Article Per Year (5 Year)
p-Index From 2020 - 2025
1.244
P-Index
This Author published in this journals
All Journal JMM (Jurnal Masyarakat Mandiri) Abdi Masyarakat Jurnal Algoritma eProceedings of Engineering Scientific Journal of Informatics
Muhajir, Daud
Unknown Affiliation
Humanities Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Education Electrical & Electronics Engineering Energy Engineering Industrial & Manufacturing Engineering Social Sciences Other

Published : 8 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 8 Documents
Search

 1 
Development of a Village Citizen Reporting Website for Smart Village Village Public Services Monika Dian Pertiwi, Kharisma; Muhajir, Daud; Ananda, Dahliar; Santik, Tita Arum Shela; Pratama, Moch. Andi Divangga; Adityo, Kahil Akbar Bayu; Jungjungan, Fadhlan Syahran
Abdi Masyarakat Vol 6, No 2 (2024): Abdi Masyarakat
Publisher : Lembaga Penelitian dan Pendidikan (LPP) Mandala

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58258/abdi.v6i2.7491

Abstract

The development of technology that has been supported by the government makes it easier for citizens to get information, and can influence society to get information and participate in public policy. The development of technology carried out by the government starts from public services on digital platforms, such as websites or village applications. Smart Village initiated by Telkom University Surabaya innovates by providing a website to help citizens with easy complaints to providing a space for community aspirations that can be accessed anywhere and anytime by the people of Panjunan Village. The Public Complaints Website has 31 features for citizens, officers and admins, these features have been tested functionally, and have been tested on several users, in the test no functional errors were found in the sense that the system is ready to use.
A Deep Learning Model Comparation for Diabetic Retinopathy Image Classification Mustaqim, Tanzilal; Safitri, Pima Hani; Muhajir, Daud
Scientific Journal of Informatics Vol. 12 No. 1: February 2025
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v12i1.20939

Abstract

Purpose: This study compares the performance of various deep learning models for diabetic retinopathy (DR) classification, emphasizing the impact of different optimization functions. Early detection of DR is vital for preventing blindness, and the research investigates how optimization functions influence the classification accuracy and efficiency of several convolutional neural networks (CNNs). This study fills a gap in the existing literature by examining how optimization functions affect model performance in conjunction with architectural considerations. Methods: This paper uses the APTOS 2019 dataset, which comprises 3,663 retinal fundus images classified into five classes of diabetic retinopathy severity. Four CNN-based models, including CNN, ResNet50, DenseNet121, and EfficientNet B0, were trained using five optimization techniques: Adam, SGD, RMSProp, AdamW, and NAdam. The performance of the experimental scenarios was evaluated through accuracy, precision, recall, F1-score, training duration, and model size. Result: EfficientNet B0 demonstrated superior computational efficiency with a minimal model size of 16.16 MB. Subsequently, DenseNet121 with the SGD optimizer achieved the highest test accuracy of 96.86%. The experimental results indicate that the optimizer significantly influences model performance. AdamW and NAdam yield superior outcomes for deeper architectures such as ResNet50 and DenseNet121. Novelty: This paper offers an analytical examination of deep learning models and optimization techniques for DR classification, helping to clarify the trade-offs between computational efficiency and classification performance. The findings contribute to the development of more accurate and efficient DR detection systems, which could be utilized in real-world, resource-limited settings.
Rancang Bangun Prototipe Sistem Deteksi Dini Retinopathic Diabetic Berbasis Website Muhajir, Daud; Mustaqim, Tanzilal; Safitri, Pima Hani; Oktavia, Vessa Rizky
Jurnal Algoritma Vol 22 No 1 (2025): Jurnal Algoritma
Publisher : Institut Teknologi Garut

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33364/algoritma/v.22-1.2255

Abstract

Diabetic Retinopathic (DR) is one of the retinal disorders caused by high blood sugar levels. There are fewer ophthalmologists available, and treating DR patients manually is a time-consuming process. Therefore, there is a need for an automatic DR early detection method using Deep Learning. The purpose of this research is to build a web-based DR early detection prototype with retinal image classification using the DenseNet121 Deep Learning model and the Stochastic Gradient Descent (SGD) optimizer to improve the accessibility and efficiency of screening. The software development method used in this research is waterfall which consists of analysis phase, design phase, implementation phase, and testing phase. To ensure the prototype runs as planned, black-box testing is carried out on each of its features to ensure system functionality in accordance with predetermined specifications. This research produces a RD early detection prototype that has been tested with all 16 test cases and has a suitable status. Future research can be carried out further system development by involving real users such as ophthalmologists and can be applied in hospitals.
PELATIHAN DAN IMPLEMENTASI APLIKASI ’KONSELINK’ UNTUK TRANSFORMASI DIGITAL LAYANAN BIMBINGAN KONSELING Hidayati, Sri; Rosidah, Nur Azizah; Muhajir, Daud; Kusumawati, Aris; Prisyanti, Affifiana; Abdillah, Rosyid; Alhari, Muhammad Ilham; Hendradi, Fransisca Aurelia Maranatha; Satriawan, Muhammad Paksi; Ramadhan, Muhammad Ayondi; Septiano, Dino Rossi Eka
JMM (Jurnal Masyarakat Mandiri) Vol 9, No 4 (2025): Agustus
Publisher : Universitas Muhammadiyah Mataram

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31764/jmm.v9i4.32709

Abstract

Abstrak: Layanan bimbingan konseling (BK) di jenjang sekolah menengah masih banyak yang dilakukan secara manual menggunakan buku pengembangan diri, yang menyebabkan kendala dalam pencatatan, analisis data, dan pemberian intervensi. Tujuan kegiatan pengabdian ini adalah mentransformasi layanan BK konvensional menjadi sistem digital melalui pengembangan dan pelatihan penggunaan aplikasi web “Konselink”. Metode yang digunakan mencakup observasi kebutuhan, pengembangan sistem, pelatihan luring, serta evaluasi. Mitra kegiatan ini adalah guru BK dan siswa di SMAN 1 Kamal. Pelatihan dilakukan di Lab TIK sekolah dan dievaluasi menggunakan kuesioner serta observasi langsung, dengan peserta pelatihan sebanyak 22 guru BK dan perwakilan siswa. Hasil kegiatan menunjukkan peningkatan keterampilan digital guru, serta skor kepuasan peserta rata-rata 4,9 dari 5. Aplikasi Konselink dinilai efektif dalam meningkatkan efisiensi pencatatan, mempermudah analisis perkembangan siswa, dan mendorong keterlibatan siswa dalam layanan konseling yang lebih fleksibel dan adaptif.Abstract: Guidance and counseling (GC) services at the secondary school level are still largely carried out manually using student development logbooks, which leads to challenges in record-keeping, data analysis, and timely intervention. The aim of this community engagement initiative was to transform conventional GC services into a digital system through the development and training in the use of a web-based application called “Konselink.” The methods employed included needs assessment, system development, offline training, and evaluation. The participants in this program consisted of guidance counselors and students. The training was conducted in the school's ICT laboratory and evaluated through questionnaires and direct observation, involving 22 participants comprising GC teachers and student representatives. The results showed an improvement in the digital skills of the teachers, with an average participant satisfaction score of 9.4 out of 10. The Konselink application was found to be effective in improving the efficiency of documentation, facilitating student development analysis, and encouraging more flexible and adaptive student engagement in counseling services.
Analisis Perbandingan Metode Preprocessing untuk Citra Retinopati Diabetik Menggunakan Deep learning Safitri, Pima Hani; Mustaqim, Tanzilal; Muhajir, Daud
Jurnal Algoritma Vol 22 No 2 (2025): Jurnal Algoritma
Publisher : Institut Teknologi Garut

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33364/algoritma/v.22-2.2324

Abstract

Diabetic retinopathy is a symptom caused by complications of diabetes that attack the eyes of sufferers. Spots on the sufferer’s retina are characteristic of the symptoms. The more spots, the more severe the diabetic retinopathy suffered. Researchers’ efforts to detect diabetic retinopathy with retinal images have begun to be developed with artificial intelligence technology, one of which is based on deep learning. The next difficulty is the poor quality of the retinal image, resulting in poor detection results. Therefore, this study proposes a comparative analysis of techniques to improve image processing accuracy for deep learning-based diabetic retinopathy detection. The data used is APTOS2019 data, which consists of 5 classes based on the severity of the disease. There are three techniques used: CLAHE, gamma correction, and Retinex. The deep learning architecture used is DenseNet121 and EfficientNetB0 because it has been widely used in medical image data. As a result, the combination of gamma correction and DenseNet121 produces the highest accuracy of 81.4%. While the lowest accuracy is obtained from the combination using Retinex. The best overall architecture is EfficientNetB0, with an average accuracy of 81.9%. Furthermore, this study can be used to improve diabetic retinopathy images so that detection can be done as early as possible.
Implementasi Prediksi Ketersediaan Stok Penjualan Di Koperasi Hita Loka Tara Dengan Metode ARIMA Agustina , Yayuk; Fenaldo Maulana, Rizky; Muhajir, Daud
eProceedings of Engineering Vol. 12 No. 5 (2025): Oktober 2025
Publisher : eProceedings of Engineering

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

Abstract

Abstrak — Koperasi Hita Loka Tara merupakan koperasi mahasiswa Universitas Telkom Surabaya yang menyediakan berbagai kebutuhan harian seperti makanan, minuman, dan alat tulis. Sistem manajemen stok yang masih manual menyebabkan rendahkan efesiensi dan ketepatan perencanaan persediaan. Untuk mengatasi hal tersebut, penelitian ini mengimplementasikan metode ARIMA (AutoRegressive Integrated Moving Average), yaitu metode statistik yang digunakan untuk menganalisis serta memprediksi data deret waktu berdasarkan pola historis. ARIMA menggabungkan komponen Autoregresi (AR), Differencing (I), dan Moving Average (MA) dalam membentuk pola estimasi nilai masa depan. Metode ini diterapkan dalam sistem prediksi penjualan berbasis web untuk mendukung pengambilan keputusan berdasarkan data yang tersedia. Data penjualan selama 50 minggu dari lima kategori produk terlaris diproses melalui tahap praproses, uji stasioneritas (ADF), analisis ACF dan PACF, pemodelan arima, serta evaluasi menggunakan MAPE dan RMSE. Hasil prediksi disajikan dalam dashboard web interaktif untuk memudahkan pemantauan. Hasil evaluasi menunjukkan bahwa kategori Minuman memiliki performa terbaik dengan MAPE 13,35% dan RMSE 147,63, sedangkan kategori AICE menunjukkan akurasi terendah dengan MAPE 75,21% dan RMSE 47,75 akibat fluktuasi data yang tinggi. Secara keseluruhan, arima efektif dalam memprediksi stok, namun kurang optimal pada data yang sangat fluktuatif. ata kunci—ARIMA, prediksi stok penjualan, sistem berbasis web, manajemen koperasi, MAPE, RMSE.K
Normalisasi Komentar Media Sosial Pasangan Calon Gubernur 2024 Dengan Statistical Machine Translation Akbar Bayu Adityo , Kahil; Rausanfita, Alqis; Muhajir, Daud
eProceedings of Engineering Vol. 12 No. 5 (2025): Oktober 2025
Publisher : eProceedings of Engineering

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

Abstract

Abstrak — Tingginya aktivitas masyarakat dalam membahas pemilihan Gubernur melalui media sosial menghasilkan data komentar dalam jumlah besar, namun komentar tersebut sering menggunakan bahasa informal, bahasa sehari-hari, singkatan, serta bercampur dengan bahasa daerah dan dialek lokal yang sulit dipahami. Hal ini menghambat pemrosesan data komentar untuk keperluan analisis atau tujuan lainnya. Proses normalisasi manual membutuhkan waktu dan sumber daya yang sangat banyak, terutama jika data yang diolah berjumlah besar. Normalisasi secara manual juga rentan terhadap inkonsistensi dan kesalahan manusia. Jumlah data komentar di media sosial yang terus meningkat membuat normalisasi manual semakin tidak mungkin dan tidak efisien untuk dilakukan, sehingga diperlukan solusi otomatisasi. Sistem normalisasi teks otomatis dikembangkan menggunakan pendekatan Phrase-Based Statistical Machine Translation dengan memanfaatkan Moses. Dataset korpus paralel dibangun dari 31.889 pasangan kalimat informal-formal, sedangkan korpus monolingual terdiri dari 1.613.381 kalimat yang diambil dari Wikipedia. Model dievaluasi menggunakan metrik BLEU untuk mengukur kualitas hasil normalisasi. Model terbaik mencapai skor BLEU 82,16 pada data test dan 81,04 pada data validasi, berhasil mengenali berbagai pola bahasa informal seperti singkatan tidak baku, kata berulang dengan angka, dan bahasa gaul. Namun, sistem memiliki keterbatasan terhadap kemampuan penanganan Out-Of-Vocabulary. Kata kunci— normalisasi teks, PBSMT, Moses, media sosial, gubernur
Pengembangan Aplikasi Android Untuk Deteksi Dan Klasifikasi Motif Batik Berbasis Convolutional Neural Network (CNN) Haikal Fikri As’ad , Muhammad; Yusuf Wicaksono, Ardian; Muhajir, Daud
eProceedings of Engineering Vol. 12 No. 5 (2025): Oktober 2025
Publisher : eProceedings of Engineering

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

Abstract

Abstrak — Batik merupakan warisan budaya Indonesia yang kaya akan motif dan nilai filosofis. Namun, masyarakat masih mengalami kesulitan dalam mengenali motif batik akibat kompleksitas visual dan keterbatasan akses informasi. Penelitian ini bertujuan mengembangkan aplikasi Android bernama Batikara yang mampu mendeteksi dan mengklasifikasikan motif batik secara otomatis menggunakan metode Convolutional Neural Network (CNN) dengan arsitektur MobileNetV2, serta memberikan informasi edukatif melalui artikel batik. Model klasifikasi MobileNetV2 memperoleh akurasi tertinggi sebesar 92%, diikuti oleh DenseNet121 dengan akurasi 90%, serta ResNet50 dengan akurasi 77%. MobileNetV2 menunjukkan performa yang paling stabil dan efisien untuk implementasi pada perangkat mobile, serta memiliki kapabilitas yang unggul dalam mengklasifikasikan motif batik yang memiliki kemiripan visual dan pola yang kompleks. Sebaliknya, ResNet50 cenderung kurang optimal dalam membedakan motif-motif batik yang serupa secara visual. Untuk deteksi objek, model SSD MobileNetV2 mencatat nilai Average Precision (AP) tertinggi 0,722 pada IoU 0,50, meskipun performa menurun pada objek kecil. Evaluasi usability melalui System Usability Scale (SUS), black box testing, dan task scenario menghasilkan skor SUS sebesar 83,17, efektivitas 88,9%, dan efisiensi 77,16%. Responden, termasuk pelaku UMKM, menilai fitur pemindaian sangat membantu dan edukatif. Aplikasi Batikara berpotensi menjadi sarana pelestarian budaya sekaligus memberdayakan UMKM batik melalui teknologi digital. Kata kunci— Aplikasi Android, Batik, CNN, Klasifikasi Citra, MobileNetV2, TensorFlow Lite
Co-Authors Adityo, Kahil Akbar Bayu Affifiana Prisyanti Agustina , Yayuk Akbar Bayu Adityo , Kahil Alqis Rausanfita Ardian Yusuf Wicaksono Aris Kusumawati Dahliar Ananda Fenaldo Maulana, Rizky Haikal Fikri As’ad , Muhammad Hendradi, Fransisca Aurelia Maranatha Jungjungan, Fadhlan Syahran Muhammad Ilham Alhari Pertiwi, Kharisma Monika Dian Pratama, Moch. Andi Divangga Ramadhan, Muhammad Ayondi Rosidah, Nur Azizah Rosyid Abdillah Safitri, Pima Hani Santik, Tita Arum Shela Satriawan, Muhammad Paksi Septiano, Dino Rossi Eka Sri Hidayati Tanzilal Mustaqim Vessa Rizky Oktavia