cover
Contact Name
muhammad siddik hasibuan
Contact Email
mhdsiddikhasibuan@gmail.com
Phone
-
Journal Mail Official
cosieaira@gmail.com
Editorial Address
Jl Pukat Banting IV NO 41 Medan Kecamatan Medan Tembung Kode Pos 20224
Location
Kota medan,
Sumatera utara
INDONESIA
Journal of Computer Science and Informatics Engineering
ISSN : -     EISSN : 28278356     DOI : -
Core Subject : Science,
Artificial Intelligence Machine Learning Natural Language Processing Computer Vision Text Speech Text Mining Data mining Cryptography Data visualization Expert System Deep Learning Fuzzy Logic IoT and smart environments Neural Networks Pattern Recognition Image Processing Optimization Digital Signal Processing Networking Technology Web intelligence
Articles 96 Documents
Optimization of Diabetes Disease Classification Using Learning Vector Quantization Algorithm(LVQ) Sianipar, Eska Avelina; Yasin S, Muhammad
Journal of Computer Science and Informatics Engineering Vol 4 No 2 (2025): April
Publisher : Ali Institute of Research and Publication

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55537/cosie.v4i2.1121

Abstract

Diabetes is a chronic disease that attacks humans. Diabetes is caused by high levels of sugar in the human body. Diabetes attacks the metabolic function of the body where the body cannot digest or use high levels of sugar in the human body. Diabetes is classified as a dangerous chronic disease because it has a very fatal impact on humans, especially if complications of the disease have occurred. This study aims to develop a method for classifying diabetes using the Learning Vector Quantization (LVQ) method. Clinical data from patients diagnosed with diabetes and patients who are not diagnosed with diabetes will be analyzed to identify patterns related to this disease. Attribute data such as pregnancies, glucose, blood pressure, skin thickness, insulin, BMI, diabetes pedigree function and age will be used as variables in this analysis. The calculation results show that the Winning Class (Lowest value from the weighted results). So the classification results show an assessment to class 2 because the closest value is 70.002265497656 while the furthest value is 182.52975490778 in class 1
E-Commerce Consumer Data Clustering Using K-Means Algorithm And Kaggle Dataset Sabrina, Ade Eka; Yasin, Muhammad
Journal of Computer Science and Informatics Engineering Vol 4 No 2 (2025): April
Publisher : Ali Institute of Research and Publication

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55537/cosie.v4i2.1123

Abstract

Online shopping has become a part of people's lifestyle. This is because of the many conveniences obtained to meet primary to tertiary needs. The current condition, consumer purchase transaction history data has not been utilized optimally so that it is less effective. This makes the company feel that it has not been optimal in meeting customer expectations in increasing consumer loyalty. In addition, the current marketing strategy is also considered ineffective because the offers made by the company to each consumer are still general, the company has not offered products or promotions that are really needed by consumers. Through this study, the author tries to provide solutions to companies to increase the efficiency of marketing strategies that greatly influence increasing consumer loyalty by conducting clustering analysis of e-commerce consumer data. The purpose of this study is to design and create an application for clustering e-commerce consumer data using the k-means algorithm and the kaggle dataset. The data used in this study is e-commerce consumer data. The output results of solving the problem of clustering e-commerce consumer data. It can be seen based on the results of the payment data grouping for Cluster 1 (Electronic Payment): Total Payment: 72,000,000.00, Cluster 2 (Shoe Payment): Total Payment: 6,600,000.00, Cluster 3 (Clothing Payment): Total Payment: 2,940,000.00 and Cluster 4 (Cosmetic Payment): Total Payment: 105,000,000.00
Smart Farming System Design Based on Long Range and Internet of Things Rifki Fauzan Anshori; Muhammad Saleh; Abqori Aula
Journal of Computer Science and Informatics Engineering Vol 4 No 2 (2025): April
Publisher : Ali Institute of Research and Publication

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55537/cosie.v4i2.1128

Abstract

Mustard greens require an efficient irrigation system to support optimal growth and increase productivity. One irrigation system that can be used is a drip irrigation system. This study designs a smart farming system based on the Internet of Things (IoT) and Long Range (LoRa) technology to automatically manage drip irrigation in mustard greens. This smart farming system uses a capacitive soil moisture sensor to monitor soil moisture, a soil pH sensor to monitor soil pH, and an HC-SR04 ultrasonic sensor to monitor the water level in the water tank. The microcontroller used in this smart farming system is ESP32 and the communication module used is LoRa SX1278. This smart farming system uses a drip irrigation system that will be displayed on the Blynk application. This system allows remote monitoring of plant conditions and automatic drip irrigation based on soil moisture levels. The results of testing the smart farming system using a drip irrigation system showed the accuracy of the ultrasonic sensor to measure water levels of 96.71%, the soil moisture sensor to measure soil moisture levels of 95.65%, and the soil pH sensor to measure soil pH of 97.46%. Communication using LoRa in areas with minimal interference can reach a distance of up to 321.66 meters. While in areas with a lot of interference it can reach a distance of up to 100 meters. The average LoRa data transmission time is 3.185 seconds
Public Complaints Application at Binjai City Police Using the Waterfall Method to Improve the Performance of Binjai District Police Dimas, Dimas; Lubis, Aidil Halim
Journal of Computer Science and Informatics Engineering Vol 4 No 3 (2025): July
Publisher : Ali Institute of Research and Publication

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55537/cosie.v4i3.1130

Abstract

The web-based public complaints application in Binjai City aims to improve the effectiveness and transparency of police services. The system allows the public to submit reports online, monitor the status of complaints, and communicate more efficiently with civil servants. With features such as real-time notifications, police database reporting, and integration, the app speeds up responses to symptoms and improves the accountability of the local police in Binjai. The use of this technology also reduces manual bureaucracy, speeds up case solutions, and increases public trust in the police. System testing shows that the app can optimize complaint workflows and provide more responsive solutions. Therefore, this app can be an innovative model in modernizing police services in the digital era.
Comparison of K-Nearest Neighbor (KNN) and Support Vector Machine (SVM) Algorithms in Predicting Customer Satisfaction Pratama, Subhan Rizky; Fajri, Ika Nur
Journal of Computer Science and Informatics Engineering Vol 4 No 3 (2025): July
Publisher : Ali Institute of Research and Publication

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55537/cosie.v4i3.1160

Abstract

This study compares the K-Nearest Neighbor (KNN) and Support Vector Machine (SVM) algorithms in predicting customer satisfaction at Warung Makan Indomie (Warmindo). The research process consists of four stages, namely: data collection, data processing, model formation, and model evaluation. This study aims to compare the performance of two classification algorithms, namely K-Nearest Neighbor (KNN) and Support Vector Machine (SVM), in predicting customer satisfaction levels based on survey data. The evaluation was carried out using accuracy metrics and classification reports to determine the level of precision, recall, and f1-score of each algorithm. The evaluation results show that both algorithms have the same accuracy of 70%. KNN excels in f1-score in class 2 (0.70), while SVM excels in precision in class 2 (0.79). with an average score of both algorithms being 0.61. These results indicate that both KNN and SVM are feasible to use, depending on the performance priority per class
Comparison of Naïve Bayes and Dempster Shafer Algorithms for the Diagnosis of ARI Diseases Haikal, Baginda Fikri; Hasibuan, Muhammad Siddik; Rifki, Mhd Ikhsan
Journal of Computer Science and Informatics Engineering Vol 4 No 3 (2025): July
Publisher : Ali Institute of Research and Publication

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55537/cosie.v4i3.1161

Abstract

Acute Respiratory Infection (ARI) has a high prevalence in Indonesia, but the manual diagnosis process faces challenges such as limited medical personnel and uncertainty in symptom analysis. This study developed and compared two AI methods, namely Naïve Bayes and Dempster-Shafer, in a web-based expert system to diagnose ARI. Symptom and disease data were collected from literature and experts, then implemented in a PHP and MySQL-based system. Naïve Bayes was used for probability-based classification, while Dempster-Shafer handled uncertainty. Testing was conducted on one case of ARI. Naïve Bayes produced a probability of 21.99% for Pneumonia, while Dempster-Shafer provided a combined probability of 61.6% for five diseases, including Colds, Acute Pharyngitis, and Epiglottitis. The results show that Naïve Bayes is suitable for consistent single diagnoses, while Dempster-Shafer is more appropriate for conditions with overlapping symptoms and uncertain data
Smart Shopping System: Integration of Web Technology and IoT for Digital Transformation of Small and Medium-Scale Retail Stores Banjarnahor, Wiwin Sry Adinda; Sinuraya, Junus; Prayudani, Santi; Tumanggor, Orli
Journal of Computer Science and Informatics Engineering Vol 4 No 2 (2025): April
Publisher : Ali Institute of Research and Publication

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55537/cosie.v4i2.1217

Abstract

The modern retail sector faces challenges in meeting consumer expectations for efficient and convenient shopping experiences. This research develops a web-based Si-Smart Shop system integrated with Internet of Things (IoT)-based smart shopping cart to optimize retail store operations. The research problem formulation includes reducing customer queue density, providing real-time price information independently, and automating the price label creation process that was previously done manually. The research methodology uses a system development approach with the waterfall model. Implementation results show that the system successfully optimizes operational workflow by reducing consumer waiting time and minimizing price calculation errors. Blackbox testing on 42 test cases demonstrates that all system functionalities meet operational criteria. The main contribution of this research is providing accessible technology solutions for various business scales with affordable investment, addressing operational problems in transaction processes and product information management, as well as improving operational efficiency and service quality in retail environments.
Comparative Performance of IndoBERT and IndoLEM Baseline Models for Post-Disaster Health Information Extraction from Indonesian Online News Istiqomah, Nalar; Novika, Fanny
Journal of Computer Science and Informatics Engineering Vol 4 No 3 (2025): July
Publisher : Ali Institute of Research and Publication

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55537/cosie.v4i3.1174

Abstract

Natural disasters often have significant impacts on public health, yet systematic monitoring of post-disaster diseases in Indonesia remains limited. This study compares the performance of two Named Entity Recognition (NER) models in extracting health impacts, affected locations, and disaster types from Indonesian-language online news articles. The first model is IndoBERT, fine-tuned using 1,137 manually validated disaster-related news articles. The second comprises baseline models from the IndoLEM benchmark, namely mBERT and XLM-RoBERTa, without domain-specific training. Evaluation results show that IndoBERT outperforms the baseline models, achieving 90.00% accuracy and an F1-score of 88.26%, compared to mBERT (72.93%) and XLM-R (76.44%). Further analysis of the extracted entities reveals spatial and temporal disease trends: floods in Java are consistently associated with diarrhea and skin diseases, while volcanic eruptions in eastern Indonesia are linked to respiratory infections and hypertension. These findings highlight the importance of selecting appropriate models to support data-driven public health monitoring systems in disaster-prone regions
Design of Web-Based Project Management System with Multi-Level Role-Based Access Control Zikra, Andi Anzanul; Darnilasari, Aulia; Yanuary, Rahmat; Sari, Komala
Journal of Computer Science and Informatics Engineering Vol 4 No 3 (2025): July
Publisher : Ali Institute of Research and Publication

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55537/cosie.v4i3.1179

Abstract

Conventional project management using WhatsApp and email creates problems in progress tracking and inefficient team coordination. Research shows 70% of projects fail to deliver what was promised to customers. This research aims to design a web-based project management system using Laravel 12, Filament Shield, and Tailwind CSS. The research methodology employs waterfall SDLC with problem identification, data collection, UML system design, and black box testing stages. The system is designed with an Role-Based Access Control (RBAC) for 6 roles: Super Admin, Project Manager, Team Lead, Developer, QA Tester, and Client. Results demonstrate successful system implementation with integrated platform features, multi-level RBAC, real-time tracking through kanban dashboard, and responsive interface. Black box testing with 21 test cases covering 5 core system functions (authentication, RBAC, project CRUD operations, task management, and kanban board) achieved 100% success rate. The system successfully addresses progress tracking and coordination issues with an integrated solution supporting organizational scalability
Application of Monitoring Handy Talkie Usage as Support For Security Operation Based Android At Polda Gorontalo Harun, Rusni; Manurapon, Reynaldi; D. Paemo, Nuranissa; Ismail, Abdul Rahman
Journal of Computer Science and Informatics Engineering Vol 4 No 3 (2025): July
Publisher : Ali Institute of Research and Publication

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55537/cosie.v4i3.1183

Abstract

Information Technology is developing rapidly and it has a significant impact on various aspects of life, including the government sector. The process of data collection that used Microsoft Excel software and paper filing have drawbacks such as difficult data collection, data loss due to natural factors, and it was difficulty in determining the condition of HTs. The purpose of this reseacrh was to designed an application of innovative monitoring HTs for security based Android. This application utilizes Android technology and HT devices to improve the effectiveness and coordination of security within the Information and Communication Technology Division of Regional Police in Gorontalo. The methodology of research used was Research and Development (R&D). The application was created using Java programming using Android Studio, PHP, and HTML. It allows Polda personnel to log in, view HT types and stock, borrow and return devices, and upload transaction receipts. The database used was MySQL. The modeling used Unified Modeling Language. This system utilizes whitebox and blackbox testing. As a sample, researchers tested the HT Borrowing flowchart, obtaining a region (R) value of 3, an Independent Path value of 3, and a Cyclometric Complexity (CC) value of 3. Test scenarios, including login, menu, HT type, and history displays, were successful, demonstrating that the system correctly authenticates users, manages HT data accurately, displays complete information, manages returns and validation appropriately, and handles usage securely

Page 8 of 10 | Total Record : 96