Article Per Year (5 Year)
p-Index From 2021 - 2026
0.444
P-Index
This Author published in this journals
All Journal INFOKUM Jurnal Pengembangan Teknologi Informasi dan Komunikasi (JUPTIK)
Muhammad Irsyad
Universitas Islam Negeri Sultan Syarif Kasim
Computer Science & IT Control & Systems Engineering Engineering Library & Information Science

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

Found 2 Documents
Search

 1 
UI/UX Redesign of INHIL Dukcapil Application Using the Design Thinking Method Rosi Yulia; Reski Mai Candra; Muhammad Irsyad; Teddie Darmizal
INFOKUM Vol. 10 No. 5 (2022): December, Computer and Communication
Publisher : Sean Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

The population and civil registration office is an agency directly related to the city government that is tasked with providing population registration services and civil registration services. Now technological advances are helping to modernize public services, including population administration services in Riau Province. There are 3 districts that have implemented services using smartphone applications. One of them is Indragiri Hilir Regency with the name Dukcapil Inhil Application. The Dukcapil Inhil application was released to make it easier for people to take care of population administration. However, in its use there are several problems. This research was conducted to develop application recommendations by analyzing and redesigning the UI and UX of the Dukcapil Inhil Application using the Design Thinking method. This method is used to take a solution-based approach that will be used for problem solving, by understanding application users to define existing problems and then make a solution to solve existing problems. Application recommendations are built in the form of prototypes using the Figma application. The prototype built was successfully tested on 10 respondents using System Usability Scale (SUS) testing and the application prototype was rated 78.
Penggunaan Dual Attention Network pada DenseNet-169 untuk Klasifikasi Multi-kelas Citra X-Ray Dada Azizah Tasykira Paramitha El Razi; Benny Sukma Negara; Muhammad Irsyad; Suwanto Sanjaya; Siti Ramadhani
Jurnal Pengembangan Teknologi Informasi dan Komunikasi (JUPTIK) Vol. 4 No. 1 (2026): JURNAL PENGEMBANGAN TEKNOLOGI INFORMASI DAN KOMUNIAKSI (JUPTIK)
Publisher : Universitas Muhammadiyah Muara Bungo

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52060/juptik.v4i1.4333

Abstract

Klasifikasi multi-kelas citra X-ray dada menjadi salah satu pendekatan yang dapat membantu membedakan kondisi COVID-19, normal, dan pneumonia secara otomatis. Namun, kemiripan karakteristik visual antar kelas dapat menyebabkan model kesulitan dalam mengekstraksi fitur yang relevan. Penelitian ini mengintegrasikan Dual Attention Network (DANet) pada DenseNet-169 untuk meningkatkan representasi fitur melalui kombinasi Grouped Channel Attention Module dan Strip Spatial Attention Module. Dataset yang digunakan terdiri atas 5.228 citra X-ray dada yang dibagi menjadi data latih dan data uji dengan rasio 80:20. Model DenseNet-169 baseline dan DenseNet-169 dengan DANet dievaluasi menggunakan akurasi, presisi, recall, F1-score, sensitivitas, spesifisitas, ROC-AUC, confusion matrix, dan Grad-CAM. Hasil pengujian menunjukkan bahwa DenseNet-169 dengan DANet memperoleh akurasi 98,47%, presisi 98,55%, recall 98,50%, F1-score 98,50%, sensitivitas 98,50%, dan spesifisitas 99,22%. Nilai ROC-AUC yang diperoleh pada kelas COVID-19, normal, dan pneumonia masing-masing sebesar 0,9994, 0,9989, dan 0,9980. Hasil tersebut menunjukkan bahwa DenseNet-169 dengan DANet memiliki kemampuan klasifikasi dan diskriminasi yang baik pada ketiga kelas. Visualisasi Grad-CAM menunjukkan bahwa DANet membantu model menghasilkan perhatian yang lebih terarah.
Co-Authors Azizah Tasykira Paramitha El Razi Negara, Benny Sukma Reski Mai Candra Rosi Yulia Siti Ramadhani Suwanto Sanjaya Teddie Darmizal