Telematika
Not a member yet
319 research outputs found
Sort by
Development of Augmented Reality-based Space Building Learning Media Pengembangan Media Pembelajaran Bangun Ruang Berbasis Augmented Reality
Full text linkKonsep bangun ruang kerap menjadi tantangan bagi siswa kelas 7 SMP. Kesulitan dalam membayangkan bentuk tiga dimensi dari gambar dua dimensi seringkali menghambat pemahaman mereka. Salah satu tantangan utama adalah rendahnya pemahaman siswa terhadap konsep abstrak dalam bangun ruang. Penelitian ini bertujuan untuk memenuhi kebutuhan siswa untuk memahami konsep bangun ruang melalui pengembangan media pembelajaran interaktif yang lebih menarik dan mudah dipahami. Penelitian ini menggunakan metode R&D dengan model pengembangan ADDIE. Hasil dari penilitian ini mendapatkan penilaian dari berbagai aspek yaitu ahli media, ahli materi, dan siswa. Hasil dari penilaiannya mengindikasikan tingkat keberhasilan yang tinggi dengan persentase Ahli Media sebesar 92,39%. Kemudian Ahli Materi menunjukkan persentase sebesar 97,05%. Sedangkan penilaian dari siswa menunjukkan persentase sebesar 95,96%. Sehingga dapat diperoleh kesimpulan bahwa pengembangan media pembelajaran bangun ruang berbasis augmented reality (AR) untuk siswa kelas 7 SMP menunjukkan hasil yang sangat layak untuk digunakan
Performance Comparison of VGG-19 and DenseNet-121 Architectures for Rice Plant Disease
Full text linkRice (Oryza sativa L.) is a major food source that often faces the challenge of crop failure due to various plant diseases. These diseases not only reduce productivity, but are also exacerbated by farmers\u27 limited knowledge in recognizing symptoms and reliance on manual diagnosis that takes a long time. This study aims to compare the performance of two Convolutional Neural Network (CNN) architectures, namely VGG-19 and DenseNet-121, in classifying rice plant diseases based on image processing. Low accuracy and overfitting are problems that are often observed when small datasets are used to train deep learning models, such as Convolutional Neural Networks (CNN). In this study, modifications were made to the VGG-19 and DenseNet-121 architectures so that the model can achieve good accuracy and reduce the risk of overfitting despite using small datasets. The dataset consists of 11,790 images in 9 classes, which are divided into 7545 training data, 1887 validation data, and 2358 testing data. After the training data is segmented, the total number of images in the dataset is 23,580. Before modification, the DenseNet-121 model achieved the highest accuracy of 50.45% and F1-score of 44.83%, while VGG-19 achieved the highest accuracy of 13.84% and F1-score of 7.39%. After making modifications to both models, the test results show that DenseNet-121 achieved an accuracy of 97.76% and F1-score of 96.31%, while VGG-19 achieved an accuracy of 84.82% and F1-score of 87.52%. The advantage of DenseNet-121 lies in its ability to process features more efficiently, resulting in more accurate predictions than VGG-19. This research contributes to the selection of the best model architecture to support automatic diagnosis of rice plant diseases, which is relevant to the agricultural sector in Indonesia.
Implementation of the Random Forest algorithm to predict rice needs in DKI Jakarta
Full text linkPurpose : to build collaborative partners between government institutions and universities in food processing, especially rice, by predicting rice needs in the DKI Jakarta area.Design/methodology/approach:The approach in this research uses the Random Forest algorithm which functions to predict rice needs in the DKI Jakarta area.Results: rice demand prediction application with evaluation values Mean Squared Error 207.86, Mean Absolute Error 9.43, MAPE 0.048, Root Mean Squared Error 14.4, accuracy 0.63Originality/value/state of the art:research using data from BAPANAS, Cipinang Main Market, with 2 datasets of rice stock, population, year and rice consumption using a random forest algorithm to predict rice needs in the DKI Jakarta area
Expert System for Coffee Leaf Disease Classification With Convolutional Neural Networks
Full text linkTujuan: Untuk mengembangkan sistem pakar menggunakan Convolutional Neural Network (CNN) untuk mengklasifikasikan Penyakit Daun Kopi ke dalam empat kategori (penambang, phoma, karat, dan tanpa penyakit), memberikan diagnosis yang cepat dan akurat bagi petani.Desain/metodologi/pendekatan: Penelitian ini mengimplementasikan sistem pakar berbasis CNN menggunakan kumpulan data 1.664 gambar daun kopi. Metodologi ini mencakup praproses data dengan standarisasi dan penambahan gambar, pengembangan model CNN dengan enam blok konvolusional, pelatihan model dengan pengoptimal Adam, dan evaluasi komprehensif menggunakan metrik validasi.Temuan/hasil: Sistem mencapai akurasi validasi 97,66% dengan waktu pemrosesan yang efisien (49-161 ms per prediksi). Model menunjukkan keandalan yang tinggi di semua kategori penyakit, dengan tingkat keyakinan secara konsisten di atas 80% dan mencapai hingga 100% untuk kondisi tertentu.Orisinalitas/nilai/keadaan terkini: Penelitian ini memperkenalkan integrasi baru sistem pakar dengan teknologi CNN untuk klasifikasi Penyakit Daun Kopi, yang menawarkan akurasi yang lebih unggul dibandingkan dengan pendekatan probabilistik tradisional. Sistem ini menyediakan visualisasi waktu nyata dan tingkat keyakinan untuk setiap prediksi, menjadikannya alat praktis bagi petani
Decision Support System For Selecting The Best Non PNS Using Fuzzy Analytical Hierarchy Process Method
Full text linkThe most important aspect of human resource management greatly impacts the success of the agency by selecting the best employees. The best employees are those who have certain skills that can contribute to the success of the agency. Non-Civil Servant Government Employees (PPNPN) are honorary personnel recognized by the Government and the State who are needed by agencies to assist a job in the unit in need. Problems occur when selecting employees with performance every month, namely the absence of a system that supports decisions to determine the selection of the best non-civil servants in Cipamokolan Bandung Village, so that the selection of the best employees is currently still done manually. Thus the selection of the best employees is not accurate and very subjective. The purpose of this research is to provide solutions to these problems by creating a decision-making system in selecting the best non-civil servants in Cipamokolan Bandung Village. The method used is by applying the Fuzzy Analytical Hierarchy Process method which can overcome uncertainty and inaccuracy in judgment. The research results obtained from the selection of non-civil servant government employees using the web-based Fuzzy Analytical Hierarchy Process method can create efficiency and effectiveness in selecting the best employees in Cipamokolan Bandung Village
Improving the Efficiency of Water Meter Reading at Perumdam Tirta Kerta Raharja Using Microcontroller-Based Implementation of the YOLOv9 Method: Peningkatan Efisiensi Pembacaan Angka Meter Air Perumdam Tirta Kerta Raharja Berbasis Mikrokontroler dengan Penerapan Metode Yolov9
Full text linkManual water meter reading remains a challenge for Perumdam Tirta Kerta Raharja due to its labor-intensive process, susceptibility to human errors, and inefficiency. This study aims to develop an automated water meter reading system using YOLOv9 and a microcontroller to improve efficiency and data accuracy. The model was trained using a dataset of water meter images under various lighting conditions and viewing angles. Evaluation results indicate that the 20-epoch configuration is the best model, achieving 99,91% accuracy, 91,16% average precision, and 91,04% average recall. The developed system successfully detects digits in real-time with high accuracy when deployed on a Raspberry Pi-based platform. However, the model still faces challenges in detecting the Background class. With further optimization, this system can be widely implemented to enhance operational efficiency in Perumdam and related industries
Anomaly Detection of Automatic Rain Gauge Measurement Using Artificial Neural Network Long Short Term Memory Method
Full text linkPurpose: The purpose of this research is to accurately detect anomalies in the results of automatic rain gauge measurements using the Long Short Term Memory (LSTM) method, so that measurement errors can be immediately identified and the equipment can be repaired immediately. Design/methodology/approach: Detection of anomalies from rain gauge measurements is carried out using quality control (QC) methods based on range and step check, spatial check and error check which produce anomaly labels which are totaled to become Total Anomaly QC. Total Anomaly QC is transformed via one-hot encoding and then the results of the Total QC data transformation are used to build an anomaly detection classification model using the LSTM algorithm.Findings/result: The model performance was tested with a confusion matrix. LSTM is able to classify data anomalies in the western, eastern and coastal clusters quite well. The accuracy value of these clusters is more than 0.9, so that >90% of the anomalies are classified correctly. The results of this research can improve BMKG\u27s ability to detect rainfall measurement anomalies from automatic rain gauges and assist in maintaining the validity of rainfall data so that equipment maintenance is carried out on time.Originality/value/state of the art: This research uses different methods and parameters from previous research. The results obtained are quite satisfactory as shown by an accuracy above 0.9
Performance Analysis and Accuracy of Machine Learning Algorithms for Heart Disease Prediction
Full text linkPurpose: This research aims to analyze the performance and accuracy of machine learning algorithms in predicting heart disease, which is a cause of death throughout the world.Design/methodology/approach: The algorithms analyzed include Logistic Regression, Naive Bayes, Support Vector Machine, K-Nearest Neighbors, Decision Tree, Random Forest, XGBoost, and Neural Network. A publicly available dataset containing patients\u27 medical records was utilized, with the methodology encompassing data collection, Exploratory Data Analysis (EDA), model training, and performance evaluation.Findings/result: The results indicate that the Random Forest algorithm achieved the highest accuracy with an accuracy of 90.16%, followed by Logistic Regression and Naive Bayes with accuracies of 85.25%. The K-Nearest Neighbors algorithm exhibits the lowest accuracy at 67.21%.Originality/value/state of the art: This research highlights the advantages of certain machine learning algorithms in predicting heart disease and contributes knowledge to early detection technology in the health sector
Implementation of Natural Language Processing with Deep Learning on Chatbot UKT (Uang Kuliah Tunggal) University
Full text linkTujuan: Penelitian yang dilakukan adalah membuat chatbot yang mampu mengklasifikasikan teks soal berdasarkan maksud/label secara lebih tepat, dan membantu memudahkan pelajar dalam memperoleh informasi tambahan terkait layanan UKT (Uang Kuliah Tunggal) dan biaya pendidikan. Chatbot ini mampu mengatasi keterbatasan yang dialami oleh Tata Usaha Fakultas Teknik (TU FT) Universitas Tidar, seperti human delay dalam merespon email atau live chat, keterbatasan jam layanan (office hour), dan keterbatasan informasi pada website kampus. Chatbot dikembangkan menggunakan pendekatan Natural Language Processing (NLP) dan algoritma Deep Learning BiDirectional Long Short-Term Memory (BiLSTM). Hasil sistem chatbot diintegrasikan ke dalam aplikasi Telegram untuk melihat tingkat kepuasan pengguna setelah berinteraksi.Desain/metodologi/pendekatan: Proses penelitian ini diawali dengan pengumpulan dataset berupa pertanyaan dan jawaban yang telah diberi tag atau label dalam format file JSON. Dataset tersebut dilakukan proses normalisasi teks atau preprocessing Natural Language Processing (NLP), dimana pada tahap ini dilakukan lower case atau case lipat, penghapusan tanda baca, penghapusan spasi berlebih, stopword dan stemming dengan perpustakaan Sastrawi, tokenisasi, dan padding. Selanjutnya dilakukan proses split dataset dengan Bagi 80% pelatihan 20% pengujian, sebelum diolah menjadi fitur ekstraksi, menggunakan FastText untuk penyematan kata. Selanjutnya dilakukan proses klasifikasi teks pertanyaan dengan model BiDirectional Long Short-Term Memory (BiLSTM) dan dilanjutkan evaluasi dengan matriks konvergensi. Tahap terakhir, yaitu integrasi chatbot ke dalam bot Telegram, kemudian dilakukan pengujian pengguna terhadap chatbot dan pengukuran tingkat kepuasan dengan metode Customer Satisfaction Score (CSAT).Temuan/Hasil: Model akurasi klasifikasi menghasilkan nilai sebesar 96,05% dan pengujian pengguna dengan penerapan metode Customer Satisfaction Score (CSAT) memberikan tingkat kepuasan rata-rata pada rentang 4 (Puas) dan 5 (Sangat Puas) dengan hasil sebesar 88,86% berdasarkan poin-poin berikut \u27Kepuasan terhadap jawaban yang diberikan\u27, \u27Pemahaman pertanyaan dan jawaban mudah dipahami\u27, dan \u27Kinerja sesuai harapan\u27.Orisinalitas/nilai/keadaan terkini: Penelitian tentang klasifikasi teks pertanyaan pada chatbot dengan pendekatan Natural Language Processing (NLP) dan model BiDirectional Long Short-Term Memory (BiLSTM) untuk menangani permasalahan pertanyaan jawab layanan UKT (Uang Kuliah Tunggal) dan biaya pendidikan Fakultas Teknik Universitas Tidar belum pernah dilakukan sebelumnya
Development of Employee Job Satisfaction Survey System with Access Management Based on Job Position Using Scrum Framework (Case Study: Era Medika Hospital)
Full text linkEra Medika Hospital is a privately-owned general hospital committed to quality service and patient safety by continuously evaluating and improving various aspects, including the quality of human resources (HR) or staff, whose service quality is greatly influenced by their job satisfaction. To assess employee job satisfaction, the HR management routinely conducts a survey in the form of a questionnaire every six months using the XYZ Application. However, from experience with this application, HR management has encountered issues such as duplicate survey submissions by employees, difficulties managing questionnaire access rights based on employee positions, and manual separation of suggestion responses by category.
This study aims to develop a web-based Employee Job Satisfaction Survey System at Era Medika Hospital with access management based on employee roles using the Scrum framework. The system development applies an agile Scrum approach, dividing the design, implementation, and testing phases into three sprints, each followed by sprint reviews with stakeholders. The developed system features solutions addressing the encountered problems, such as login/sign-in using employee identification numbers (NIP) to ensure each employee submits the survey only once, questionnaire data management including access control by position, and automatic categorization of suggestion responses starting from the survey input.Evaluation results indicate the system meets both functional and non-functional requirements as expected by users. Achieving the research objectives, the system is expected to operate optimally, support data-driven decision-making, and improve employee satisfaction and productivity, ultimately contributing to enhancing the quality of healthcare services provided by Era Medika Hospital