Afina Lina Nurlaili
Universitas Pembangunan Nasional “Veteran” Jawa Timur

Published : 5 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 5 Documents
Search

Implementasi Process Mining pada Proses Praktik Kerja Lapangan (PKL) Afina Lina Nurlaili; - Muhsin; Eristya Maya Safitri
Techno.Com Vol 20, No 4 (2021): November 2021
Publisher : LPPM Universitas Dian Nuswantoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33633/tc.v20i4.5256

Abstract

Proses yang tidak terdokumentasi dapat menjadi masalah apabila tidak ada suatu prosedur yang mendasarinya, khususnya pada proses yang melibatkan banyak aktivitas. Prosedur dibutuhkan untuk mengatur kegiatan dalam mencapai tujuan tertentu. Dalam hal memenuhi tujuan tersebut, dibuatlah panduan berupa Standar Operasional Prosedur (SOP) yang dapat memberikan manfaat bagi organisasi mana pun yang menerapkannya. SOP juga menjadi dasar dalam proses evaluasi antara kenyataan di lapangan dengan prosedur yang telah dibuat ke dalam SOP sehingga diperlukan suatu teknik untuk mengevaluasi antara SOP dengan kenyataan. Process mining merupakan teknik evaluasi antara suatu model proses dengan data peristiwa atau event log yang terdapat dalam sistem informasi. Berbagai metode process mining telah dikenalkan yaitu algoritma Alpha dan Heuristic Miner. Algoritma Alpha dan Heurisctic pada penelitian ini akan dihitung nilai fitness dan presisi. Fitness dilakukan dengan mengukur kesesuaian antara event log dan model proses. Presisi mengukur apakah suatu algoritma tepat untuk menyelesaikan kasus tertentu. Berdasarkan evaluasi yang dilakukan, algoritma Alpha tidak mampu menggambarkan proses sesuai dengan event log PKL. Hal ini disebabkan karena varian kasus mengandung proses loop/perulangan. Hal ini juga menunjukkan event log yang ada pada proses PKL belum menerapkan SOP PKL. Sedangkan Heuristic Miner mengabaikan proses minor menyebabkan proses-proses yang tidak banyak terjadi, tidak digambarkan ke dalam model proses. Secara keseluruhan proses model yang terbentuk menggunakan algoritma Alpha yang paling mendekati dengan kenyataan karena memiliki fitness 0,96.
Optimization of Earthquake B-Value Prediction in Java Using GRU and Particle Swarm Optimization Kesya Nursyahada; Basuki Rahmat; Afina Lina Nurlaili
bit-Tech Vol. 8 No. 1 (2025): bit-Tech
Publisher : Komunitas Dosen Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32877/bt.v8i1.2521

Abstract

Accurate prediction of earthquake parameters is essential for seismic risk assessment and disaster mitigation, particularly in tectonically active regions such as Java Island, Indonesia. This study presents a novel predictive model for estimating the earthquake b-value a fundamental seismological parameter representing the logarithmic relationship between earthquake frequency and magnitude by integrating a Gated Recurrent Unit (GRU) neural network with Particle Swarm Optimization (PSO). The model is trained using earthquake catalog data from 1962 to 2024, sourced from the Indonesian Meteorological, Climatological, and Geophysical Agency (BMKG). The GRU architecture is selected for its effectiveness in modeling temporal dependencies in seismic time series data. PSO is employed to optimize essential hyperparameters, including the number of GRU units, learning rate, and dropout rate. The optimized model achieves notable improvements in predictive performance: Mean Squared Error (MSE) is reduced from 0.00435 to 0.00030, Root Mean Squared Error (RMSE) from 0.0509 to 0.0173, and Mean Absolute Percentage Error (MAPE) from 3.42% to 1.12%. Training time is also reduced from 57 seconds to 33 seconds, indicating greater computational efficiency. The optimal PSO settings include an inertia weight of 0.8, cognitive and social coefficients of 1.0, 40 particles, and 10 iterations. The primary novelty of this study lies in its targeted application of PSO-optimized GRU architecture for b-value prediction in a seismically complex region. These results demonstrate that evolutionary optimization significantly enhances deep learning performance, providing a robust and efficient framework to support earthquake forecasting and risk mitigation efforts in high-risk zones such as Java Island.
Comparative Analysis of IndoBERT, IndoBERTweet, and XLM-RoBERTa for Detecting Online Gambling Comments on YouTube Kevin Iansyah; Afina Lina Nurlaili; Muhammad Muharrom Al Haromainy
bit-Tech Vol. 8 No. 2 (2025): bit-Tech
Publisher : Komunitas Dosen Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32877/bt.v8i2.3257

Abstract

The proliferation of online gambling promotions in YouTube comment sections poses significant challenges for content moderation on Indonesian digital platforms. Although transformer models have proven effective for various Indonesian-language NLP tasks, no systematic comparative evaluation exists for detecting online gambling promotions on YouTube, nor has research explored model sensitivity to hyperparameters in this context. This research identifies the optimal transformer model and configuration for detecting Indonesian-language online gambling promotion comments on YouTube. A total of 26,455 YouTube comments were collected from February to July 2025 and stratified into balanced training (18,926 comments) and validation sets (3,340 comments), plus an imbalanced testing set (4,189 comments with 28.05% promotions and 71.95% non-promotions) reflecting realistic platform conditions. Nine fine-tuning experiments were conducted with three transformer models (IndoBERT, IndoBERTweet, XLM-RoBERTa) using three learning rates (1e-5, 2e-5, 3e-5). Evaluation employed accuracy, precision, recall, F1-score, and AUC-ROC metrics. Results show IndoBERT with learning rate 1e-5 achieved best performance (F1-score 99.57%, recall 99.49%), outperforming IndoBERTweet (F1-score 98.58%) and XLM-RoBERTa (F1-score 99.28%). Interestingly, the formal corpus-trained model (IndoBERT) proved more effective than the social media model (IndoBERTweet), indicating that gambling promotion language patterns tend to be structured despite appearing in informal contexts. IndoBERT demonstrated greatest stability to learning rate variations (standard deviation 0.0011), while XLM-RoBERTa offered fastest inference time (2.48 ms) with optimal performance-efficiency balance. These findings provide practical recommendations for automated content moderation systems on Indonesian social media platforms, with IndoBERT for maximum accuracy scenarios and XLM-RoBERTa for large-scale real-time deployment.
PERBANDINGAN STRATEGI FINE-TUNED DAN PRE-TRAINED EFFICIENTNET-B0 UNTUK KLASIFIKASI PENYAKIT DAUN PADI Egar Firmansyah; Faisal Muttaqin; Afina Lina Nurlaili
Jurnal Informatika Teknologi dan Sains (Jinteks) Vol 8 No 2 (2026): EDISI 28
Publisher : Program Studi Informatika Universitas Teknologi Sumbawa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51401/jinteks.v8i2.7409

Abstract

Penyakit daun padi menjadi salah satu penyebab utama menurunnya produktivitas padi di Indonesia sehingga diperlukan pendekatan deteksi yang cepat dan akurat. Penelitian ini membandingkan performa dua strategi pelatihan pada arsitektur EfficientNet-B0, yaitu pre-trained yang membekukan seluruh lapisan base model dan fine-tuned yang melatih ulang keseluruhan parameter, untuk mengklasifikasikan enam kelas citra daun padi meliputi Bacterial Leaf Blight, Brown Spot, Healthy, Leaf Blast, Rice Hispa, dan Sheath Blight. Sebanyak 4.770 citra digunakan dengan pembagian 3.333 data training, 958 data validasi, dan 479 data testing. Konfigurasi hyperparameter terbaik diperoleh melalui empat tahap pengujian bertahap yang mencakup variasi batch size, epoch, learning rate, dan optimizer. Hasil evaluasi menunjukkan bahwa model pre-trained gagal total akibat fenomena class collapse dengan akurasi hanya 19,62% karena seluruh citra diprediksi pada satu kelas yang sama. Di sisi lain, model fine-tuned berhasil meraih akurasi 99,79% dengan f1-score sebesar 0,9981 pada data testing, di mana 478 dari 479 citra berhasil diklasifikasikan secara tepat. Hasil ini menegaskan bahwa proses fine-tuning memegang peranan kritis dalam mengadaptasi fitur EfficientNet-B0 terhadap domain penyakit daun padi.
Pengembangan Sistem Manajemen Inventaris dengan Integrasi OCR dan QR Code untuk Validasi Identitas Mahasiswa Muhammad Faizul Ulum; Retno Mumpuni; Afina Lina Nurlaili
JOINS (Journal of Information System) Vol 11 No 1 (2026): (Desember 2025 - Mei 2026)
Publisher : Fakultas Ilmu Komputer, Universitas Dian Nuswantoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33633/joins.v11i1.15972

Abstract

Inventory management at the Islamic Spiritual Activity Unit (UKKI) of UPN "Veteran" Jawa Timur still relies on manual methods, which are prone to human error and borrower identity manipulation. This study aims to develop a web-based inventory management information system integrating Optical Character Recognition (OCR) technology for identity validation and Quick Response (QR) Code for transaction security. Software development utilized the Rapid Application Development (RAD) method through an iterative approach with end-users. The results demonstrated that the OCR module, supported by pre-processing algorithms and Levenshtein Distance, successfully extracted and validated Student Identity Card (KTM) data automatically, achieving an acceptable similarity score (R) of over 0.87. Furthermore, implementing QR Codes as digital tokens proved effective in minimizing recording errors and ensuring officer accountability during handovers. Black-Box testing confirmed that 100% of the features, including dynamic stock management, operate precisely. In conclusion, this system successfully replaces conventional methods while enhancing time efficiency, data transparency, and organizational asset security.