaran Digital Menggunakan Metode Adaptive Moving Self-Organizing Maps Prasetyo, Onky; Supianto, Ahmad Afif; Anam, Syaiful; Pardede, Hilman Ferdinandus; Zilvan, Vicky; Kusumo, R. Budiarianto Suryo
Jurnal Teknologi Informasi dan Ilmu Komputer Vol 9 No 1: Februari 2022
Publisher : Fakultas Ilmu Komputer, Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25126/jtiik.2022915579

Abstract

Digitalisasi proses pembelajaran memungkinkan untuk dihasilkannya rekaman terhadap setiap aktivitas siswa selama belajar. Rekaman yang dihasilkan tersebut dapat digunakan untuk mengelompokkan siswa berdasarkan pola dari proses belajar yang dilakukan. Hasil pengelompokkan yang peroleh dapat digunakan untuk melakukan penyesuaian komponen pembelajaran ataupun metode pembelajaran bagi siswa. Salah satu metode pengelompokan yang sering digunakan adalah Self-Organizing Maps (SOM), SOM merupakan metode jaringan syaraf tiruan dengan tujuan untuk mempertahankan topologi data ketika data input multidimensi diubah menjadi data output dengan dimensi yang lebih rendah. Neuron SOM pada dimensi input diperbaharui sepanjang proses pelatihan, sedangkan neuron pada dimensi output tidak mendapatkan pembaruan sama sekali, hal ini menyebabkan struktur neuron yang digunakan pada tahapan inisialisasi akan tetap sama hingga akhir proses pengelompokan. Pada penelitian ini menggunakan metode Adaptive Moving Self-Organizing Maps (AMSOM) yang menggunakan struktur neuron lebih fleksibel, dengan dimungkinkannya terjadi perpindahan, penambahan dan penghapusan dari neuron menggunakan data 12 assignments dari media pembelajaran MONSAKUN. Hasil penelitian menunjukkan terdapat perbedaan yang signifikan secara statistik antara nilai quantization error dan nilai topographic error dari algoritme AMSOM dengan algoritme SOM. Metode AMSOM menghasilkan rata-rata nilai quantization error 27 kali lebih kecil dan rata-rata nilai topographic error 54 kali lebih kecil dibandingkan dengan metode SOM.AbstractThe digitization of the learning process makes it possible to produce recordings of each student's activity during learning. The resulting record can be used to group students based on the pattern of the learning process. The grouping results can be used to make adjustments to the learning components or learning methods for students. One of the most frequently used clustering methods is Self-Organizing Maps (SOM), SOM is a neural network method to maintain data topology when multidimensional input data is converted into output data with lower dimensions. The SOM neurons in the input dimension are updated throughout the training process, while the neurons in the output dimension do not get updated at all, this causes the neuron structure used in the initialization stage to remain the same until the end of the grouping process. In this study, the Adaptive Moving Self-Organizing Maps (AMSOM) method uses a more flexible neuron structure, allowing for the transfer, addition and deletion of neurons using 12 assignments of data from MONSAKUN learning media. The results showed that there was a statistically significant difference between the quantization error and the topographic error of the AMSOM algorithm and the SOM algorithm. The AMSOM method produces an average quantization error 27 times smaller and an average topographic error 54 times smaller than the SOM method.
Co-Authors Ade Ramdan Ahmad Afif Supianto Amri, M Faizal Dahlan, Rico Fani Fauziah, Fani Fitra Abdurrachman Bachtiar Gultom, Dito William Hamonangan Heri Syahrian, Heri Heryana, Ana Krisnandi, Dikdik Oka Mahendra Pardede, Hilman F. Pardede, Hilman Ferdinandus Pramanik, Subrata Prasetyo, Onky Suryawati, Endang Syaiful Anam Vicky Zilvan, Vicky Yuliani, Asri Rizki Yuwana, Raden Sandra