Article Per Year (5 Year)
p-Index From 2021 - 2026
2.702
P-Index
This Author published in this journals
All Journal JUTI: Jurnal Ilmiah Teknologi Informasi JITK (Jurnal Ilmu Pengetahuan dan Komputer) Abdi Masyarakat Jurnal Pengabdian kepada Masyarakat Nusantara Jurnal Pengabdian Kepada Masyarakat Jurnal Abdi Masyarakat Indonesia Jurnal Algoritma eProceedings of Engineering Journal of Fuzzy Systems and Control (JFSC) Scientific Journal of Informatics Control Systems and Optimization Letters Jurnal ilmiah teknologi informasi Asia Journal of Soft Computing Exploration
Pima Hani Safitri
Telkom University
Aerospace Engineering Arts Humanities Automotive Engineering Chemistry Civil Engineering, Building, Construction & Architecture Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Dentistry Economics, Econometrics & Finance Education Electrical & Electronics Engineering Energy Engineering Environmental Science Health Professions Industrial & Manufacturing Engineering Languange, Linguistic, Communication & Media Law, Crime, Criminology & Criminal Justice Library & Information Science Mathematics Nursing Public Health Social Sciences Veterinary Other

Published : 20 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 20 Documents
Search

 1   2 
Deteksi Kantuk Menggunakan Ear Dan Mar Berbasis Deep Learning Pada Citra Sekuensial Galih Ananta, Whisnumurty; Wali Satria Bahari Johan, Ahmad; Hani Safitri, Pima
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 — Pada tahun 2023, kecelakaan lalu lintas mencapai angka tertinggi dalam lima tahun terakhir, dengan total 148.575 kasus, dimana 20% di antaranya disebabkan oleh kantuk, yang meningkatkan risiko kecelakaan hingga tiga kali lipat akibat penurunan kewaspadaan pengemudi. Penelitian ini mengusulkan metode ekstraksi fitur geometris Eye Aspect Ratio (EAR) dan Mouth Aspect Ratio (MAR) dari citra wajah sekuensial secara real-time menggunakan MediaPipe. EAR dan MAR dihitung berdasarkan koordinat landmark mata dan mulut, kemudian disusun dalam urutan temporal untuk menggambarkan perubahan kondisi subjek seiring waktu. Representasi ini efektif dalam menggambarkan transisi kantuk, yang dapat digunakan sebagai input dalam model deteksi berbasis deep learning. Penelitian ini melibatkan lima komponen utama: pengujian metode klasifikasi, pengolahan input data, perbaikan citra, augmentasi data, dan pipeline model. Data dari National Tsing Hua University Drowsiness Dataset (NTHU-DDD) dikelompokkan dalam window 60 frame, dengan fitur EAR dan MAR diekstraksi menggunakan MediaPipe. Hasil penelitian menunjukkan bahwa model CNN LSTM efektif dalam memproses fitur EAR dan MAR secara sekuensial. Representasi penuh dengan input implisit (120, 1) memberikan performa terbaik, sementara teknik augmentasi SMOTE meningkatkan performa dengan menyeimbangkan distribusi kelas. Model CNN-LSTM-120FT tanpa perbaikan citra atau augmentasi menunjukkan performa paling stabil, dengan accuracy 85,59% dan precision 92,31%. Kata kunci— kecelakaan lalu lintas, kantuk, EAR, MAR, deep learning, citra sekuensial
Implementasi Sistem Pakar untuk Mendiagnosa Penyakit Kulit pada Manusia yang Disebabkan oleh Alergi dengan Metode Dempster Shafer Dafi’us Shidqi , Mochamad; Wali Satria Bahari Johan, Ahmad; Hani Safitri, Pima
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 — Penyakit kulit akibat alergi merupakan salah satu permasalahan kesehatan yang umum terjadi, terutama di wilayah tropis seperti Indonesia. Gejala yang bervariasi dan kemiripan dengan penyakit lain seringkali menyulitkan proses diagnosis secara akurat. Penelitian ini bertujuan untuk mengembangkan sistem pakar berbasis metode Dempster Shafer guna mendiagnosis penyakit kulit yang disebabkan oleh alergi. Metode ini dipilih karena kemampuannya dalam menangani ketidakpastian informasi dan menghasilkan tingkat keyakinan terhadap diagnosis yang diberikan. Sistem dikembangkan menggunakan bahasa pemrograman Python dengan dukungan antarmuka GUI berbasis Tkinter. Pengujian dilakukan menggunakan 37 data kasus yang divalidasi oleh pakar, dan hasil evaluasi sistem menunjukkan bahwa metode ini mampu memberikan diagnosis yang akurat dengan nilai akurasi yang tinggi. Selain memberikan informasi diagnosis awal beserta tingkat kepercayaannya, sistem ini juga menyediakan rekomendasi penanganan awal. Dengan demikian, sistem pakar ini diharapkan dapat menjadi solusi diagnosis yang efisien, terutama bagi masyarakat yang memiliki keterbatasan akses ke layanan kesehatan, sekaligus berfungsi sebagai alat bantu edukasi dalam mengenali gejala penyakit kulit akibat alergi secara mandiri. Kata kunci— sistem pakar, penyakit kulit, alergi, dempster-shafer, diagnosis, kepercayaan
Klasifikasi Ekspresi Wajah Menggunakan HFT CNN Dan Siamese Network Pada Citra Wajah Satria Putra Buana, Elang; Yusuf Wicaksono, Ardian; Hani Safitri, Pima
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 — Ekstraksi fitur yang kurang optimal merupakan salah satu kendala utama dalam klasifikasi ekspresi wajah menggunakan metode tradisional. Penelitian ini bertujuan untuk meningkatkan akurasi pengenalan ekspresi wajah dengan menerapkan pendekatan berbasis deep learning yang secara khusus menargetkan bagian-bagian penting wajah. Metode yang diusulkan menggabungkan arsitektur Siamese Neural Network (SNN) untuk mengukur kemiripan antar ekspresi, serta multi-level feature extraction (HFT-CNN) untuk melakukan ekstraksi fitur secara mendalam dan terfokus pada tiga area utama wajah, yaitu keseluruhan wajah, mata dan alis, serta mulut. Ketiga channel ini digabungkan untuk membentuk representasi fitur yang lebih kaya dan informatif. Hasil implementasi menunjukkan bahwa arsitektur HFT-CNN mampu mencapai akurasi hingga 99%, sedangkan model SNN Triple mencatatkan akurasi sebesar 91%. Meskipun demikian, hasil prediksi dari kedua model belum sepenuhnya stabil dalam berbagai kondisi pengujian, yang mengindikasikan masih adanya keterbatasan dalam hal generalisasi terhadap variasi ekspresi wajah. Selain itu, proses pengumpulan dan preprocessing data turut berpengaruh terhadap performa model, sehingga seleksi data secara manual tetap diperlukan guna memastikan kualitas dan relevansi data yang digunakan dalam pelatihan Kata kunci— convolutional neural networks, ekspresi wajah, pembelajaran mesin, siamese networks.
GECOM: GREEN COMMUNICATION CONCEPTS FOR ENERGY EFFICIENCY IN WIRELESS MULTIMEDIA SENSOR NETWORK Muhammad Ihsan Diputra; Ahmad Akbar Megantara; Pima Hani Safitri; Didik Purwanto
JUTI: Jurnal Ilmiah Teknologi Informasi Vol. 18, No. 2, July 2020
Publisher : Department of Informatics, Institut Teknologi Sepuluh Nopember

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12962/j24068535.v18i2.a942

Abstract

Wireless multimedia sensor network (WMSN) is one of broad wide application for developing a smart city. Each node in the WMSN has some primary components: sensor, microcontroller, wireless radio, and battery. The components of WMSN are used for sensing, computing, communicating between nodes, and flexibility of placement. However, the WMSN technology has some weakness, i.e. enormous power consumption when sending a media with a large size such as image, audio, and video files. Research had been conducted to reduce power consumption, such as file compression or power consumption management, in the process of sending data. We propose Green Communication (GeCom), which combines power control management and file compression methods to reduce the energy consumption. The power control management method controls data transmission. If the current data has high similarity with the previous one, then the data will not be sent. The compression method compresses massive data such as images before sending the data. We used the low energy image compression algorithm algorithm to compress the data for its ability to maintain the quality of images while producing a significant compression ratio. This method successfully reduced energy usage by 2% to 17% for each data.   
Peningkatan Sensitivitas Deteksi Diabetic Retinopathy melalui Mekanisme Hierarchical Self-Attention pada Swin Transformer Mustaqim, Tanzilal; Safitri, Pima Hani; Oktavia, Vessa Rizky
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.22-2.2986

Abstract

Diabetic Retinopathy (DR) is a complication of diabetes that can cause blindness if not detected early. CNN has limitations in capturing scattered lesions due to its narrow receptive field, while Vision Transformers are generally less computationally efficient. The objective of this study is to develop an approach that can capture long-range spatial dependencies while maintaining computational efficiency for resource-limited clinical applications. The Swin Transformer-Tiny was implemented with a shifted window-based hierarchical self-attention mechanism on the APTOS 2019 dataset (3,663 retinal images), with pre-processing (CLAHE, gamma correction, Gaussian filtering) and data augmentation. The model was trained using SGD with CosineAnnealingLR and evaluated based on accuracy, precision, recall, and F1-score with a focus on minimizing false negatives. Swin Transformer-Tiny achieved an accuracy of 84.99%, precision of 84.89%, and recall of 84.99%, surpassing EfficientNet-B0 by 1.32% in F1-score and outperforming ResNet50 by 5.60%. The attention mechanism reduces false negatives by 1.28% compared to conventional CNNs while maintaining linear computational complexity. This research contributes to showing that hierarchical self-attention in Swin Transformer effectively improves DR detection sensitivity by overcoming the limitations of CNN receptive fields, while maintaining computational efficiency for clinical implementation.
ADAPTIVE AL-QUR’AN MEMORIZATION RECOMMENDATION SYSTEM BASED ON FUZZY LOGIC COGNITIVE MEMORY AND PROFILE MATCHING Afifah Fikriyah Dhiya'ulhaq; Muhammad Dzulfikar Fauzi; Pima Hani Safitri
JITK (Jurnal Ilmu Pengetahuan dan Teknologi Komputer) Vol. 11 No. 3 (2026): JITK Issue February 2026
Publisher : LPPM Nusa Mandiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33480/jitk.v11i3.8048

Abstract

Memorizing mutasyabihat verses in the Qur’an is particularly challenging due to similarities in structure, linguistic patterns, and semantic density that place a heavy load on short-term memory. Conventional memorization approaches do not account for individual cognitive differences when dealing with verse complexity. This study proposes an adaptive recommender system based on cognitive modeling to align verse group selection with the user’s memory profile.The system models memory capacity as a multidimensional profile using fuzzy inference derived from three quantitative indicators: continuous memory score, total correct recall, and average response time. This profile is matched with verse group feature vectors through a profile matching approach and a weighted Euclidean distance similarity measure within a Multi-Attribute Decision Making (MADM) framework. Four verse characteristics are considered: thematic (35%), semantic (25%), linguistic (25%), and pattern (15%).An adaptive calibration phase combines 20% of the initial cognitive profile with 80% of actual memorization performance, reflecting the dominance of behavioral evidence over initial assessment. System evaluation employs the Top-N Accuracy method commonly used in recommender systems.Testing with 29 participants resulted in a Top-3 success rate of 66% and an overall Top-N accuracy of 62.07%. These results indicate that cognitive profile–based multidimensional similarity can adaptively match verse complexity to individual memory capacity. This study demonstrates that fuzzy cognitive modeling and profile matching can be effectively implemented in adaptive personalized learning systems to optimize memorization of mutasyabihat verses
Hybrid GA-GWO with Dual-Vector Encoding for Indonesian School Timetabling Akbar Muhammad Sadat; Alqis Rausanfita; Pima Hani Safitri
Journal of Fuzzy Systems and Control Vol. 4 No. 2 (2026): Vol. 4 No. 2 (2026)
Publisher : Peneliti Teknologi Teknik Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59247/jfsc.v4i2.418

Abstract

School timetabling is a complex combinatorial optimization problem that involves assigning subjects, teachers, and classes to predefined time slots while satisfying numerous institutional constraints. In many Indonesian junior high schools, scheduling is still performed using manual approaches, which are often time-consuming and prone to conflicts. Compared with university timetabling, school timetabling presents additional challenges due to fixed class groups, rigid subject allocations, teacher availability constraints, and institutional regulations. To address these challenges, this study proposes a hybrid optimization framework that combines a Guided Genetic Algorithm (GA) and Grey Wolf Optimizer (GWO) for the school timetable. The proposed framework incorporates dual-vector solution encoding to provide a structured representation of scheduling components and support efficient constraint handling during the optimization process. In addition, a majority-voting and guided mutation strategy is employed to enhance the balance between exploration and exploitation. The proposed method was evaluated using real-world scheduling data from an Indonesian junior high school consisting of 27 classes, 54 teachers, 13 subjects, and 36 time slots. Experimental results show that the proposed hybrid GA-GWO achieved a fitness improvement of 95.84%, reducing the fitness value from 16,120 to 670, compared with improvements of 89.83% and 94.31% obtained by Traditional GA and Guided GA, respectively. Although the proposed method required approximately 28 minutes of execution time, it produced the highest overall timetable quality among the evaluated approaches. These findings demonstrate that the integration of dual-vector encoding, majority voting, and guided mutation within a hybrid GA-GWO framework can effectively improve timetable optimization for real-world Indonesian school scheduling environments.
Digitalisasi Layanan Perizinan dan Monitoring Siswa melalui SIPAS ANTARTIKA di SMK Antartika 1 Sidoarjo Pima Hani Safitri; Muhammad Fadli Zamzami; Imam Prasetyo Suherman; Khansa Nailah Anjani; Nur Aulia Dinda Putri; Ananda Bintang Saputra; Dimas Adiputra; Nilla Rachmaningrum
Jurnal Abdi Masyarakat Indonesia Vol 6 No 4 (2026): JAMSI - Juli 2026
Publisher : CV Firmos

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54082/jamsi.3093

Abstract

Digitalisasi layanan administrasi sekolah merupakan salah satu upaya untuk meningkatkan efektivitas pengelolaan data dan pelayanan kepada warga sekolah. SMK Antartika 1 Sidoarjo masih menghadapi kendala pada proses perizinan siswa yang dilakukan secara manual sehingga monitoring keluar masuk siswa belum optimal. Kegiatan pengabdian kepada masyarakat ini bertujuan mengimplementasikan SIPAS ANTARTIKA, yaitu sistem perizinan berbasis web yang dilengkapi fitur pengajuan izin daring, persetujuan berjenjang, validasi menggunakan QR Code, dan monitoring siswa secara real-time. Metode pelaksanaan meliputi observasi kebutuhan mitra, implementasi sistem, pelatihan pengguna, serta evaluasi menggunakan kuesioner skala Likert. Hasil kegiatan menunjukkan bahwa SIPAS ANTARTIKA mampu meningkatkan efektivitas proses administrasi perizinan dan mempermudah monitoring siswa. Evaluasi terhadap 24 responden menunjukkan bahwa 84,17% responden memberikan penilaian setuju dan sangat setuju terhadap sistem dan pelaksanaan kegiatan. Dengan demikian, SIPAS ANTARTIKA menjadi solusi yang efektif dalam mendukung transformasi digital layanan administrasi di lingkungan sekolah.
Broad Learning System: A Derivation-Based Mathematical Formulation Dimas Chaerul Ekty Saputra; Dyah Putri Rahmawati; Affifah Mutiara Pertiwi; Muhammad Ijaz Shafarin; Kharisma Monika Dian Pertiwi; Thinzar Aung Win; Irianna Futri; Pima Hani Safitri
Control Systems and Optimization Letters Vol 4, No 1 (2026)
Publisher : Peneliti Teknologi Teknik Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59247/csol.v4i1.294

Abstract

Broad Learning System is a wide learning framework that constructs nonlinear feature representations while enabling efficient model training through analytical solutions. This paper presents a derivation-based formulation of Broad Learning System that explains the mathematical structure underlying the learning process. The model constructs an expanded feature representation through feature mapping nodes followed by enhancement nodes that further enrich the learned representation. The learning problem is then expressed as a linear model in the constructed feature space, and the output weights are obtained using ridge regularized least squares optimization. This formulation allows the training process to be solved directly using matrix operations without iterative gradient based procedures. In addition, an incremental learning mechanism is introduced to enable efficient parameter updates when new samples or additional nodes are incorporated into the model. The presented formulation highlights how Broad Learning System combines nonlinear feature construction with computationally efficient closed form learning, providing a clear theoretical interpretation of the learning process.
Indonesian sign language (BISINDO) gesture detection using Yolov11-Pose Achmad Dany Alfansyah; Ardian Yusuf Wicaksono; Pima Hani Safitri
Journal of Soft Computing Exploration Vol. 7 No. 3 (2026): September 2026
Publisher : SHM Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52465/joscex.v7i3.69

Abstract

Sign language serves as the primary means of communication for deaf individuals, yet existing recognition systems for Bahasa Isyarat Indonesia (BISINDO) have not fully exploited keypoint-based pose estimation, particularly for dynamic alphabetic gestures. This study developed a BISINDO gesture detection system using the YOLOv11m-Pose algorithm integrated with MediaPipe hand landmark extraction, Letterbox image preprocessing, and a multi-phase class representation strategy that divided dynamic letters into two movement-phase classes. The main novelty of this study lies in the integration of YOLOv11-Pose with MediaPipe hand keypoint extraction and a multi-phase representation strategy for dynamic letters, which has not been previously applied to BISINDO alphabet detection. A dataset of 1.450 images across 29 classes was collected, augmented to 8.700 samples, and split into training, validation, and test sets. Evaluation on 870 test images yielded an overall accuracy of 99,43%, a macro precision of 99,44%, macro recall of 99,43%, and a mAP50 of 99,11%. All six dynamic letter classes achieved perfect prediction scores, confirming the effectiveness of the multi-phase representation approach. These results demonstrated that the proposed system was capable of reliable BISINDO alphabet detection and provided a solid foundation for further development toward full support for sign language communication.
Co-Authors Achmad Dany Alfansyah Affifah Mutiara Pertiwi Afifah Fikriyah Dhiya'ulhaq Ahmad Akbar Megantara Ahmad Wali Satria Bahari Johan Akbar Muhammad Sadat Alqis Rausanfita Ananda Bintang Saputra Anditya Arifianto Ardian Yusuf Wicaksono Arip Ramadan Dafi’us Shidqi , Mochamad Desita Nur Rachmaniar DEWI RAHMAWATI Didik Purwanto Dimas Adiputra Dimas Chaerul Ekty Saputra Dyah Putri Rahmawati Firdaus, Raihan Galih Ananta, Whisnumurty Imam Prasetyo Suherman Irianna Futri Khansa Nailah Anjani Kharisma Monika Dian Pertiwi Krishna Lokajaya, Gabriel Azarya Kurniawan Nur Ramadhani Muhajir, Daud Muhammad Dzulfikar Fauzi Muhammad Fadli Zamzami Muhammad Ihsan Diputra Muhammad Ijaz Shafarin Muhammad Ilham Alhari Nilla Rachmaningrum Nur Aulia Dinda Putri Putra, Muhammad Alifian Saputra, Ananda Bintang Satria Putra Buana, Elang Sebastian, Gerrard Tanzilal Mustaqim Thinzar Aung Win Vessa Rizky Oktavia Yulianto, Ilham Zulkarnaen, Feiticeira