Article Per Year (5 Year)
p-Index From 2021 - 2026
0.882
P-Index
This Author published in this journals
All Journal Sinkron : Jurnal dan Penelitian Teknik Informatika RABIT: Jurnal Teknologi dan Sistem Informasi Univrab JURNAL TEKNOLOGI DAN ILMU KOMPUTER PRIMA (JUTIKOMP) Jurnal Minfo Polgan (JMP)
Maya Sofhia
Universitas Prima Indonesia
Computer Science & IT Engineering Library & Information Science Mathematics Social Sciences

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

Found 5 Documents
Search

 1 
Raw Material Weighing Application Through Visual-Based RS-232 Cable Port Sofhia, Maya; Manawan, Junio Fegri Wira Manawan
Sinkron : jurnal dan penelitian teknik informatika Vol. 7 No. 1 (2023): Articles Research Volume 7 Issue 1, 2023
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/sinkron.v8i1.12158

Abstract

Officers who record incoming weighing data using a manual weighing machine experience difficulties when interacting with the weighing device. It is difficult to press the buttons, the storage memory cannot be more than three digits, and the display is difficult for officials to understand which can hinder the performance of recording the scales. Lack of capacity to store scale data on Officers who record incoming weighing data using a manual weighing machine experience difficulties when interacting with the weighing device. It is difficult to press the buttons, the storage memory cannot be more than three digits, and the display is difficult for officials to understand which can hinder the performance of recording the scales. Lack of capacity to store scale data on machine, so it can only store a maximum of 3 data scales. Inflexible on-machine data storage system. That is, the data scales that have been stored cannot be moved apart from within the machine itself. The large size of the machine is enough to take up space. So it is necessary to design a signal connection path from the scales to the computer via cable. With a computerized weighing application through the RS-232 communication port, where data input can be done using a visual-based weighing application. This data is then processed and produces an accurate report according to the data recorded by the scales. The testing process is carried out by entering data on the scales 19 times along with the check-in and check-out process for each incoming truck of raw materials for transportation. The testing process is carried out so that the application can run properly. machine, so it can only store a maximum of 3 data scales. Inflexible on-machine data storage system. That is, the data scales that have been stored cannot be moved apart from within the machine itself. The large size of the machine is enough to take up space. So it is necessary to design a signal connection path from the scales to the computer via cable. With a computerized weighing application through the RS-232 communication port, where data input can be done using a visual-based weighing application. This data is then processed and produces an accurate report according to the data recorded by the scales. The testing process is carried out by entering data on the scales 19 times along with the check-in and check-out process for each incoming truck of raw materials for transportation. The testing process is carried out so that the application can run properly.
Comparative Evaluation of YOLOv8 and YOLOv11 for Student Behavior Detection in Classroom CCTV Environments Sofhia, Maya
Sinkron : jurnal dan penelitian teknik informatika Vol. 10 No. 1 (2026): Article Research January 2026
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/sinkron.v10i1.15648

Abstract

Monitoring student behavior during classroom learning is important for supporting learning quality and teacher performance. This study presents a pilot comparison between YOLOv8 and YOLOv11 for detecting student classroom behaviors from CCTV images. Six elementary behaviors are consistently defined and used throughout the work: lookup, raise-hand, read, stand, turn-head, and write. The available SCB dataset contains 4,934 labeled images, but this study deliberately uses a front-facing subset of 100 images that best represent clear posture and behavior. After augmentation, the dataset grows to 220 images, split into 180 training, 30 validation, and 10 testing images. Both models are trained for 25 epochs on a T4 GPU with comparable configurations. At the detector level, YOLOv11 achieves higher mean average precision (mAP) of 42.9% compared to 28.9% for YOLOv8. At the behavior level, overall classification accuracy on the test set is 43.3% for YOLOv8 and 37.5% for YOLOv11. These results indicate a trade-off: YOLOv11 provides stronger bounding-box detection performance, while YOLOv8 produces slightly more stable behavior-level predictions on this very small and imbalanced dataset. The study emphasizes that these findings are exploratory baselines rather than definitive benchmarks, because the dataset is small and no statistical significance testing is performed. Future work must use a larger portion of the SCB dataset, more balanced class distributions, repeated experiments, and statistical analysis to obtain more robust conclusion.
KLASIFIKASI EKSPRESI WAJAH UNTUK IDENTIFIKASI INDIKASI KECEMASAN MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORK (CNN) Aldrian Fidel Pranata Sunaryo; Maya Sofhia; Eighia Nita Perangin Angin; Esa Andriani Barus
Rabit : Jurnal Teknologi dan Sistem Informasi Univrab Vol 11 No 2 (2026): Juli
Publisher : LPPM Universitas Abdurrab

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36341/rabit.v11i2.8065

Abstract

This study aims to develop a facial expression classification system to identify indications of anxiety using the Convolutional Neural Network (CNN) method, utilizing the FER2013 dataset, which consists of seven emotion categories: angry, disgust, fear, happy, sad, surprise, and neutral. The research stages include data preprocessing, CNN model training, model performance evaluation, and real-time testing using a webcam. The developed system is capable of recognizing facial expressions and classifying them into PANIC, NORMAL, and UNCERTAIN statuses based on the confidence scores generated by the model. The expressions of fear, sad, angry, and disgust are interpreted as indicators of anxiety, whereas happy, neutral, and surprise are interpreted as normal conditions. Based on the testing results using the FER2013 test data, the CNN model achieved an accuracy of 92.58%, a precision of 93.48%, a recall of 92.58%, and an F1-score of 92.27%. The results demonstrate that the CNN method can be effectively utilized for facial expression classification and to provide early indications of anxiety based on the detected expressions. However, the system's performance remains influenced by factors such as lighting, camera quality, facial positioning, and limitations in the quantity and distribution of data within the utilized dataset. Therefore, further development is required to enhance the accuracy and reliability of the system in detecting indications of anxiety in real-time.
Prediksi Penderita Penyakit Anemia Berdasarkan Konsentrasi Hemoglobin Dengan Algoritma Artifical Neutral Network (ANN) Maya Sofhia; Putri Rahmadhani Br Tarigan; Eldita Haggai Sihombing; Yohana Tampubolon; Devi Kenes Andari Br Ginting
Jurnal Minfo Polgan Vol. 15 No. 2 (2026): Artikel Penelitian
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/jmp.v15i2.16331

Abstract

Anemia merupakan salah satu masalah kesehatan global yang ditandai dengan rendahnya kadar hemoglobin dalam darah, sehingga mengurangi kemampuan darah dalam mengangkut oksigen ke seluruh tubuh. Kondisi ini dapat berdampak pada penurunan kualitas hidup dan produktivitas apabila tidak terdeteksi sejak dini. Oleh karena itu, diperlukan metode prediksi yang akurat untuk membantu proses deteksi awal anemia. Penelitian ini bertujuan untuk membangun model prediksi status anemia dengan konsentrasi hemoglobin sebagai variabel utama menggunakan algoritma Artificial Neural Network (ANN). Dataset yang digunakan terdiri dari 1.421 data sampel dengan beberapa parameter pemeriksaan darah yang diperoleh dari platform Kaggle (Anemia Dataset). Atribut dataset meliputi kadar hemoglobin (Hb), Mean Corpuscular Hemoglobin (MCH), Mean Corpuscular Hemoglobin Concentration (MCHC), Mean Corpuscular Volume (MCV), gender, serta status anemia. Dalam penelitian ini, kadar hemoglobin digunakan sebagai indikator utama, sedangkan parameter darah lainnya dimanfaatkan sebagai variabel pendukung untuk meningkatkan akurasi model prediksi. Tahapan penelitian meliputi pra-pemrosesan data, normalisasi Min-Max, pembagian data menjadi data latih (80%) dan data uji (20%), pelatihan model ANN, serta evaluasi kinerja menggunakan metrik klasifikasi. Berdasarkan hasil pengujian perbandingan, model ANN berhasil mendeteksi status anemia dengan tingkat akurasi sempurna 100% pada pengujian perbandingan model. Evaluasi detail arsitektur ANN (100 epoch, ReLU, Sigmoid) menghasilkan nilai akurasi sebesar 99,3%, presisi 98,46%, recall 100%, dan F1-score 99,22%. Nilai recall sebesar 1,0 menunjukkan bahwa seluruh kasus anemia berhasil teridenteksi tanpa adanya kesalahan False Negative. Dengan demikian, model yang dihasilkan diharapkan dapat digunakan sebagai alat bantu sistem pendukung keputusan awal bagi tenaga medis dalam proses skrining dini anemia.
Comparative Analysis of Local Outlier Factor (LOF) and Elliptic Envelope Algorithms in Outlier Detection of Customer Purchase Behavior Data Viola -; Maya Sofhia; Juliansyah Putra Tanjung; Marciello Farrel
JURNAL TEKNOLOGI DAN ILMU KOMPUTER PRIMA (JUTIKOMP) Vol. 9 No. 1 (2026): Jurnal Teknologi dan Ilmu Komputer Prima (JUTIKOMP)
Publisher : Fakultas Teknologi dan Ilmu Komputer Universitas Prima Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34012/jutikomp.v8i2.7909

Abstract

Outlier detection is a crucial stage in customer data analysis, as the presence of anomalies can significantly affect analytical outcomes and decision-making processes. This study aims to compare the performance of the Local Outlier Factor (LOF) and Elliptic Envelope (EE) algorithms in detecting outliers in customer behavior data. The dataset consists of 2,240 customer records with eight key features representing demographic characteristics, purchasing behavior, and responses to marketing campaigns. The research stages include data preprocessing, feature selection, normalization, and the application of the LOF and EE algorithms. The results indicate that LOF detects a larger number of outliers compared to EE, while EE tends to be more selective and focuses on extreme global deviations. A number of overlapping data points were also detected by both algorithms, indicating the presence of anomalies with consistent characteristics. Feature characteristic analysis reveals that each algorithm has a distinct and complementary detection focus. These findings suggest that a comparative approach between LOF and EE provides a more comprehensive understanding of customer anomaly patterns and supports the selection of appropriate outlier detection methods aligned with specific analytical objectives.
Co-Authors Aldrian Fidel Pranata Sunaryo Devi Kenes Andari Br Ginting Eighia Nita Perangin Angin Eldita Haggai Sihombing Esa Andriani Barus Juliansyah Putra Tanjung Manawan, Junio Fegri Wira Manawan Marciello Farrel Putri Rahmadhani Br Tarigan Viola - Yohana Tampubolon