Article Per Year (5 Year)
p-Index From 2021 - 2026
2.136
P-Index
This Author published in this journals
All Journal JITK (Jurnal Ilmu Pengetahuan dan Komputer) JSAI (Journal Scientific and Applied Informatics) JPMTT (Jurnal Pengabdian Masyarakat Teknologi Terbarukan)
Vina Ayumi
Universitas Dian Nusantara
Computer Science & IT Decision Sciences, Operations Research & Management Economics, Econometrics & Finance Education

Published : 15 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 15 Documents
Search

 1   2 
MOBILENET PERFORMANCE IMPROVEMENTS FOR DEEPFAKE IMAGE IDENTIFICATION USING ACTIVATION FUNCTION AND REGULARIZATION Handrie Noprisson; Vina Ayumi; Mariana Purba; Nur Ani
JITK (Jurnal Ilmu Pengetahuan dan Teknologi Komputer) Vol. 10 No. 2 (2024): JITK Issue November 2024
Publisher : LPPM Nusa Mandiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33480/jitk.v10i2.5798

Abstract

Deepfake images are often used to spread false information, manipulate public opinion, and harm individuals by creating fake content, making developing deepfake detection technology essential to mitigate these potential dangers. This study utilized the MobileNet architecture by applying regularization and activation function methods to improve detection accuracy. ReLU (Rectified Linear Unit) enhances the model's efficiency and ability to capture non-linear features, while Dropout and L2 regularization help reduce overfitting by penalizing large weights, thereby improving generalization. Based on experimental results, the MobileNet model optimized with ReLU and Dropout achieved an accuracy of 99.17% in the training phase, 85.34% in validation, and 70.60% in testing, whereas the MobileNet model optimized with ReLU and L2 showed lower accuracy in the training and validation phases compared to Dropout but achieved higher accuracy in testing at 72.18%. This study recommends MobileNet with ReLU and L2 due to its better generalization ability when testing data (resulting from reduced overfitting).
PENGGUNAAN SISTEM EVALUASI PEMBELAJARAN AL QUR’AN BERBASIS GOOGLE FORM PADA SISWA SMA DAARUL FIKRI Uswatun Hasanah; Vina Ayumi
JPMTT (Jurnal Pengabdian Masyarakat Teknologi Terbarukan) Vol. 6 No. 1 (2026): April
Publisher : Lembaga Penelitian Pengabdian Masyarakat Penerbitan dan Percetakan Indonesian Scholar Khiar Wafi (LPPMPP IKHAFI)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54650/jpmtt.v6i1.641

Abstract

Evaluasi pembelajaran Al-Qur’an di tingkat sekolah menengah masih banyak dilakukan secara manual, sehingga menimbulkan kendala dalam rekapitulasi nilai, risiko kehilangan data, dan keterlambatan pelaporan perkembangan siswa. Digitalisasi sistem penilaian menjadi kebutuhan mendesak untuk meningkatkan efisiensi, akurasi, dan transparansi administrasi pembelajaran keagamaan. Kegiatan pengabdian ini bertujuan mengimplementasikan sistem evaluasi pembelajaran Al-Qur’an berbasis Google Form pada siswa SMA Daarul Fikri. Metode yang digunakan adalah pendekatan service learning, yang mengintegrasikan pemberian solusi teknologi dengan pemberdayaan mitra melalui pelatihan teknis. Tahapan kegiatan meliputi observasi sistem manual, identifikasi kebutuhan penilaian (makhraj, tajwid, dan kelancaran bacaan), perancangan instrumen evaluasi digital, integrasi Google Form dengan Google Sheets, serta sosialisasi dan pelatihan penggunaan sistem melalui perangkat mobile. Program dilaksanakan di Pesantren Terpadu Daarul Fikri, Cikarang Barat, Kabupaten Bekasi, dengan sasaran pengajar Al-Qur’an pada Sub-Unit Kurikulum SMA. Hasil pelatihan menunjukkan bahwa sistem digital mampu mempercepat rekapitulasi nilai secara otomatis dan real-time, meningkatkan keamanan penyimpanan data, serta mempermudah monitoring perkembangan siswa. Implementasi ini juga meningkatkan kompetensi digital guru dan mendukung modernisasi tata kelola evaluasi pembelajaran Al-Qur’an secara berkelanjutan
Analisis Pengaruh Konektivitas dan Standar Sosial pada Penerapan Knowledge Sharing System di Universitas XYZ Vina Ayumi
JSAI (Journal Scientific and Applied Informatics) Vol 8 No 1 (2025): Januari
Publisher : Fakultas Teknik Universitas Muhammadiyah Bengkulu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36085/jsai.v8i1.7910

Abstract

This study analyzes the factors influencing the implementation of the Tacit Knowledge Sharing System at XYZ University by considering three main variables: Social Interaction, Trust, and Norm of Reciprocity. The study employs a purposive sampling method, involving 420 students from various faculties as respondents. Data collection was conducted through questionnaires distributed using the drop-off/pick-up method. Data analysis was performed using Structural Equation Modeling (SEM) based on SMART-PLS to evaluate the relationships between variables as well as the validity and reliability of the research constructs. The results indicate that Social Interaction and Norm of Reciprocity significantly influence the implementation of the Tacit Knowledge Sharing System, whereas Trust does not show a significant effect. These findings suggest that an academic environment that fosters social interaction and encourages a reciprocal culture enhances the effectiveness of knowledge sharing. The implications of this study provide insights for universities in designing policies to strengthen knowledge-sharing systems to improve academic collaboration
Model Deep Learning Berbasis Word2Vec dan LSTM untuk Klasifikasi Umpan Balik Kompetensi Profesional Dosen Vina Ayumi; Mariana Purba; Abd Rahman
JSAI (Journal Scientific and Applied Informatics) Vol 8 No 2 (2025): Juni
Publisher : Fakultas Teknik Universitas Muhammadiyah Bengkulu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36085/jsai.v8i2.8762

Abstract

This study aimed to develop a deep learning model based on Word2Vec and Long Short-Term Memory (LSTM) to classify sentiment in student feedback on lecturers' professional competence. Manual analysis of large volumes of evaluation text data required significant time and resources, thus an automated method was needed to assist the sentiment classification process. Word2Vec was used to represent words as fixed-dimension numerical vectors, which then served as input to the LSTM model. The LSTM model was selected for its ability to process sequential data and retain relevant long-term contextual information in the text. The dataset consisted of 6,124 evaluation texts, divided into 3,800 positive and 2,324 negative samples. The dataset was split into training (70%), validation (10%), and testing (20%) subsets. The model was trained for 50 epochs, achieving a training accuracy of 81.20% and a validation accuracy of 77.10%. Evaluation using a confusion matrix on the testing data showed that the model correctly classified 587 positive and 359 negative samples, while producing 106 false positives and 173 false negatives. These results indicated that the combination of Word2Vec and LSTM was effective in classifying sentiment in lecturer competence evaluation texts, with a testing accuracy of 77.2%.
Klasifikasi Teks Umpan Balik Kompetensi Sosial Dosen di Perguruan Tinggi Menggunakan Word2Vec dan CNN-1D Vina Ayumi; Mariana Purba; Siska Mailana
JSAI (Journal Scientific and Applied Informatics) Vol 8 No 2 (2025): Juni
Publisher : Fakultas Teknik Universitas Muhammadiyah Bengkulu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36085/jsai.v8i2.8763

Abstract

The advancement of artificial intelligence technology supported the development of automatic sentiment classification. This study aimed to develop a deep learning model based on Word2Vec and one-dimensional Convolutional Neural Networks (CNN-1D) to classify the sentiment of textual feedback regarding lecturers’ social competence in higher education. The dataset consisted of 6,124 feedback texts collected from student questionnaires at Universitas Sjakhyakirti. The data were proportionally divided into 70% for training, 10% for validation, and 20% for testing. The developed Word2Vec-CNN-1D model demonstrated performance with a training accuracy of 85.10% and a validation accuracy of 79.10%. During the testing phase, the model achieved an accuracy of 76.2% in classifying the feedback texts into positive and negative classes. Evaluation metric analysis showed that for the positive class, the model attained a precision of 0.827, recall of 0.760, and F1-score of 0.792, while for the negative class, it obtained a precision of 0.679, recall of 0.761, and F1-score of 0.717. The results indicated that the Word2Vec and CNN-1D model was more effective at identifying positive sentiments, whereas the performance for the negative class could still be improved in the classification of textual feedback on lecturers’ social competence.
Klasifikasi Teks Umpan Balik Kompetensi Kepribadian di Perguruan Tinggi Menggunakan Ekstraksi Fitur TF-IDF dan Algoritma Logistic Regression Vina Ayumi; Mariana Purba; Siska Mailana
JSAI (Journal Scientific and Applied Informatics) Vol 8 No 2 (2025): Juni
Publisher : Fakultas Teknik Universitas Muhammadiyah Bengkulu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36085/jsai.v8i2.8764

Abstract

This study aimed to develop and evaluate a text classification model to identify sentiment in feedback on lecturers’ personality competencies at a university using TF-IDF feature extraction and Logistic Regression (LR) algorithms. The data originated from student evaluations of lecturers’ personality competencies at Universitas Sjakhyakirti, consisting of a total of 6,112 texts labeled as positive sentiment (3,700) and negative sentiment (2,412). The dataset was then divided into three parts: training (70%), validation (10%), and testing (20%). The research stages included text preprocessing, which involved data cleaning, letter normalization, and the removal of common words, followed by term weighting using the TF-IDF method and classification using the LR model to categorize texts as positive or negative sentiment. The model was evaluated using accuracy, precision, recall metrics, and a confusion matrix. Experimental results showed that at the 50th epoch, the model achieved a training accuracy of 81.90% and a validation accuracy of 78.30%, while on the testing data, the TF-IDF-LR model reached an accuracy of 75.1%.
Application of Random Contrast and Brightness Range Methods on Phytomedicine Leaf Image Dataset Mariana Purba; Vina Ayumi; Sarwati Rahayu; Umniy Salamah; Inge Handriani; Ida Farida
JSAI (Journal Scientific and Applied Informatics) Vol 8 No 2 (2025): Juni
Publisher : Fakultas Teknik Universitas Muhammadiyah Bengkulu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36085/jsai.v8i2.8766

Abstract

This study aimed to enhance the performance of deep learning models in detecting and classifying medicinal plant leaf images by applying two data augmentation techniques, namely Random Contrast Augmentation (RCA) and Brightness Range Augmentation (BRA). The RCA technique randomly adjusted the contrast of images by calculating the pixel average and modifying each pixel value based on a contrast factor, thereby increasing the variation in image lighting. Meanwhile, BRA randomly altered the brightness of the images to simulate varying lighting conditions. The research process began with the collection of medicinal plant leaf image datasets, which were then divided into three parts: training data, validation data, and testing data. The dataset was then pre-processed to prepare the images before applying the augmentation. Augmentation techniques were employed to enrich the dataset by generating modified copies of images using RCA and BRA techniques. The application of both augmentation techniques resulted in a training dataset of 2,400 images, 300 validation images, and 300 testing images.
Penerapan Metode Augmentasi pada Dataset Farmakognosi Menggunakan Teknik Flip Secara Horizontal dan Vertikal Mariana Purba; Vina Ayumi; Nur Ani
JSAI (Journal Scientific and Applied Informatics) Vol 8 No 2 (2025): Juni
Publisher : Fakultas Teknik Universitas Muhammadiyah Bengkulu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36085/jsai.v8i2.8769

Abstract

This study aimed to apply image augmentation techniques, namely horizontal flip and vertical flip, to a pharmacognosy dataset to increase the diversity of training data in a pharmacognosy image recognition system. By applying these two techniques, this study focused on finalizing a pharmacognosy image dataset that could be used to train machine learning models. The application of these augmentation techniques improved the accuracy and generalization ability of the model in recognizing pharmacognosy images taken from various viewpoints and orientations. This study used two image augmentation techniques, vertical flip augmentation (VFA) and horizontal flip augmentation (HFA), to expand the pharmacognosy image dataset. Each augmentation technique produced four times the number of modified images from the original images with more and more diverse data variations. With the application of the vertical flip augmentation technique, the training dataset consisted of 2,400 images, a validation dataset of 300 images, and a testing dataset of 300 images, for a total of 3,000 data sets. Similarly, the horizontal flip augmentation technique yielded the same amount of data: 2,400 data points for training, 300 data points for validation, and 300 data points for testing. These two techniques increased the total number of training and testing data points to 3,000.
Implementasi Dataset Augmentation pada Citra Etnofimedisin Menggunakan Teknik Rotation dan Channel Shift Mariana Purba; Vina Ayumi; Wachyu Hari Haji
JSAI (Journal Scientific and Applied Informatics) Vol 8 No 2 (2025): Juni
Publisher : Fakultas Teknik Universitas Muhammadiyah Bengkulu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36085/jsai.v8i2.8776

Abstract

This study aimed to increase the quantity and variety of ethnopharmacological image datasets using image augmentation techniques, specifically rotation range augmentation (RRA) and channel shift range augmentation (CSA). The dataset augmentation was conducted to enrich the training data for the development of machine learning models used to recognize medicinal plant images. The RRA technique rotated images by random angles, providing variations in object orientation, while CSA altered the color channel values to simulate changes in lighting and the natural colors of plants. The research process included dataset collection, data preprocessing, application of both augmentation techniques, and division of the dataset into training, validation, and testing data. The results showed that the CSA technique produced 2,400 training data, 300 validation data, and 300 testing data, while the RRA technique produced the same amount of data. Therefore, the total data generated from both augmentation techniques amounted to 6,000 images, which could improve the accuracy and performance of deep learning models in recognizing ethnopharmacological images.
Pemodelan Feature Modelling dan User Interface Untuk Sistem Manajemen Data Etnofarmakologi Mariana Purba; Vina Ayumi; Anita Ratnasari
JSAI (Journal Scientific and Applied Informatics) Vol 8 No 2 (2025): Juni
Publisher : Fakultas Teknik Universitas Muhammadiyah Bengkulu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36085/jsai.v8i2.8777

Abstract

This research aimed to develop software for a pharmacognosy data management system by utilizing a feature modeling approach to map and categorize application features based on user requirements. The feature grouping consisted of mandatory and optional features designed to meet the diverse needs of users. The research methods used included literature study, data collection, requirement analysis, feature modeling, user interface modeling, and evaluation. The results of this study led to the development of a pharmacognosy data management system to manage data on medicinal plants, users, and reviews. The main features developed included user management and plant management, both of which were mandatory features for managing users and medicinal plants. Under Plant Management, there were optional features like plant part management and mandatory features like Plant Usage Management to support the management of medicinal plant usage information. Additionally, image data management was added as an alternative feature for managing images of medicinal plants. Thus, the system was expected to serve as a reference for software development that can adapt to user needs and the ongoing technological advancements.
Co-Authors Abd Rahman Anita Ratnasari Ida Farida Inge Handriani Inge Handriani Mariana Purba Mariana Purba Mariana Purba Noprisson, Handrie Nur Ani Nur Ani Nur Ani Sarwati Rahayu Sarwati Rahayu Siska Mailana Siska Mailana Umniy Salamah Uswatun Hasanah Wachyu Hari Haji