Article Per Year (5 Year)
p-Index From 2020 - 2025
1.704
P-Index
This Author published in this journals
All Journal Insect (Informatics and Security) : Jurnal Teknik Informatika Journal of Information Systems and Informatics bit-Tech Journal of Computer Science, Information Technology and Telecommunication Engineering (JCoSITTE) International Journal of Informatics, Information System and Computer Engineering (INJIISCOM) TRANSFORMASI : JURNAL PENGABDIAN PADA MASYARAKAT Jurnal Islam Nusantara Jurnal Informatika Progres Malcom: Indonesian Journal of Machine Learning and Computer Science
Aras, Suhardi
Unknown Affiliation
Religion Humanities Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Education Electrical & Electronics Engineering Engineering Industrial & Manufacturing Engineering Mechanical Engineering Social Sciences

Published : 11 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 11 Documents
Search

 1   2 
Sentiment Analysis on Shopee Product Reviews Using IndoBERT Aras, Suhardi; Yusuf, Muhammad; Ruimassa, Reinhard Yohanis; Wambrauw, Elli Agustinus Billi; Pala'langan, Elsa Bura
Journal of Information System and Informatics Vol 6 No 3 (2024): September
Publisher : Universitas Bina Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51519/journalisi.v6i3.814

Abstract

A marketplace is a place in cyberspace where there are commercial activities between buyers and sellers. Products offered from the marketplace have reviews to review. Shopee is the most visited marketplace by people and offers various products. Product reviews can provide benefits for other consumers in assessing the products offered. By utilizing NLP technology in particular, this study can classify positive sentiment and negative sentiment in product review data. The IndoBERT model is a model that can be used in NLP technology by utilizing the relationship between each input and output element as well as the weights to be calculated simultaneously. By utilizing this technology, sentiment analysis on Shopee product reviews provides maximum accuracy until 93% with different training conditions. This provide that IndoBERT model can show that the performance of the indoBERT model in this research is very good.
Digital Skill Trainin for Deaf Community in Sorong City Aras, Suhardi; Yusuf, Muhammad; Surahmanto, Muhammad; Pangri, Muzakkir
TRANSFORMASI : JURNAL PENGABDIAN PADA MASYARAKAT Vol 4, No 3 (2024): Desember
Publisher : UNIVERSITAS MUHAMMADIYAH MATARAM

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31764/transformasi.v4i3.28601

Abstract

This community service program aims to make a direct contribution to the community. The main focus is to enhance participants' understanding of the application of artificial intelligence (AI) to support learning and earn income through micro-job services on various platforms. There were 40 participants in this activity, which was held in person on November 9, 2024 at Hangout Avenue, Sorong City. The methods used included the delivery of materials through lectures, question and answer sessions, and hands-on training in the use of AI applications and micro-job services. The results showed a significant increase in participants' understanding and skills in the use of digital technology to support the economy and education. This program has an important role in the empowerment of the Deaf community, students, and college students so that they can make the best use of emerging AI applications to complete microjobs on various platforms and increase their income
Sistem Deteksi Simbol Isyarat Pada SIBI (Sistem Isyarat Bahasa Indonesia) Menggunakan Mediapipe Berbasis Android Soekarta, Rendra; Aras, Suhardi; Rezki, Rezki; Ainun K.D.P, Nofryanti
Insect (Informatics and Security): Jurnal Teknik Informatika Vol. 10 No. 2 (2024): Oktober 2024
Publisher : Universitas Muhammadiyah Sorong

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33506/insect.v10i2.4079

Abstract

Penyandang disabilitas, khususnya tunarungu, menghadapi tantangan besar dalam berkomunikasi, terutama di perguruan tinggi yang seringkali belum sepenuhnya inklusif. Penelitian ini bertujuan untuk mengembangkan sistem deteksi Bahasa Isyarat Indonesia (SIBI) berbasis Android menggunakan framework MediaPipe dan teknologi deep learning. Dataset penelitian mencakup 24 kelas huruf abjad statis yang diambil menggunakan kamera smartphone dan sumber daring. Proses pengembangan melibatkan ekstraksi fitur menggunakan MediaPipe dan integrasi ke platform Android. Pengujian menunjukkan model mencapai akurasi 90,85%, meskipun terdapat tantangan pada kelas dengan pola isyarat yang mirip. Hasil usability testing dengan 60 pengguna mencatat rata-rata kepuasan 83,51%, menunjukkan respons positif terhadap kemudahan penggunaan dan manfaat aplikasi. Aplikasi ini memungkinkan deteksi isyarat secara akurat dan cepat, serta stabil di berbagai kondisi. Hasil penelitian ini penting untuk meningkatkan apksesibilitas komunikasi penyandang tunarungu melalui solusi teknologi yang mudah diakses dan aplikatif.
AI-Based Chatbot System for Education and Recommendations on the Use of Native Papuan Herbal Plants using the Large Language Models method Aras, Suhardi; Kelian, Maya Nurliati; S, Ilham; Faridah, Alfiyyah
International Journal of Informatics, Information System and Computer Engineering (INJIISCOM) Vol. 6 No. 1 (2025): INJIISCOM: VOLUME 6, ISSUE 1, JUNE 2025
Publisher : Universitas Komputer Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34010/injiiscom.v6i1.13575

Abstract

This study develops an AI-based chatbot to provide education and recommendations for using Papuan herbal tea. The goal is to raise awareness and dispel traditional knowledge about Papuan herbal tea, which has historically been weak in local communities. The methods used include problem analysis, literature review, data collection, system testing using Large Language Models (LLM), and implementation using Google Generative AI (Gemini-Pro). This system is designed to analyze text queries and provide relevant answers using text-based query syntax. Pengujian is carried out using the BERTScore method to assess the system's adherence to the reference data. The study's results indicate very good performance, with a rata-rata Precision of 0.58357/58 %, a Recall of 0.66781/66%, and an F1-Score of 0.62205/62. %. More precise recall indicates that the system can capture a lot of pertinent information, even when there are performance differences between questions. This study demonstrates the potential of AI technology to improve understanding between local communities and the general public as well as to support local Papuan kearifan events
Bettering Academic Services through the Use of an Integrated Academic Information System Hartinah, Hartinah; Aras, Suhardi; Nugroho, Moch Nur Syawalludin; Mutmainna, Mutmainna; Jafar, Lili Septiani
JURNAL ISLAM NUSANTARA Vol 8, No 2 (2024)
Publisher : Lembaga Ta'lif wa An-Nasyr (LTN) PBNU

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33852/jurnalnu.v8i2.556

Abstract

Nowadays, humans cannot be separated from information technology; this significantly impacts various fields, especially educational administration. Academic services that were initially manual have now transformed into digital, where academic data, both information and other administration, can now be handled by information systems. This study aimed to determine the effectiveness of implementing the integrated academic information system (SIAT) that has been running in Islamic universities. Data collection was carried out using a descriptive qualitative method concerning the field research approach combined with data triangulation; data collection was obtained by distributing questionnaires, qualitative interviews, and observations. The findings were that SIAT could increase the effectiveness and efficiency of academic services at the Faculty, ease of accessing information, especially in KRS filling activities, KHS checking, lecture monitoring, lecturer grade input, and distribution of lecture schedules can be done in one system. The obstacles felt were few, especially if the system was undergoing maintenance, data loading was hampered, and some students did not understand how to operate the information system. The solution that can be done is to hold periodic socialization regarding the implementation of SIAT and to update the system so that it remains optimal in its operation.
Deteksi Keretakan Permukaan Gedung Menggunakan Algoritma YOLO Berbasis Web Soekarta, Rendra; Aras, Suhardi; Fitrah Pasaribu, Andi Muhammad
Insect (Informatics and Security): Jurnal Teknik Informatika Vol. 11 No. 1 (2025): Maret 2025
Publisher : Universitas Muhammadiyah Sorong

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33506/insect.v11i1.4339

Abstract

Keretakan permukaan gedung merupakan salah satu permasalahan yang sering terjadi pada bangunan, yang disebabkan oleh faktor-faktor seperti bencana alam, perubahan temperatur, kelembaban tinggi, dan penggunaan bahan bangunan berkualitas rendah. Permasalahan ini dapat mengancam keselamatan dan kenyamanan pengguna bangunan jika tidak segera ditangani. Metode konvensional dalam mendeteksi keretakan pada bangunan sering kali melibatkan inspeksi manual yang memakan waktu dan berisiko tinggi. Penelitian ini bertujuan untuk mengembangkan sistem deteksi keretakan permukaan gedung menggunakan algoritma YOLO (You Only Look Once) berbasis web. Sistem ini dirancang untuk meningkatkan efisiensi dan akurasi dalam mendeteksi keretakan dengan memanfaatkan teknologi pemrosesan citra dan pembelajaran mendalam (deep learning). Algoritma YOLO dipilih karena kemampuannya dalam melakukan deteksi objek secara dengan tingkat akurasi yang tinggi. Hasil evaluasi menunjukkan bahwa sistem deteksi keretakan permukaan gedung menggunakan algoritma YOLOv8 berbasis web berhasil mencapai tingkat akurasi sebesar 54,8%, dengan nilai precision sebesar 62,9%, recall sebesar 81,1%, F1-score sebesar 70,1%, dan nilai mAP sebesar 75,1%. Dengan demikian, sistem ini diharapkan dapat menjadi solusi yang efektif dan efisien dalam memantau kondisi bangunan, mengurangi risiko kecelakaan kerja, dan meningkatkan keselamatan serta kenyamanan pengguna gedung.
Development of a Chatbot Education System for Protected Marine Animals in Papua Using the Large Language Models Method Anang, Alfikran; Maimunah, Anggun; Tawainella, Annisa Iriani; Aras, Suhardi
Journal of Computer Science, Information Technology and Telecommunication Engineering Vol 6, No 1 (2025)
Publisher : Universitas Muhammadiyah Sumatera Utara, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30596/jcositte.v6i1.20677

Abstract

The research aims to raise awareness about protected marine species in Papua through a chatbot providing accessible information. This strengthens public and student understanding of marine conservation. Methods include problem analysis, literature study, data collection, and system analysis and design using PyMuPDF and a Large Language Model (LLM). Data from scientific journals, books, e-books, and websites is processed through chunking and embedding to create vector representations. The FITZ library stores these vectors, enabling the chatbot to find similarities and provide relevant answers. Results show the chatbot accurately delivers information on Papua's protected marine animals. Testing with BERTScore indicated a high semantic correlation between chatbot responses and reference data. User satisfaction surveys reveal positive contributions to understanding marine conservation. The chatbot is relevant and satisfying, though performance and functionality could be improved with advanced technology. Further research should enhance data quality and answer consistency to better meet user needs.
IMPLEMENTASI ALGORITMA YOLO UNTUK MENDETEKSI JENIS TANAMAN HIAS BERBASIS ANDROID Soekarta, Rendra; Aras, Suhardi; Rahman, Muh Fadhil
PROGRESS Vol 17 No 1 (2025): April
Publisher : P3M STMIK Profesional Makassar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56708/progres.v17i1.456

Abstract

Ornamental plants possess high aesthetic value and environmental benefits, yet identifying their species often poses a challenge, especially for beginners. This study aims to develop an Android-based application employing the You Only Look Once version 8 (YOLOv8) algorithm to detect ornamental plant species through leaf images in real-time. The dataset comprises 1,096 images of ornamental plant leaves, including snake plant (Sansevieria), aloe vera (Aloe vera), and coral cactus (Cereus peruvianus). The data were annotated using bounding box techniques, and the model was trained on Google Colab with an 80:20 split between training and testing datasets. The training resulted in an accuracy rate of 96% based on the mean Average Precision (mAP) metric. The application was developed using Android Studio with a user-friendly interface, enabling real-time detection on Android devices with a minimum RAM specification of 3 GB. Application testing involved black-box testing to ensure functionality and usability testing with 31 respondents, revealing a user satisfaction rate of 87%. Some challenges encountered included the impact of lighting on detection accuracy and result variability across different devices. This study contributes to the utilization of artificial intelligence technology for biodiversity education and supports environmental conservation efforts
Deep Learning-Based Approach for Identifying and Counting Wooden Blocks with YOLO Aras, Suhardi; Soekarta, Rendra; Umasugi, Edwin
bit-Tech Vol. 8 No. 1 (2025): bit-Tech
Publisher : Komunitas Dosen Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32877/bt.v8i1.2627

Abstract

The wood processing industry in Indonesia, especially in the Southwest Papua region, faces ongoing challenges in accurately counting wooden logs, a task traditionally performed manually. Manual methods are time-intensive and prone to error, leading to inefficiencies in operations and weaknesses in resource management. This study addresses these challenges by applying a deep learning-based object detection approach, specifically the You Only Look Once version 8 (YOLOv8) algorithm, to automate the detection and counting of wooden beams in real time. YOLOv8 was chosen for its ability to perform high-speed and accurate detection even under varying environmental conditions, such as different lighting levels and camera angles. The model was trained on a custom dataset consisting of 265 annotated images of wooden beams, with a split of 70% for training, 20% for validation, and 10% for testing. Performance evaluation using a confusion matrix revealed a detection accuracy of 94%. These findings suggest that YOLOv8 is highly effective in supporting automation within wood processing workflows. By reducing dependency on manual labor and minimizing counting errors, the system contributes to more accurate inventory tracking and enhanced productivity. This research demonstrates the potential of integrating AI-driven models into mobile and industrial applications for improved efficiency in forestry-related sectors.
Analisis Perbandingan Kinerja Algoritma K-NN dan K-Means untuk Sistem Rekomendasi Mouse Gaming : Comparative Performance Analysis of K-NN and K-Means Algorithms for Gaming Mouse Recommendation System Aras, Suhardi; Malino, Agniel Lorensyus; Paliyama, Yuchiro Heizkia Reenhard; Ramadhan, Ridho Bintang
MALCOM: Indonesian Journal of Machine Learning and Computer Science Vol. 5 No. 4 (2025): MALCOM October 2025
Publisher : Institut Riset dan Publikasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57152/malcom.v5i4.2160

Abstract

Kebutuhan akan perangkat keras seperti mouse untuk bermain gim yang mendukung performa meningkat seiring dengan pertumbuhan industri gim dan teknologi digital. Fitur seperti DPI tinggi, tombol makro, sensor presisi, dan desain ergonomis menjadi daya tarik utama bagi pengguna. Namun, banyaknya variasi produk membuat konsumen kesulitan memilih mouse yang sesuai dengan preferensi mereka. Penelitian ini bertujuan untuk membangun sistem rekomendasi mouse gaming berbasis pembelajaran mesin dengan membandingkan dua algoritma, yaitu K-Nearest Neighbor (K-NN) dan K-Means. Metode penelitian meliputi pengumpulan data dari Kaggle, preprocessing data (pembersihan, normalisasi), reduksi dimensi dengan PCA, pengelompokan menggunakan K-Means, serta klasifikasi menggunakan K-NN dengan penerapan SMOTE untuk mengatasi ketidakseimbangan kelas. K-NN, sebagai algoritma supervised learning, memanfaatkan kedekatan antar data untuk menentukan label, sedangkan K-Means merupakan algoritma unsupervised yang mengelompokkan data berdasarkan kemiripan fitur. Evaluasi dilakukan menggunakan metrik akurasi, presisi, recall, dan F1-score. Hasil menunjukkan bahwa K-Means efektif dalam segmentasi produk, sementara K-NN memberikan akurasi tinggi dalam klasifikasi. Kombinasi keduanya menghasilkan sistem rekomendasi yang lebih akurat dan relevan bagi pengguna.
Co-Authors Ainun K.D.P, Nofryanti Anang, Alfikran Bazhar, Muhammad Faridah, Alfiyyah Fitrah Pasaribu, Andi Muhammad Hartinah Jafar, Lili Septiani Kelian, Maya Nurliati Maimunah, Anggun Malino, Agniel Lorensyus Muhammad Yusuf Mutmainna, Mutmainna Muzakkir Pangri Pala'langan, Elsa Bura Paliyama, Yuchiro Heizkia Reenhard Rahman, Muh Fadhil Ramadhan, Ridho Bintang REZKI, REZKI Ruimassa, Reinhard Yohanis S, Ilham Soekarta, Rendra Surahmanto Syawalludin, Syawalludin Tanra, Putriana Tawainella, Annisa Iriani Umasugi, Edwin Wambrauw, Elli Agustinus Billi