Garuda - Garba Rujukan Digital

SMART ATTENDANCE TRACKING SYSTEM EMPLOYING DEEP LEARNING FOR FACE ANTI-SPOOFING PROTECTION Bani Nurhakim; Ahmad Rifai; Dian Ade Kurnia; Dadang Sudrajat; Ujang Supriatna
JITK (Jurnal Ilmu Pengetahuan dan Teknologi Komputer) Vol. 10 No. 3 (2025): JITK Issue February 2025
Publisher : LPPM Nusa Mandiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33480/jitk.v10i3.5992

Conventional attendance systems face challenges in accuracy and efficiency, often vulnerable to spoofing and data manipulation. This study addresses these issues by developing a smart attendance system integrating Deep Learning-based facial recognition with anti-spoofing technology. The system ensures secure and reliable attendance authentication while automating and enhancing management processes. Utilizing a convolutional neural network (CNN) architecture, the system processes raw facial images directly without additional feature extraction, improving accuracy and efficiency. A novel training strategy, termed 50 Random Samples-30 Sub-epochs Count-1 Epoch, is introduced to optimize the training process. This strategy involves random sampling during each forward pass and grouping 30 passes as one epoch, enabling the use of complex CNN architectures and automatic dataset expansion. The system achieves 98.90% accuracy in identifying genuine attendance, maintaining a confidence level above 80%, significantly reducing spoofing risks and errors. This innovative solution has significant implications, particularly for educational institutions. It automates attendance tracking, minimizes manual effort, reduces errors, and supports disciplinary enforcement through accurate data. Moreover, its scalability allows for application across various environments, offering benefits to a wide range of institutions. By enhancing data accuracy and operational efficiency, this system sets a foundation for smarter, more reliable attendance management, strengthening administrative practices in education and beyond.

Peningkatan Kreativitas Karang Taruna Melalui Pelatihan Desain Grafis dan Konten Digital Ahmad Faqih; Ahmad Rifai; Mohamad Riad Solihin; Muhammad Daffa Ayyasy
AMMA : Jurnal Pengabdian Masyarakat Vol. 1 No. 03 (2022): AMMA : Jurnal Pengabdian Masyarakat
Publisher : CV. Multi Kreasi Media

Show Abstract | Download Original | Original Source | Check in Google Scholar

The role of youth in village development is becoming increasingly important in the digital era, particularly in promoting local potential, social activities, and creative economic initiatives. One of the main challenges faced by youth organizations such as Karang Taruna is the limited ability to produce engaging digital content and graphic designs, despite the great potential of social media as a means of promotion and communication. This Community Service Program (PKM) aims to enhance the capacity of Karang Taruna members in creating creative content and graphic design using simple and accessible digital applications. The program was carried out in several stages: identifying participants’ needs, developing training modules, conducting in-person training sessions, and providing post-training assistance. The training materials included the basics of graphic design, understanding visual elements (color, typography, layout), simple photography and videography techniques using smartphones, and the use of design applications such as Canva, CapCut, and Pixellab. Participants also practiced creating social media content to promote village activities, local MSMEs, and social campaigns managed by Karang Taruna. The results show a significant improvement in participants' skills in designing posters, Instagram feeds, and short videos for publication purposes. Some of the participants' works have been uploaded to Karang Taruna’s official social media accounts and received positive responses from the community. This program not only enhanced technical skills but also fostered confidence, creativity, and a spirit of collaboration among members. It makes a tangible contribution to empowering village youth through digital literacy and creative media. Moving forward, this training can be developed into a sustainable program to strengthen the village’s digital identity and promote local potential through community-based efforts.

Penerapan Model LSTM Univariat dengan Walk-Forward Validation untuk Estimasi Harga Saham Nokia Ahmad Rifai; Roni Saputra; Dian Ade Kurnia; Fatihanursari Dikananda
TEMATIK Vol. 13 No. 1 (2026): Tematik : Jurnal Teknologi Informasi Komunikasi (e-Journal) - Juni 2026
Publisher : LPPM POLITEKNIK LP3I BANDUNG

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.38204/tematik.v13i1.3000

Prediksi harga saham merupakan permasalahan yang kompleks karena karakteristik data deret waktu finansial yang bersifat non-linear, volatil, dan dinamis. Meskipun algoritma Long Short-Term Memory (LSTM) terbukti efektif dalam menangkap pola temporal, banyak penelitian sebelumnya menggunakan pendekatan multivariat yang melibatkan variabel dengan korelasi sangat tinggi sehingga berpotensi menimbulkan redundansi informasi dan meningkatkan kompleksitas model. Penelitian ini mengusulkan model LSTM univariat untuk memprediksi harga saham Nokia Corporation (NOK) dengan menggunakan harga penutupan sebagai variabel masukan tunggal. Data historis harian periode 1 Oktober 2015 hingga 24 Oktober 2025 sebanyak 2.532 observasi diperoleh dari Yahoo Finance. Sebelum proses pemodelan, dilakukan analisis korelasi terhadap variabel Open, High, Low, Close, dan Volume. Hasil analisis menunjukkan bahwa variabel harga memiliki korelasi yang sangat tinggi (r > 0,99), sedangkan variabel Volume memiliki korelasi yang sangat rendah terhadap variabel harga (−0,052 ≤ r ≤ −0,043). Berdasarkan hasil tersebut, harga penutupan dipilih sebagai fitur utama dalam pemodelan. Untuk mengevaluasi performa model pada kondisi prediksi yang realistis, diterapkan metode Walk-Forward Validation (WFV) sebanyak 30 iterasi. Hasil pengujian menunjukkan bahwa model memperoleh nilai MSE sebesar 0,0260, RMSE sebesar 0,1613, MAE sebesar 0,1086, MAPE sebesar 2,75%, dan koefisien determinasi (R²) sebesar 0,9446. Hasil tersebut menunjukkan bahwa model mampu menjelaskan 94,46% variasi harga saham dengan tingkat kesalahan prediksi yang rendah. Penelitian ini menyimpulkan bahwa model LSTM univariat yang didukung oleh proses seleksi fitur yang sistematis dan validasi temporal yang robust mampu menghasilkan prediksi harga saham yang andal dengan kompleksitas yang lebih rendah dibandingkan pendekatan multivariat konvensional.

Analisis Sentimen Ulasan by.U dengan Pelabelan Rating dan Leksikon Menggunakan Multinomial Naïve Bayes Fatihanursari Dikananda; Bani Nurhakim; Dian Ade Kurnia; Ahmad Rifai; Mugi Praseptiawan
TEMATIK Vol. 13 No. 1 (2026): Tematik : Jurnal Teknologi Informasi Komunikasi (e-Journal) - Juni 2026
Publisher : LPPM POLITEKNIK LP3I BANDUNG

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.38204/tematik.v13i1.2996

Perkembangan layanan telekomunikasi digital mendorong bertambahnya jumlah ulasan pengguna yang digunakan sebagai bahan informasi guna mendukung pengambilan keputusan berbasis data. Penelitian ini bertujuan menganalisis sentimen ulasan aplikasi by.U menggunakan dua metode pelabelan data, yaitu rating-based labeling dan lexicon-based labeling, menggunakan algoritma Multinomial Naïve Bayes (MNB). Metode penelitian menerapkan framework Knowledge Discovery in Databases yang meliputi tahapan selection, preprocessing, transformation, data mining, dan evaluation. Dataset penelitian diperoleh dari Google Play sebanyak 8.000 ulasan berbahasa Indonesia. Tahap prapemrosesan mencakup cleaning, case folding, normalisasi, tokenisasi, stopword removal, serta stemming. Representasi fitur dilakukan menggunakan TF-IDF, sedangkan penyeimbangan data diterapkan melalui metode SMOTE. Evaluasi model dilakukan menggunakan metrik accuracy, precision, recall, dan F1-score dengan skema 10-fold cross validation. Hasil penelitian menunjukkan bahwa pendekatan lexicon-based labeling memberikan performa yang lebih baik dibandingkan rating-based labeling. Pendekatan rating-based menghasilkan accuracy sebesar 82,59%, precision 83,79%, recall 82,59%, dan F1-score 82,43%. Sementara itu, pendekatan lexicon-based memperoleh accuracy sebesar 88,96%, precision 89,69%, recall 88,96%, serta F1-score 88,91%. Temuan tersebut menunjukkan bahwa strategi pelabelan memiliki pengaruh terhadap performa klasifikasi sentimen. Pendekatan berbasis leksikon dinilai lebih efektif karena mampu memahami konteks linguistik dan ekspresi emosional pengguna secara lebih baik dibandingkan pendekatan berbasis rating.

Analisis Kualitas Jaringan Hotspot Menggunakan Metode Quality of Service (QoS) dalam Mendukung Kegiatan Belajar Mengajar Di Sekolah Menengah Kejuruan Negeri 1 Gebang Mochammad Fatha Mudzhaffar; Martanto; Arif Rinaldi Dikananda; Ahmad Rifai
Jurnal Dinamika Informatika Vol. 14 No. 1 (2025): Jurnal Dinamika Informatika Volume 14 Nomor 1
Publisher : Program Studi Informatika Universitas PGRI Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31316/jdi.v14i1.513

Hotspot adalah jaringan nirkabel yang menyediakan akses internet kepada pengguna melalui perangkat Wi-Fi. Kualitas jaringan hotspot sangat penting dalam mendukung berbagai aktivitas, termasuk kegiatan belajar mengajar. Untuk menilai performa jaringan, metode Quality of Service (QoS) digunakan sebagai pendekatan standar dalam mengukur parameter-parameter utama jaringan, seperti throughput, packet loss, delay, dan jitter. Penelitian ini bertujuan untuk menganalisis kualitas jaringan hotspot di SMK Negeri 1 Gebang menggunakan metode QoS. Hasil penelitian menunjukkan bahwa nilai throughput berada dalam kategori "Buruk" hingga "Sangat Buruk" pada jam-jam trafik tinggi (12:00-15:00), dengan nilai berkisar antara 150-318 kbps, sehingga memerlukan optimasi jaringan. Di sisi lain, parameter packet loss tercatat 0%, yang menempatkannya dalam kategori "Sangat Baik." Nilai delay berkisar antara 10,12 ms hingga 30,01 ms, menunjukkan responsivitas jaringan yang baik dalam kategori "Sangat Baik." Sementara itu, nilai jitter berada dalam kategori "Baik" meskipun mengalami sedikit fluktuasi pada jam sibuk. Secara keseluruhan, meskipun performa jaringan dinilai baik dalam aspek packet loss, delay, dan jitter, peningkatan kualitas throughput sangat diperlukan untuk memastikan koneksi yang stabil dan berkualitas, khususnya pada jam trafik tinggi. Temuan ini memberikan dasar untuk pengembangan strategi optimasi jaringan guna mendukung kegiatan pendidikan secara lebih efektif.

Pendampingan Legalitas Usaha Dan Penguatan Profie Digital Bagi UMKM Kota Cirebon Ahmad Rifai; Nana Suarna; Difa Aulia Farradila; Muhammad Zeya Sebastian
AMMA : Jurnal Pengabdian Masyarakat Vol. 4 No. 5 : Juni (2025): AMMA : Jurnal Pengabdian Masyarakat
Publisher : CV. Multi Kreasi Media

Show Abstract | Download Original | Original Source | Check in Google Scholar

Micro, Small, and Medium Enterprises (MSMEs) play a strategic role in supporting regional economic development; however, many still face challenges related to business legality and digital business identity. These limitations restrict their access to government assistance programs, financing opportunities, business partnerships, and broader market promotion. This community service program aimed to enhance the capacity of MSMEs through business legality assistance and digital business profile development for five MSMEs in Cirebon City, namely Warung Kopi/Angkringan, Wonton, Orenz Drink, Noenk Ice, and Afiqah Banana. The program employed a participatory approach consisting of needs assessment, business legality education, assistance in obtaining the Business Identification Number (Nomor Induk Berusaha/NIB), digital business profile development, implementation support, and monitoring and evaluation. The results demonstrated that all participating MSMEs improved their understanding of the importance of business legality, organized their business administration more systematically, and developed comprehensive digital business profiles containing business identity, product information, visual documentation, contact details, and communication channels. Furthermore, the assistance enhanced participants' ability to utilize digital business profiles as effective information and promotional media, thereby strengthening their business identity and increasing consumer accessibility. The integrated mentoring model successfully combined administrative strengthening with digital transformation within a systematic empowerment framework. This program contributes to improving MSMEs' capacity to establish better business governance, enhance consumer trust, and strengthen the competitiveness of local products in a sustainable manner.

Enhancing Election Staff Selection through Decision Tree-Based Classification Rizal Rayyan Firdaus; Nana Suarna; Irfan Ali; Ahmad Rifai
Journal of Artificial Intelligence and Engineering Applications (JAIEA) Vol. 4 No. 2 (2025): February 2025
Publisher : Yayasan Kita Menulis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59934/jaiea.v4i2.768

The selection of competent election committee members is a critical aspect in ensuring the success of a fair and transparent election process. However, the subjective nature of this selection process necessitates a data-driven approach to optimize the selection of officials who meet the required competency criteria. This research aims to classify the competencies of prospective election committee members using the Decision Tree algorithm based on demographic data and technological attributes of the population. The study employs the Knowledge Discovery in Databases (KDD) methodology, which includes the stages of data selection, preprocessing, transformation, data mining, and evaluation. In this process, data collected through various attributes are processed to build a classification model. The Decision Tree algorithm is applied to extract patterns from the data, resulting in a decision tree that can classify individuals into different competency classes based on existing features. The research findings indicate that the Decision Tree algorithm effectively classifies respondents into several competency classes that represent varying levels of skills and interest in the election process. The model shows that Class 4 is the dominant class, indicating that most respondents have moderate competency in technological skills and interest in elections. Class 3 represents individuals with higher technological skills but moderate interest, while Classes 2 and 1 represent individuals with varying combinations of interest and skills. This study demonstrates that using the Decision Tree algorithm in the KDD process is highly effective in objectively classifying the competencies of prospective election committee members. By analyzing the interactions among relevant attributes, the model provides insights that can improve the accuracy of election official selection. This data-driven approach can be adapted to other contexts requiring competency classification, offering broader benefits for various criteria-based selection systems.

K-Means Algorithm for Clustering High-Achieving Student at Madrasah Tsanawiyah Yami Waled Muhammad Hilman; Martanto; Arif Rinaldi Dikananda; Ahmad Rifai
Journal of Artificial Intelligence and Engineering Applications (JAIEA) Vol. 4 No. 3 (2025): June 2025
Publisher : Yayasan Kita Menulis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59934/jaiea.v4i3.771

This study aims to apply the K-Means algorithm to cluster students based on their mathematics grades at Madrasah Tsanawiyah Islamiyyah Yami Waled. By categorizing students into clusters of low, medium, and high academic achievement, the institution can develop more effective and targeted learning strategies. The data consisted of semester mathematics grades from 112 students, analyzed using the K-Means clustering algorithm. Clusters were evaluated using the Davies-Bouldin Index (DBI), with results showing three distinct clusters: Cluster 0 (low achievers, 54 students), Cluster 1 (medium achievers, 37 students), and Cluster 2 (high achievers, 21 students). The DBI score of 0.893 indicates good clustering quality, providing valuable insights for personalized learning approaches.

Optimization of Kebaya Product Grouping Using K-Means Algorithm for Marketing Strategy of Rental Services at Gifaattire Store Nuraeni; Martanto; Arif Rinaldi Dikananda; Ahmad Rifai
Journal of Artificial Intelligence and Engineering Applications (JAIEA) Vol. 4 No. 3 (2025): June 2025
Publisher : Yayasan Kita Menulis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59934/jaiea.v4i3.776

This study aims to implement the K-Means algorithm to improve the kebaya clustering model to support the rental marketing strategy at Gifaattire Store. The K-Means algorithm was used to analyze eight months of historical kebaya rental data, focusing on the attributes of kebaya type and color. Using the Knowledge Discovery in Database (KDD) approach, the research conducted data selection, preprocessing, transformation, data mining, and evaluation of clustering results. Davies-Bouldin Index (DBI) was utilized to assess the quality of clustering, resulting in an optimal value of 6 clusters with a DBI of 0.580. The results showed that each cluster has unique characteristics that reflect customer demand patterns. Cluster 0, the largest cluster, indicates kebayas with high demand but limited color variations. In contrast, Cluster 1 indicates kebayas with a wide variety of colors but specific demand. This information enables Gifaattire Store to design more targeted data-driven marketing strategies and improve stock management efficiency. The research contributes to the development of literature on the application of K-Means in the fashion rental sector and offers practical insights into understanding customer preferences.

Simple Additive Weighting Method for Improving Decision Support Systems Laptop Selection Ika Riantika; Martanto; Arif Rifaldi Dikananda; Ahmad Rifai
Journal of Artificial Intelligence and Engineering Applications (JAIEA) Vol. 4 No. 2 (2025): February 2025
Publisher : Yayasan Kita Menulis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59934/jaiea.v4i2.790

The development of information technology significantly benefits various activities, particularly for students, by facilitating access to information and supporting academic tasks. However, students majoring in Information Technology often face challenges in selecting a suitable laptop due to the wide range of options with varying specifications and prices. This study aims to develop a Decision Support System (DSS) based on the Simple Additive Weighting (SAW) method to assist in choosing the best laptop. The SAW method was selected for its ability to evaluate multiple criteria through a weighting process. The study utilizes five main criteria: price, processor, RAM, storage type, and storage capacity. Data were collected through interviews and observations at the "IComp" laptop store. The analysis process involves matrix normalization and preference value calculation to determine recommendations. The DSS recommends the best laptop based on the highest preference score: Lenovo IP Flex 5 (0.78), followed by Lenovo IP3 (0.77) and HP Pav14 (0.76). The results indicate that these laptops offer an optimal balance between performance and price. The web-based sy stem designed accelerates the evaluation process, enhances objectivity, and improves user accessibility. The implementation of the SAW method proves effective and accurate in determining the best laptop, particularly in scenarios combining cost and benefit criteria. The system successfully meets the needs of Information Technology students by providing relevant and reliable results. This study successfully develops a DSS using the SAW method for selecting the best laptop. The system designed is effective and reliable for multi-criteria decision-making. Future research can integrate real-time data and broader user surveys to improve result generalization, making it applicable to other product selection contexts.

Title

Found 10 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 10 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 10 Documents
Search