Article Per Year (5 Year)
p-Index From 2021 - 2026
4.691
P-Index
This Author published in this journals
All Journal Semesta Teknika Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) Insect (Informatics and Security) : Jurnal Teknik Informatika ILKOM Jurnal Ilmiah JURIKOM (Jurnal Riset Komputer) Systemic: Information System and Informatics Journal Jurnal Pengabdian Dharma Laksana JURNAL MAHAJANA INFORMASI Jurnal Ilmiah Betrik : Besemah Teknologi Informasi dan Komputer Respati Jurnal Tekinkom (Teknik Informasi dan Komputer) Bubungan Tinggi: Jurnal Pengabdian Masyarakat JITU : Journal Informatic Technology And Communication Jurasik (Jurnal Riset Sistem Informasi dan Teknik Informatika) Abdimas: Papua Journal Of Community Service Jurnal Pengabdian Masyarakat Journal of Community Empowerment FRAMEWORK Jurnal Ilmu Komputer dan Informatika Jurnal Keperawatan Sehat Mandiri
Muhammad Rizki Setyawan
Program Studi Teknik Informatika, Universitas Muhammadiyah Sorong
Humanities Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Economics, Econometrics & Finance Education Electrical & Electronics Engineering Engineering Environmental Science Health Professions Languange, Linguistic, Communication & Media Law, Crime, Criminology & Criminal Justice Library & Information Science Nursing Public Health Social Sciences Other

Published : 34 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 34 Documents
Search

 1   2   3   4 
Sentiment Analysis towards Jokowi Post-Presidential Term Using CNN-BiLSTM with Multi-head Attention on Platform X Setyawan, Muhammad Rizki; Putra, Fajar Rahardika Bahari; Ramadhani, Ardhina
ILKOM Jurnal Ilmiah Vol 17, No 2 (2025)
Publisher : Prodi Teknik Informatika FIK Universitas Muslim Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33096/ilkom.v17i2.2843.150-161

Abstract

The development of social media has changed the way the public expresses political opinions, especially regarding the evaluation of President Joko Widodo’s (Jokowi) leadership after his term. Platform X (formerly Twitter) has become the primary source of public opinion data, but the use of informal language and sarcasm makes accurate sentiment analysis challenging. This study creates a sentiment analysis model that uses deep learning with a CNN-BiLSTM structure and a multi-head attention mechanism. The dataset consists of 52,643 tweets that have been labeled and embedded using IndoBERT. To address class imbalance, the SMOTE method was applied to the training data, enabling the model to better learn from minority classes. The results indicate that the model achieves a high accuracy of 98.78%, with an average precision, recall, and F1-score of 0.98. These findings indicate that the model is not only accurate but also reliable in distinguishing each sentiment class. A comparison with other model variants suggests that the complete combination of CNN-BiLSTM and Multi-Head Attention delivers the best performance, although the improvement is relatively small.
SOSIALISASI DAN PELATIHAN PENGGUNAAN LMS BERBASIS MOODLE PADA YAYASAN SEKOLAH ADVENT DI SAUSAPOR Sundari Sundari; Muhammad Syahrul Kahar; Zulkarnain Sangadji; Muhammad Rizki Setyawan; Ahmad Ilham; Ihsan Febriadi
Journal of Community Empowerment Vol 4, No 2 (2025): September
Publisher : Universitas Muhammadiyah Mataram

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31764/jce.v4i2.34137

Abstract

ABSTRAK                                                                 Transformasi digital menuntut peningkatan literasi digital guru, khususnya dalam pemanfaatan Learning Management System (LMS). Di Distrik Sausapor, Papua Barat Daya, meskipun infrastruktur jaringan memadai, keterampilan guru dalam menggunakan LMS berbasis Moodle masih terbatas. Kegiatan pengabdian ini dilaksanakan bersama Yayasan Pendidikan Advent Sausapor sebagai mitra, melibatkan 25 guru dari jenjang SD, SMP, dan SMA pada Juli 2025. Metode yang digunakan adalah Participatory Action Research (PAR) melalui empat tahap: (1) perencanaan, meliputi koordinasi dengan mitra untuk menentukan kebutuhan, peserta, dan jadwal; (2) persiapan, penyusunan modul pelatihan serta instrumen pre-test dan post-test dengan dukungan fasilitas komputer dan internet sekolah; (3) pelaksanaan, berupa dua hari kegiatan sosialisasi dan praktik pembuatan kelas virtual, unggah materi ajar, penyusunan kuis, forum diskusi, dan sistem penilaian otomatis; (4) evaluasi, menggunakan pre-test dan post-test. Hasil menunjukkan peningkatan rata-rata sebesar 74,1% pada seluruh aspek pengetahuan dan keterampilan. Temuan ini menegaskan bahwa peserta tidak hanya memahami konsep dasar Moodle, tetapi juga mampu mengaplikasikannya secara efektif dalam proses pembelajaran. Kegiatan ini memberikan kontribusi nyata dalam memperkuat kompetensi guru serta menjadi langkah awal membangun ekosistem pembelajaran digital yang inklusif dan adaptif. Kata kunci: LMS Moodle; literasi digital; pembelajaran digital; pelatihan guru; PAR ABSTRACTThe digital transformation demands an increase in teachers’ digital literacy, particularly in utilizing Learning Management Systems (LMS). In Sausapor District, Southwest Papua, although network infrastructure is adequate, teachers’ skills in using Moodle-based LMS remain limited. This community service program was carried out in collaboration with the Sausapor Advent Education Foundation as a partner, involving 25 teachers from elementary, junior high, and senior high schools in July 2025. The method employed was Participatory Action Research (PAR) through four stages: (1) planning, including coordination with partners to determine needs, participants, and schedule; (2) preparation, which involved developing training modules and pre-test and post-test instruments supported by the school’s computer and internet facilities; (3) implementation, consisting of two days of activities such as socialization, virtual class creation, digital material uploads, quiz development, discussion forums, and automatic grading systems; (4) evaluation, using pre-tests and post-tests. The results indicated an average improvement of 74.1% across all aspects of knowledge and skills. These findings confirm that participants not only understood the basic concepts of Moodle but were also able to apply them effectively in the learning process. This activity provides a tangible contribution to strengthening teachers’ competencies and serves as an initial step in building an inclusive and adaptive digital learning ecosystem.. Keywords: LMS Moodle; digital literacy; digital learning; teacher training; PAR
Perancangan Electronic Nose (E-Nose) untuk Analisis dan Klasifikasi Aroma Daging Menggunakan PCA dan LDA Muhammad Rizki Setyawan; Abdul Fadlil; Anton Yudhana
JITU Vol 10 No 1 (2026)
Publisher : Universitas Boyolali

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36596/jitu.v10i1.2254

Abstract

Meat is a vital food commodity prone to adulteration through species mixing or chemical contamination such as formalin and borax. This study aimed to design and test an Electronic Nose (E-Nose) system for aroma pattern analysis and meat classification using Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA). Samples included pure meat (beef, chicken, pork), mixed meat, and chemically contaminated meat. Aroma data were captured using an array of gas sensors sensitive to Volatile Organic Compounds (VOCs) and standardized prior to analysis. PCA reduced eight sensor features into three principal components explaining a total variance of 79.63%. PC1, PC2, and PC3 accounted for 46.10%, 20.58%, and 12.96% of variance, respectively, showing clustering patterns among samples with minor overlap. LDA provided clearer class separation with three discriminant components LD1, LD2, and LD3 explaining 77.13%, 16.63%, and 4.59% of between-class variance, totaling 98.34%. LD1 separated pure, mixed, and contaminated meat, LD2 distinguished variations due to contaminant type and species, and LD3 refined separation of similar classes. Classification evaluation achieved an overall accuracy of 82%. Most classes were well classified, while classes 1 and 10 experienced misclassification due to similar aroma patterns. The findings confirm that E-Nose combined with PCA and LDA is a rapid, non-destructive, and efficient method for detecting meat authenticity and adulteration, showing strong potential for food quality monitoring in the field
Implementasi Algoritma YOLOv12 Untuk Deteksi Tingkat Kesiapan Panen Kangkung Potong Secara Realtime Berbasis Android Muhammad Rizki Setyawan; Fajar Rahardika; Suhardi Aras; Dimas Adi Suseno; Andini Setyaningsi
Insect (Informatics and Security): Jurnal Teknik Informatika Vol. 12 No. 2 (2026): Oktober 2026
Publisher : Universitas Muhammadiyah Sorong

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33506/insect.v12i2.5910

Abstract

Penentuan tingkat kesiapan panen kangkung potong umumnya masih dilakukan secara manual melalui pengamatan visual petani yang bersifat subjektif sehingga rentan menimbulkan ketidakkonsistenan dan memengaruhi kualitas serta nilai jual hasil panen. Penelitian ini bertujuan mengimplementasikan algoritma YOLOv12 untuk mendeteksi tingkat kesiapan panen kangkung potong ke dalam dua kelas, yaitu siap panen dan belum panen, secara real-time pada perangkat Android. Dataset terdiri atas 3.556 citra yang dikumpulkan dari repositori publik, kemudian dianotasi, dipra-pemrosesan, dan dibagi dengan proporsi 70:20:10, dengan augmentasi diterapkan hanya pada subset data latih. Model YOLOv12s dilatih selama 100 epoch pada Google Colab, lalu dikonversi ke format ONNX agar dapat dijalankan langsung di perangkat melalui ONNX Runtime. Hasil evaluasi menunjukkan model memperoleh precision 0,841, recall 0,888, mAP50 0,914, dan mAP50-95 0,539, dengan performa terbaik pada kelas siap panen (recall 0,990). Sistem berhasil diimplementasikan menjadi aplikasi Android Kangkung Detector dan diuji menggunakan black box testing yang seluruh fungsinya berjalan sesuai harapan, serta usability testing terhadap 25 responden dengan rata-rata penerimaan 92,08%. Dengan demikian, YOLOv12 berhasil diimplementasikan pada perangkat Android untuk mendeteksi tingkat kesiapan panen kangkung potong dengan akurasi yang baik pada data uji, sehingga berpotensi membantu petani menentukan waktu panen secara lebih objektif dan konsisten.
Co-Authors Abdul Fadlil Abdul Fadlil Ahistasari, Asih Ahmad Ilham Ahmad Ilham Amri, Irman Andini Setyaningsi Anton Yudhana Aronggear, Karim Paul Bahari Putra, Fajar Rahardika Bekti Wiji Lestari Dewi Astria Faroek Dimas Adi Suseno Elisabeth Serlina Elensia Bupu Ermin Ermin Ermin, E Fajar Rahardika Fajar, Fajar Rahardika Bahari Putra Fakhri , La Jupriadi Fitriyani Tella Hermansa Hermansa Hilda Histiarini, Aprisa Rian Ilham, Ahmad Kahar, Muhammad Syahrul Kayatun, Siti Nur La Jupriadi Fakhri Marlinda Indah Eka Budiarti, Marlinda Indah Eka Mirga Maulana Rachmadhani Mohammad Arief Nur Wahyudien Mohammad Arief Nur Wahyudien Muh. Fadli Hasa Muhammad Jundullah Muzakkir Pangri Nabila Novita Sari Nurul Widya Har Putra, Fajar Rahardika Bahari Rachmadanur Rachmadanur Rahardika Bahari Putra, Fajar Rahmatullah Bin Arsyad Ramadhani, Ardhina Rawi, Rais Dera Pua Reflita Reflita Rinanda Tri Setiawan Sangadji, Zulkarnain Simori, Pascalina Magrice Slameto, Andika Agus Soekarta, Rendra Soleha Rahayu Muklis Hayu Suhardi Aras Sundari Sundari Surahmanto Suseno, Dimas Adi Tasman Tasman Teguh Hidayat Iskandar Alam Tella, Fitriyani tivani nurizqi Umar Rusli Marasabessy Verra Widhi Astuti Wattimena, Mikhael