Claim Missing Document
Check
Articles

Found 7 Documents
Search

Implementation of PageRank Algorithm for Visualization and Weighting of Keyword Networks in Scientific Papers Lubis, Adyanata; Prasiwiningrum, Elyandri
Journal of Applied Data Sciences Vol 4, No 4: DECEMBER 2023
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v4i4.138

Abstract

Papers are written works that contain thoughts about a particular problem or topic that are written systematically accompanied by logical analysis. Scientific papers are often found on the internet or in libraries for various titles of scientific papers, citations or references can be found in every scientific paper and can be obtained easily, but to display all citations in scientific papers in the form of visualization cannot be done easily. Visualizing the citation network of scientific papers in the form of a graph, with nodes representing research papers and edges representing the relationship between researchers' scientific papers and other scientific papers based on scientific paper citations. This research uses the pagerank algorithm to create a keyword network that has a high relationship and potential relevance in a data library. This research is the first research in using the pagerank algorithm and testing its accuracy by comparing with KNN and linear clustering. The presentation displays the citation of scientific papers based on the size of the node by showing the number of citations of the scientific paper. It can be concluded that all processes in the system have run according to design, and functionally the visualization system and weighting of the scientific paper citation network are in accordance with the design. The results obtained from 51 articles, this algorithm produces a visual user interest of 81.60%, compared to the accuracy of the data suitability produced by the linear clustering and KNN algorithms in the form of 71.22% and 61.34%, helping to facilitate the search for scientific papers in large quantities.
Pengendalian Persediaan Darah untuk Pasien dengan Hemoglobin Rendah Menggunakan Metode Backpropagation Prasiwiningrum, Elyandri
Jurnal Informatika Ekonomi Bisnis Vol. 4, No. 3 (September 2022)
Publisher : SAFE-Network

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (550.194 KB) | DOI: 10.37034/infeb.v4i3.153

Abstract

The Blood Transfusion Unit (UTD) of the Rokan Hulu Regional General Hospital (RSUD) has an important role to fulfill the demand for blood from patients. Patients who need blood donation are patients with low hemoglobin levels. The problem faced by the UTD-RS is that they have not been able to meet the needs of each patient's blood request optimally. The reason is because it is not able to predict the need for blood that will come. To see the pattern of blood demand and then determine the appropriate inventory control method to assist the planning process for the fulfillment of blood supply at UTD in the next period. Materials (data) and Methods: The data processed in this study were patient data and blood demand data from January 2020 to January 2021. The data were sourced from the Laboratory Installation and UTD at the Rokan Hulu Hospital. The data is divided into training data and testing data. Then the blood demand data is processed by normalizing it first and then the prediction process is carried out using the Backpropagation method. Then analyzed and tested with the help of Matlab software. This study uses the best network architecture pattern produced is 5-5-1 with an accuracy rate of 68% and a Mean Squared Error value of 0.198. The backpropagation method used is able to help UTD Rokan Hulu Hospital to find out the blood needs that must be met so that the blood supply can be controlled. So that every blood request from patients with low hemoglobin can be met quickly.
Application Method Certainty Factor in Electrical Damage Zulkifli, Akhmad; Riandini, Meisarah; Hayadi, B. Herawan; Prasiwiningrum, Elyandri
Journal of ICT Applications System Vol 2 No 1 (2023): Journal of ICT Aplications and System
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56313/jictas.v2i1.236

Abstract

Electricity is need main For life people human . Electricity is used man For various type activity human . Electricity plays a big role for life , like For lighting , cooking , and so on . Almost all activity daily use electricity . Almost every home in Indonesia, both in the city nor village Already trellis with electricity . For stream and distribute electricity to each home , office nor distant institutions _ away , then needed Transformer Distribution . Transformer Distribution This own objective use special that is, to lower voltage tall to voltage low , so that the voltage used in accordance with equipment ratings electricity customer or load in general . For help in handle problem damage Transformer distribution , then one is needed branch from Knowledge computer that is System Expert . System Expert is system based computer that uses knowledge , facts , and techniques reasoning in solve problem , which usually is only can completed by one expert in field certain . (Putri, 2020). The method used in research _ This is Certainty Factor. Study This apply certainty factor method For role in diagnose damage to electricity . Based on results discussion on with choose one _ damage namely P1 ( Oil transformer go out from the transformer body ) on the study case obtained decision level accuracy that is as big That's 5.650198%. means system expert certainty factor method can overcome damage and deliver results diagnosis good at damage electricity
Projection and Visualization of Health Worker Data in Indonesia (2015–2017) Using Google Looker Studio Prasiwiningrum, Elyandri
Journal of ICT Applications System Vol 2 No 2 (2023): Journal of ICT Aplications and System
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56313/jictas.v2i2.398

Abstract

Projection of health worker data in Indonesia during 2015–2017 highlights the need for efficient workforce planning to address rising healthcare demands. This study examines the distribution and projection of ten healthcare professions, including nurses, midwives, pharmacists, and general practitioners, using Google Looker Studio for visualization. Secondary data analysis was utilized to process information obtained from credible sources, ensuring relevance and accuracy. The challenge lies in understanding workforce disparities and predicting future needs to optimize hospital operations effectively. Google Looker Studio was employed to create interactive dashboards, simplifying data interpretation and enhancing decision-making capabilities. Results indicate a continuous increase in healthcare personnel over time, with 2,172 professionals recorded during the observed period. Visualization provides insights into workload distribution and the adequacy of healthcare workers across regions. This research offers a scalable solution for projecting workforce trends and supports long-term healthcare planning in Indonesia
Classification Of Palm Oil Maturity Using CNN (Convolution Neural Network) Modelling RestNet 50 Prasiwiningrum, Elyandri; Adyanata Lubis
Decode: Jurnal Pendidikan Teknologi Informasi Vol. 4 No. 3: NOVEMBER 2024
Publisher : Program Studi Pendidikan Teknologi Infromasi UMK

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51454/decode.v4i3.822

Abstract

Accurate classification of palm fruit maturity levels is very important to optimize harvest time and increase production efficiency in the palm oil industry. Traditional methods that rely on visual assessment of factors such as fruit shedding and skin discoloration are prone to human error. To overcome this limitation, this research applies deep learning techniques, specifically using Convolutional Neural Network (CNN) with ResNet-50 architecture, to classify Fresh Fruit Bunches (FFB) into two stages of maturity: unripe and ripe. The model is trained and validated using a combination of data augmentation techniques to improve model performance. Various configurations were tested, including variations in data sharing, optimizer, and learning rate. The optimal configuration—90/10 training and validation data split, Adam optimizer, and learning rate of 0.0001—resulted in excellent model performance. The ResNet-50 model achieved 97% accuracy, with 96% precision, 98% recall, and an F1 score of 97%. This metric reflects the high reliability of the model in classifying palm fruit maturity levels, significantly reducing classification errors compared to traditional methods. This research highlights the transformational potential of deep learning to improve maturity classification in the palm oil industry, by offering a more efficient, accurate and automated approach. Further research should focus on expanding the dataset to increase model robustness as well as exploring real-time implementation to further improve decision making in palm oil production. This approach promises to increase agricultural efficiency by ensuring optimal harvest timing and better resource management.
Electre Method Decision Support System for Concrete Type Selection in Building Structures Prasiwiningrum, Elyandri; Rasna, Rasna; Heka Ardana, Putu Doddy; Nugroho, Fifto; Aini, Qurrotul
International Journal of Engineering, Science and Information Technology Vol 5, No 2 (2025)
Publisher : Malikussaleh University, Aceh, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52088/ijesty.v5i2.843

Abstract

Choosing the correct type of concrete in building construction is a crucial step that significantly affects the structure's quality, durability, and safety. Concrete, as the most widely used building material, has various types with different characteristics, such as compressive strength, tensile strength, modulus of elasticity, and durability. During this time, people who make concrete often ignore the strength of the concrete itself. They do not care what will happen if the manufacture of concrete is not by the recommended concrete construction. Concrete is the primary material for constructing a building such as a building. The quality of concrete is determined by its constituent materials, which include hydraulic cement, coarse aggregate, fine aggregate, water, and other additives. The concrete mix determines the strength of the concrete. If the building is built with unsuitable concrete, it will be quickly destroyed during natural disasters such as earthquakes. Based on this, the concrete type selection must be precise and accurate. Decision Support Systems are used in this research to provide additional input to decision-makers. Decision support systems can deliver maximum results by using algorithms or methods. The Electre method is one of the multicriteria decision-making methods based on ranking by pairwise comparisons of existing alternatives based on appropriate criteria. Overall, this research is expected to significantly improve the quality of decision-making in selecting concrete types, resulting in a safer, more durable, and more efficient building structure. The results obtained after inputting criteria values and alternative values are concrete types such as reinforced concrete, precast concrete, and lightweight concrete.
Automatic Food Label Detection in Images Using Convolutional Neural Network with Food-101 Dataset Natasya, Ccely; Aisyah, Nur; Prasiwiningrum, Elyandri; Yulfita Aini
Journal of ICT Applications System Vol 4 No 1 (2025): Journal of ICT Aplications and System
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56313/jictas.v4i1.432

Abstract

automatic detection of food labels from digital images has emerged as a crucial application in dietary analysis, nutrition monitoring, and smart culinary systems. This study presents the implementation of a Convolutional Neural Network (CNN) model for food label recognition using the Food-101 dataset, which consists of over 101,000 images from 101 distinct food categories. The proposed system follows a systematic pipeline that includes image resizing, normalization, and data augmentation to enhance model robustness and performance. The CNN architecture is designed with multiple convolutional and pooling layers, followed by dense and softmax output layers for final classification. The training was conducted using the Adam optimizer with a learning rate of 0.0001, batch size of 32, and dropout regularization to prevent overfitting. Experimental results demonstrate a classification accuracy of 24.45% after one training epoch, highlighting both the capability and limitations of the baseline CNN model. Despite moderate accuracy, the model successfully identifies visually distinguishable food items and sets a foundation for future improvements through transfer learning and fine-tuning. This research confirms the potential of CNN-based models for food label detection and provides insights for the development of more accurate food recognition systems in health, dietary, and culinary applications