Garuda - Garba Rujukan Digital

Penerapan Sistem Pembelajaran Interaktif Berbasis Augmented Reality Khusus Difabel Putra, Ova Nurisma; Sumardi, Idi
Jurnal ICT : Information Communication & Technology Vol 20, No 2 (2021): JICT-IKMI, Desember 2021
Publisher : STMIK IKMI Cirebon

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36054/jict-ikmi.v20i2.398

Education is a necessity that must be possessed by every human being, regardless of whether normal or disabled, especially for the disabled. This research aims to improve the quality of education for the disabled by providing confidence and the ability to receive lessons to be more interesting and interactive about what is conveyed by the teacher. The application of information technology in interactive learning systems is very helpful in the learning process for people with disabilities. The latest technology used in the delivery of information is the technology of Augmented Reality (AR). AR is a technology that combines three-dimensional (3D) virtual objects into a real three-dimensional environment at the same time. The 3-dimensional model is commonly used as a teaching aid to make students better understand the material provided. AR provides a more realistic interaction and is a progress of a promising technology method that can motivate users to engage in a more active learning system. Augmented Reality can be built using the help of Vuforia and Unity 3D software. In this study explains the basic concepts of making AR applications, with the Marker Based Tracking method. While the development model used is the waterfall model which consists of five phases, namely analysis, design, implementation, testing and maintenance. The results of this study are interactive learning media applications with Augmented Reality that will be used by the disabled, especially in the city of Bandung and in general in all schools in Indonesia, through their supervisors, and the results of this study serve as a comparison for other researchers in the selection of topics, studies, the method to be applied

A Methodology for Characterizing Real-Time Multimedia Quality of Service in Limited Bandwidth Network Yoanes Bandung; Idi Sumardi
TELKOMNIKA (Telecommunication Computing Electronics and Control) Vol 14, No 4: December 2016
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/telkomnika.v14i4.3295

This paper presents how to characterize the quality of multimedia which consists of audio and video that are transmitted in real-time communication through the Internet with limited bandwidth. We developed a methodology of characterizing the multimedia Quality-of-Service (QoS) by measuring network parameters (i.e., bandwidth capacity, packet loss rate (PLR), and end-to-end delay) of testbed network and simulating the audio-video delivery according to the measured network parameters. The analysis of network parameters was aimed to describe the network characteristics. Multimedia QoS was characterized by conducting a simulation using data which was collected from the previous network characterization. A simulation network model was built using OMNet++ representing a delivery of audio-video in real-time while a background traffic was generated to represent a real condition of the network. Apllying the methodology in a network testbed in Indonesia’s rural area, the simulation results showed that audio-video could be delivered with accepted level of user satisfaction.

Sistem Pakar Troubleshooting Kerusakan Hardware Laptop Dengan Metode Backward Chaining Berbasis Android henry chrystianto; Idi Sumardi
Jurnal Ilmiah Infrastruktur Teknologi Informasi Vol 2, No 1 (2021): Juni 2021
Publisher : Universitas Teknokrat Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33365/jiiti.v2i1.939

Laptop is a very helpful device for human work. Almost all fields use laptops to complete human work. Expert system technology that serves as a substitute for someone who is an expert in their field. Expert systems arise because of problems in a specific specific field, where users want a solution to the problem solved by approaching expert ways in solving problems. There are several methods in expert systems. This study will discuss the Expert System for Troubleshooting Hardware Damage on Laptops, and the inference method used is Backward chaining. Tracking is based on data or facts and then leads to conclusions in the form of conclusions about symptoms of laptop damage or problems that attack the laptop. The system that has been built is tested by comparing the suitability of the system output with the results of expert diagnoses. The test results from 15 case data obtained an accuracy rate of 86.6%. With these accuracy results, it shows that the system is able to help the role of experts

Implementasi Aplikasi e-RW untuk meningkatkan pelayanan Secara Real Time dalam mewujudkan Bandung Smart-City Diash Firdaus; Idi Sumardi
Simpatik: Jurnal Sistem Informasi dan Informatika Vol. 1 No. 1 (2021): Juni 2021
Publisher : LPPM Universitas Bina Sarana Informatika

Rukun Warga (RW) adalah pembagian wilayah di Indonesia di bawah Dusun atau Lingkungan. Rukun Warga bukanlah termasuk pembagian administrasi pemerintahan, dan pembentukannya adalah melalui musyawarah masyarakat setempat dalam rangka pelayanan kemasyarakatan yang ditetapkan oleh Desa atau Kelurahan. RW Elektronik merupakan Implementasi bidang IoT yang di terapkan di lingkungan RW. Untuk Mendapatkan informasi yang cepat dan akurat serta memberikan kemudahan dalam akses berbagai data yang berhubungan dengan RW. Pada lingkungan RW memiliki beberapa masalah seperti Masalah informasi yang beredar di RW masih berupa informasi dari media social yang belum jelas kebenarannya, pelayanan administrasi RW masih menggunakan system manual, informasi seputar RW tidak tersampikan secara realtime, masalah kemanan RW masih menggunakan system manual (Security) yang memiiliki kelemahan dalam segi fleksibilitas dan pengolahan hasil eknomi kreatif RW yang belum terdistribusi secara luas. Aplikasi RW elektronik ini dapat menjadi salah satu solusi yang dapat digunakan utuk meningkatkan pelayanan dan mempermudah dalam pengolahan data yang berhubungan dengan ruang lingkup RW, memberikan system kemanan RW menggunakan smart CCTV yang terintegrasi langsung kedalam aplikasi, memberikan informasi secara realtime memanfaatkan push notification serta Aplikasi ini dapat dijadikan sarana jual beli dengan memanfaatkan menu Toko Online RW elektronik untuk meningkatkan pendistribusian hasil ekonomi kreatif dari warga pada RW terkait

Integrating Retrieval-Augmented Generation with Large Language Model Mistral 7b for Indonesian Medical Herb Firdaus, Diash; Sumardi, Idi; Kulsum, Yuni
JISKA (Jurnal Informatika Sunan Kalijaga) Vol. 9 No. 3 (2024): September 2024
Publisher : UIN Sunan Kalijaga Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14421/jiska.2024.9.3.230-243

Large Language Models (LLMs) are advanced artificial intelligence systems that use deep learning, particularly transformer architectures, to process and generate text. One such model, Mistral 7b, featuring 7 billion parameters, is optimized for high performance and efficiency in natural language processing tasks. It outperforms similar models, such as LLaMa2 7b and LLaMa 1, across various benchmarks, especially in reasoning, mathematics, and coding. LLMs have also demonstrated significant advancements in addressing medical queries. This research leverages Indonesia’s rich biodiversity, which includes approximately 9,600 medicinal plant species out of the 30,000 known species. The study is motivated by the observation that LLMs, like ChatGPT and Gemini, often rely on internet data of uncertain validity and frequently provide generic answers without mentioning specific herbal plants found in Indonesia. To address this, the dataset for pre-training the model is derived from academic journals focusing on Indonesian medicinal herbal plants. The research process involves collecting these journals, preprocessing them using Langchain, embedding models with sentence transformers, and employing Faiss CPU for efficient searching and similarity matching. Subsequently, the Retrieval-Augmented Generation (RAG) process is applied to Mistral 7b, allowing it to provide accurate, dataset-driven responses to user queries. The model's performance is evaluated using both human evaluation and ROUGE metrics, which assess recall, precision, F1 measure, and METEOR scores. The results show that the RAG Mistral 7b model achieved a METEOR score of 0.22%, outperforming the LLaMa2 7b model, which scored 0.14%.

Peningkatan Keamanan Server GraphQL Terhadap Serangan DDOS Dengan Tipe Batch Attack Menggunakan Metode Rate Limiting Diash Firdaus; Sumardi, Idi; Nugraha, Ginanjar
Cyber Security dan Forensik Digital Vol. 7 No. 2 (2024): Edisi November 2024
Publisher : Fakultas Sains dan Teknologi UIN Sunan Kalijaga Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14421/csecurity.2024.7.2.4718

GraphQL telah memperkenalkan pergeseran paradigma tentang bagaimana aplikasi berkomunikasi dengan data, menawarkan opsi yang lebih efisien dan ampuh dibandingkan dengan RESTful API tradisional. Namun, atribut yang membuat GraphQL fleksibel dan efisien juga dapat membuatnya rentan terhadap ancaman siber yang ditargetkan, termasuk serangan batch. Eksploitasi ini memanfaatkan kemampuan untuk menggabungkan beberapa kueri atau mutasi ke dalam satu permintaan HTTP, yang dapat menyebabkan server kelebihan beban. Di berbagai industri, termasuk di Facebook, tempat kelahiran GraphQL, teknologi ini digunakan untuk menangani pertukaran data yang rumit antara aplikasi dan basis pengguna yang luas di seluruh dunia. Pembatasan kecepatan muncul sebagai penanggulangan yang tangguh terhadap ancaman serangan batch. Dengan membatasi frekuensi permintaan yang dapat dilakukan pengguna dalam interval waktu tertentu, pembatasan laju melindungi kinerja dan waktu aktif server sekaligus menggagalkan penyalahgunaan. Pendekatan ini tidak hanya membantu dalam manajemen sumber daya server yang bijaksana tetapi juga bertindak sebagai pencegah terhadap aktor jahat yang ingin memanfaatkan sistem. Data empiris mengungkapkan bahwa pembatasan laju efektif dalam mengurangi beban CPU dan Memori secara substansial, mengurangi penggunaan CPU rata-rata dari 4,8% menjadi 0,86% dan penggunaan Memori dari 87MB menjadi 49,6MB selama serangan. Sebaliknya, server tanpa pembatasan kecepatan mengalami lonjakan konsumsi CPU dan Memori setiap beberapa detik, sedangkan dengan pembatasan kecepatan, lonjakan seperti itu terbatas pada 5 detik awal. Bukti ini menggarisbawahi bahwa pembatasan kecepatan memungkinkan server untuk mempertahankan kinerja dan ketersediaan dalam menghadapi potensi serangan. Kata kunci: DdoS, GraphQL, Batch Attack ------------------------------------------------------- Abstract GraphQL has introduced a paradigm shift in how applications communicate with data, offering a more streamlined and potent option compared to traditional RESTful APIs. However, the very attributes that make GraphQL flexible and efficient can also render it vulnerable to targeted cyber threats, including batch attacks. These exploits leverage the capability to bundle multiple queries or mutations into a single HTTP request, which can lead to server overload. Across various industries, including at Facebook, the birthplace of GraphQL, this technology is employed to handle intricate data exchanges between applications and a vast user base worldwide. Rate limiting emerges as a formidable countermeasure to the threat of batch attacks. By capping the frequency of requests a user can initiate within a specified time interval, rate limiting safeguards server performance and uptime while thwarting misuse. This approach not only aids in the judicious management of server resources but also acts as a deterrent against malicious actors seeking to take advantage of the system. The empirical data reveals that rate limiting is effective in substantially reducing the strain on CPU and Memory, decreasing average CPU usage from 4.8% to 0.86% and Memory usage from 87MB to 49.6MB during an attack. In contrast, servers without rate limiting experience a surge in CPU and Memory consumption every few seconds, whereas with rate limiting, such a spike is confined to the initial 5 seconds. This evidence underscores that rate limiting enables servers to sustain performance and availability in the face of potential attacks. Keywords: DdoS, GraphQL, Batch Attack

Deteksi Serangan Pada Jaringan Internet Of Things Medis Menggunakan Machine Learning Dengan Algoritma XGBoost: Attack Detection On Internet Medical Of Things Using Machine Learning With Xgboost Algorithm Diash Firdaus; Afin, Afin; Sumardi, Idi; Chazar, Chalifa
Cyber Security dan Forensik Digital Vol. 8 No. 1 (2025): Edisi Mei 2025
Publisher : Fakultas Sains dan Teknologi UIN Sunan Kalijaga Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14421/csecurity.2025.8.1.5036

Internet of Things (IoT) telah memberikan dampak besar pada sektor kesehatan, memungkinkan pengumpulan data pasien secara real-time dan meningkatkan efisiensi layanan kesehatan. Namun, adopsi perangkat IoT medis juga membawa tantangan baru terkait keamanan, terutama serangan Distributed Denial of Service (DDoS) yang dapat mengganggu layanan kritis. Penelitian ini melakukan deteksi terhadap lima jenis serangan, yaitu ARP Spoofing, Recon Attack, MQTT Attack, TCP/IP DoS, dan DDoS, menggunakan model machine learning dengan algoritma XGBoost. Dataset yang digunakan adalah CICIoMT2024, yang dirancang khusus untuk menilai keamanan perangkat medis terhubung, melibatkan 40 perangkat IoMT. XGBoost menunjukkan performa terbaik dengan akurasi, recall, presisi, dan F1-score yang unggul, mencapai akurasi 99.8%, presisi 92.4%, recall 96%, dan F1-score 93.8%. Sebelumnya, algoritma lain seperti Logistic Regression dan Naive Bayes menunjukkan akurasi masing-masing sebesar 79% dan 92% dalam mendeteksi serangan serupa, hal ini menunjukan keterbatasan dalam menangani pola yang lebih kompleks. Hasil ini menegaskan efektivitas XGBoost dalam mendeteksi ancaman keamanan dalam ekosistem IoT medis, memberikan perlindungan lebih baik terhadap potensi gangguan pada layanan kesehatan kritis. Kata kunci: Machine Learning, Keamanan Siber, xgboost, deteksi, Internet Medical of Things ------------------------- Abstract The Internet of Things (IoT) has significantly impacted the healthcare sector, enabling real-time patient data collection and enhancing service efficiency. However, the adoption of medical IoT devices also introduces new security challenges, particularly Distributed Denial of Service (DDoS) attacks that can disrupt critical services. This study detects five types of attacks: ARP Spoofing, Recon Attack, MQTT Attack, TCP/IP DoS, and DDoS, using machine learning models with the XGBoost algorithm. The dataset used is CICIoMT2024, specifically designed to assess the security of connected medical devices, involving 40 IoMT devices. XGBoost demonstrated the best performance with superior accuracy, recall, precision, and F1-score, achieving 99.8% accuracy, 92.4% precision, 96% recall, and 93.8% F1-score. Previously, other algorithms such as Logistic Regression and Naive Bayes showed accuracies of 79% and 92% respectively in detecting similar attacks, but with limitations in handling more complex patterns. These results underscore the effectiveness of XGBoost in detecting security threats in the medical IoT ecosystem, providing enhanced protection against potential disruptions to critical healthcare services. Keywords: Machine Learning, Cybersecurity, xgboost, detection, Internet Medical of Things

Image-Based Malware Multiclass Classification Using Vision Transformer Architecture: Multiclass Klasifikasi Malware Berbasis Gambar Menggunakan Vision Transformer Architecture Diash Firdaus; Sumardi, Idi; Chazar, Chalifa; Dafy, Muhamad Zufar
Cyber Security dan Forensik Digital Vol. 8 No. 1 (2025): Edisi Mei 2025
Publisher : Fakultas Sains dan Teknologi UIN Sunan Kalijaga Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14421/csecurity.2025.8.1.5107

Perkembangan malware yang semakin canggih telah menjadi ancaman serius bagi keamanan siber global, mengakibatkan kerugian finansial yang signifikan. Metode deteksi tradisional seperti deteksi berbasis tanda tangan dan analisis dinamis memiliki keterbatasan dalam mendeteksi varian malware baru. Sebagai solusi inovatif, analisis malware berbasis gambar mengubah file biner malware menjadi representasi gambar, memanfaatkan pemrosesan citra digital dan pembelajaran mesin untuk identifikasi yang lebih efisien. Penelitian ini menggunakan arsitektur Vision Transformer (ViT) untuk klasifikasi malware multikelas berbasis gambar, menawarkan pendekatan yang lebih efektif dibandingkan CNN tradisional seperti EfficientNet dan VGG16. ViT muncul sebagai pendekatan baru yang menarik karena fleksibilitasnya dalam memahami hubungan objek dalam gambar dan mendeteksi pola penting. Dengan kemampuannya mempelajari hubungan jangka panjang antar data, ViT dapat mendeteksi perbedaan halus antara berbagai jenis malware dan mencapai akurasi lebih tinggi. Dataset yang digunakan adalah Malimg, yang merupakan hasil konversi malware biner menjadi format gambar. Hasil penelitian menunjukkan Vision Transformers mencapai akurasi pelatihan 99.96%, validasi 98.05%, dan pengujian 97.49%, meningkatkan akurasi dibandingkan CNN. Keberhasilan ini menunjukkan kemajuan signifikan dalam akurasi deteksi, mengindikasikan arah menjanjikan untuk penelitian dan aplikasi keamanan siber di masa depan. Studi ini menekankan pentingnya teknik pembelajaran mesin yang canggih untuk meningkatkan deteksi malware. Kata kunci: Vision Transformers, Klasifikasi Malware, Deep learning ------------------------- The increasing sophistication of malware has become a serious threat to global cybersecurity, resulting in significant financial losses for individuals and organizations. Traditional detection methods, such as signature-based detection and dynamic analysis, face limitations in identifying new or modified malware variants. As an innovative solution, image-based malware analysis converts malware binary files into image representations, leveraging digital image processing and machine learning for safer and more efficient identification. This study employs the Vision Transformer (ViT) architecture for multiclass image-based malware classification, offering a more effective approach compared to traditional CNNs. The Vision Transformer (ViT) has emerged as an exciting new approach, gaining attention for its flexibility in understanding object relationships within images and detecting important patterns. ViT, with its ability to learn long-range relationships between data, can detect subtle differences between various types and subtypes of malware, achieving higher classification accuracy. The results of this study show that Vision Transformers achieve the highest training accuracy of 99.96%, the highest validation accuracy of 98.05%, and a testing accuracy of 97.49%. The success of Vision Transformers in malware classification indicates significant advancements in detection accuracy, suggesting a promising direction for future research and applications in cybersecurity. This study underscores the importance of leveraging advanced machine learning techniques to enhance malware detection capabilities Keywords: Vision Transformers, Malware Classification, Deep learning

Preparation of Information Security Risk Management Based on Iso / IEC 27001: 2022 at Diskominfo West Java Province Nugraha, Ginanjar; Nurhasanah, Ina Siti; Sumardi, Idi; Nugraha, Yuda Prasetia
International Journal of Marketing & Human Resource Research Vol. 6 No. 1 (2025): International Journal of Marketing and Human Resource Research
Publisher : Training & Research Institute - Jeramba Ilmu Sukses

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47747/ijmhrr.v6i1.2596

Information security and awareness of the dangers of information leakage are the most important things in information technology, especially information that is classified and has strategic value. Information security risk management is an approach organizations use to identify, distribute, measure, and manage risks related to information security, which, if left unchecked, can paralyze existing business process activities in the organization. In carrying out its business processes, the West Java Province Diskominfo still has risk problems, namely that information security incidents often disrupt institutional business processes, where some incidents can be handled directly (reactively) in the field. However, several other incidents require planning and time. There are quite a few solutions, and there is no proper supervision and planning in managing data and information security, so Information Security Risk Management based on ISO/IEC 27001:2022 is needed. The results of this research show that there are forty-one information security risks in the West Java Province Diskominfo, and recommendations have been given for each risk in accordance with the ISO/IEC 27001:2022 standard.

Title

Found 9 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 9 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 9 Documents
Search