Article Per Year (5 Year)
p-Index From 2021 - 2026
1.383
P-Index
This Author published in this journals
All Journal Jurnal Ilmu Komputer dan Informasi Proceeding of the Electrical Engineering Computer Science and Informatics ILKOMNIKA: Journal of Computer Science and Applied Informatics Jurnal Repositor Journal of Integrated Agribusiness Literatus Innovation in Research of Informatics (INNOVATICS) JOURNAL SCIENTIFIC OF MANDALIKA (JSM) Nusantara Journal of Computers and its Applications The Indonesian Journal of Computer Science
Syadza Anggraini
Universitas Muhammadiyah Sampit
Agriculture, Biological Sciences & Forestry Humanities Biochemistry, Genetics & Molecular Biology Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Education Electrical & Electronics Engineering Engineering Environmental Science Languange, Linguistic, Communication & Media Law, Crime, Criminology & Criminal Justice Library & Information Science Social Sciences Other

Published : 12 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 12 Documents
Search

 1   2 
Enhancing YOLOv5s with Attention Mechanisms for Object Detection in Complex Backgrounds Environment Impron, Ali; Lestari, Dina; Sutriani, Linda; Anggraini, Syadza; Rizal, Randi
Innovation in Research of Informatics (Innovatics) Vol 7, No 2 (2025): September 2025
Publisher : Department of Informatics, Siliwangi University, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37058/innovatics.v7i2.16833

Abstract

Enhancing performance for object detection in complex environments is essential for real-world applications that represent complexities, such as stacking objects in the same location or environment. Models for detecting objects developed to this day still have difficulties in detecting objects with environments that have complex backgrounds. The reason is that the model often experiences a decrease in accuracy when the object to be detected is occlusion by other objects and is small in size. Therefore, in this study, a model improvement method was carried out in detecting objects in a complex environment. The algorithm used in this study is YOLOv5s. Optimization is carried out by adding a CBAM (Convolutional Block Attention Module) attention mechanism layer which is integrated with the C3 layer (C3CBAM) in the backbone of the YOLOv5s model architecture. In addition, a P2 feature map is also added to the architecture head. The optimization results carried out were quite satisfactory, namely there was an increase in the precision value by 1.6 %, at mAP@0.5 an increase of 1.4 %, and also mAP@50-95 increased by 0.1%. This proves that the enhancement method applied to YOLOv5s in this study can improve the performance of the model. However, with the addition of the attention mechanism layer, it turns out that it can increase the computational load. Therefore, for future research, a method can be applied to reduce computing load, one of the methods is knowledge distillation.
Peringkasan Hybrid Teks Berita Bahasa Indonesia Berbasis Improved TextRank dan Transformer Anggraini, Syadza; Impron, Ali; Sutriani, Linda; Putra, Kurniawan Tri
ILKOMNIKA Vol 7 No 3 (2025): Volume 7, Number 3, December 2025
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28926/ilkomnika.v7i3.806

Abstract

Peringkasan teks otomatis berbahasa Indonesia masih menghadapi tantangan dalam menghasilkan ringkasan yang informatif namun tetap koheren secara semantik. Sebagian besar penelitian sebelumnya hanya menggunakan metode ekstraktif seperti TextRank atau metode abstraktif seperti mT5-small tanpa mengoptimalkan hubungan semantik antar kalimat. Terdapat masalah di antaranya metode ekstraktif cenderung kaku dan tidak mengubah susunan kata dalam kalimat, sedangkan metode abstraktif bisa menyebabkan risiko kesalahan fakta ataupun output yang kurang relevan jika teks terlalu panjang. Untuk mengatasi masalah tersebut tersebut, penelitian ini mengusulkan metode peringkasan teks hybrid yang menggabungkan Improved TextRank dengan mT5-small. Pada tahap awal, dilakukan praproses dan ekstraksi kalimat dengan representasi semantik berbasis embedding. Hasil ekstraksi dimasukkan sebagai input di model mT5-small untuk menghasilkan ringkasan secara abstractive melalui proses parafrasa dan penyusunan ulang kalimat. Penelitian dilakukan terhadap 1000 dokumen berita dataset IndoSum dengan metrik evaluasi ROUGE. Hasil evaluasi menunjukkan bahwa metode usulan mencapai nilai ROUGE sebesar 0.687, 0.451, dan 0.634, melampaui performa TextRank klasik 0.472, 0.307, 0.441 dan mT5-Small 0.553, 0.362, 0.508 untuk hasil evaluasi ROUGE 1, 2 dan L secara berturut-turut. Hasil ini membuktikan bahwa integrasi sentence embedding dan pendekatan hybrid efektif meningkatkan kualitas ringkasan dari segi relevansi semantik. Sehingga pendekatan ini berpotensi menjadi dasar pengembangan model peringkasan teks Bahasa Indonesia yang lebih robust dan semantik.
Co-Authors Agus Zainal Arifin Chicka Willy Yanti Diana Purwitasari Dina Lestari, Dina Dini Adni Navastara, Dini Adni Gita Indah Marthasari Impron, Ali Khairiyyah Nur Aisyah Khairiyyah Nur Aisyah, Khairiyyah Nur Linda Sutriani Novi Nur Putriwijaya Nur Hayatin Putra, Kurniawan Tri Rarasmaya Indraswari Riantri Barus, Riantri Rizal, Randi suraji suraji Veny Betsy Saragih