Tamuntuan, Virginia
Unknown Affiliation

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search

Analisis Perbandingan Kinerja Algoritma Klasifikasi Pada Mahasiswa Berpotensi Dropout Tamuntuan, Virginia; Kusrini, Kusrini; Kusnawi, Kusnawi
Building of Informatics, Technology and Science (BITS) Vol 6 No 2 (2024): September 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i2.5658

Abstract

This research aims to compare the performance levels of two data mining classification algorithms, namely Support Vector Machine and Neural Network Backpropagation, using the K-fold cross-validation method. The data used consists of graduates from 2019 to 2023 at STMIK Multicom Bolaang Mongondow. A total of 80% of the 200 data points were used as training data, while the remaining 20% were used as testing data. K-fold cross-validation was conducted with K set to 5. The results of the study indicate that the Support Vector Machine algorithm achieved an accuracy of 80%, recall of 80%, and precision of 35%, while the Neural Network Backpropagation algorithm achieved an accuracy of 77%, recall of 63%, and precision of 44%.
Identification of Lumpy Skin Disease in Cattle with Image Classification using the Convolutional Neural Network Method Sentoso, Thedjo; Ardiansyah, Fachri; Tamuntuan, Virginia; Wangsa, Sabda Sastra; Kusrini, Kusrini; Kusnawi, Kusnawi
Sistemasi: Jurnal Sistem Informasi Vol 13, No 3 (2024): Sistemasi: Jurnal Sistem Informasi
Publisher : Program Studi Sistem Informasi Fakultas Teknik dan Ilmu Komputer

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32520/stmsi.v13i3.2569

Abstract

One of the problems often faced by cattle farmers is related to diseases in their cattle where one of the cattle diseases whose transmission rate is very fast is Lumpy Skin Disease (LSD). Currently, to identify the health of livestock, especially in cattle, is still very dependent on experts and of course this takes time, resulting in delays in the prevention and treatment of diseases in cattle, especially this LSD disease. The Convolutional Neural Network (CNN) algorithm is one of the algorithms can used for image classification of cows whether the cow is healthy or Lumpy. The stages of this research start from problem identification, literature study, data collection, algorithm implementation, testing, and performance evaluation results of the algorithm on cattle disease data. In this research, testing was conducted using three architectures for CNN: VGG16, VGG19, and ResNet50. The results of the experiment showed that VGG16 was the most effective architecture compared to VGG19 and ResNet50, with a training accuracy of 95.31% and a loss value of 0.1292, as well as a testing accuracy of 96.88% and a loss value of 0.102.
User Interface Design of a Web- and Mobile-Based Sales Information System for GBL Chicken Modayag MSME Tamuntuan, Virginia; Mamangki, Hamdi; Tuntun, Ritham
Bulletin of Network Engineer and Informatics Vol. 3 No. 2 (2025): BUFNETS (Bulletin of Network Engineer and Informatics) October 2025
Publisher : GWEX NET PUBLISHER

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59688/rnbbp832

Abstract

GBL Chicken Modayag is a micro-scale culinary business that still relies on manual transaction recording and has not optimally utilized digital media for sales and marketing activities. This condition leads to a high risk of data loss, limited promotional reach, and ineffective dissemination of menu and promotional information to customers. This study aims to design a web and mobile-based sales information system interface as a digital solution to these problems. The research method includes direct observation and interviews with the business owner to analyze the existing system, followed by the design of a proposed system using flowcharts, Entity Relationship Diagrams (ERD), and user interface design developed with Figma. The results of this study produce a sales information system design that supports menu management, online and offline ordering, transaction processing, sales reporting, and communication through a chat feature involving multiple user roles, namely admin, cashier, courier, and customer. The proposed system design is expected to improve sales management efficiency, reduce the risk of transaction data loss, and expand the marketing reach of MSMEs through web and mobile-based digital platforms. However, this study is limited to the design stage and does not include system implementation or usability evaluation, so the results serve as a conceptual reference for future development and testing.