Article Per Year (5 Year)
p-Index From 2021 - 2026
2.392
P-Index
This Author published in this journals
All Journal Jurnal Pekommas Journal of Sustainable Engineering: Proceedings Series Jurnal Ilmiah Realtech EPI International Journal of Engineering Journal of Electrical Engineering and Computer (JEECOM)
Steven Pandelaki
Program Studi Teknik Informatika, Universitas Katolik De La Salle Manado
Chemical Engineering, Chemistry & Bioengineering Civil Engineering, Building, Construction & Architecture Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Electrical & Electronics Engineering Energy Engineering Industrial & Manufacturing Engineering Languange, Linguistic, Communication & Media Mechanical Engineering

Published : 16 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 16 Documents
Search

 1   2 
Prediksi Tingkat Kesejahteraan Masyarakat Di Kelurahan North Wangurer Menggunakan Regresi Linear Berganda Wikarsa, Liza; Pandelaki, Steven; Sumajouw, Karen
Jurnal Pekommas Vol 8 No 2 (2023): December 2023
Publisher : Sekolah Tinggi Multi Media “MMTC” Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56873/jpkm.v8i2.5094

Abstract

The level of social welfare in a country can determine the quality and condition of the country itself. North Wangurer sub-district is in Madidir District, Bitung City, with a population of around 750 households. The level of community welfare in this sub-district is solely based on the monthly income obtained from each community which is regarded to be ineffective. Thus, this research aimed to predict the level of community welfare at this sub-district using the multiple linear regression method. It was hoped that this research could provide insights for the North Wangurer sub-district office to make more effective policies/decisions to increase the level of welfare in hope of eradicating poverty and equitably distributing social assistance to the targetted households. There were four independent variables employed in this research such as income, education, occupation, and the number of family members. Meanwhile, the dependent variable was the level of community welfare consisting of Pra-KS, KS-I, KS-II, KS-III, and KS-III Plus as acknowledged by BKKBN. The results revealed that the level of community welfare for the North Wangurer sub-district was in Prosperous Family III Plus (Level 5). Most of the families (98,7%) in this sub-district can meet all basic needs, social psychology and its development, and self-accountability (self-esteem).
Implementasi Algoritma Boyer-Moore Majority Vote pada Sistem Manajemen Inventory Berbasis Internet of Things Lahama, Artquito Beltsazar; Sitanayah, Lanny; Pandelaki, Steven
Jurnal Pekommas Vol 9 No 1 (2024): Juni 2024
Publisher : Sekolah Tinggi Multi Media “MMTC” Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56873/jpkm.v9i1.5105

Abstract

Gold jewelry stores have to deal with various problems involving inventory control and inventory analytics in their day-to-day operations. Many Indonesian gold retailers tackle these problems by using a paperbased recording method. This method is not time efficient and adds more workload into the inventory management processes, such as stocktaking and popular item category classification. Stocktaking gets more difficult as more items are added to the inventory. These added workload and time increase the risk for items to go missing without being immediately documented. This paper proposes an Internet of Things solution that makes inventory stocktake more streamlined by introducing an RFID system with a web application that allows an operator to identify items during a stocktake instantly. In addition, store managers can also find trends in item categories using the Boyer-Moore majority vote algorithm. Our evaluation show that this system can effectively record data items in one scan with an average accuracy of 93.16% for a tray of 5 up to 25 items.
Pengembangan Aplikasi berbasis Android untuk Mengenali Jenis Lesi Kulit Menggunakan Convolutional Neural Network Sengkey, Blessynta Christesa; Pandelaki, Steven; Paseru, Debby
Jurnal Pekommas Vol 9 No 1 (2024): Juni 2024
Publisher : Sekolah Tinggi Multi Media “MMTC” Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56873/jpkm.v9i1.5557

Abstract

Skin lesions are skin abnormalities or disorders in the form of changes, damage, abnormal growth of the skin, such as changes in texture, color, appearance of lumps and spots on the skin. This disease certainly disrupts people's activities and behavior every day because of the reactions it causes, such as sensations of itching, pain, stinging and excessive heat. However, knowledge of the types of skin lesions by the lay public is still lacking and a system is needed that can provide information regarding primary skin lesions. Image processing as part of machine learning can recognize types of primary skin lesions through applications that use Convolutional Neural Network (CNN). This method can perform good feature extraction and classification, so it is very suitable for image detection. Research was carried out on 4 classes of lesions, namely macular, urticarial, popular and vesicular. Based on the test results with the CNN model, it was found that the average accuracy value was 95% with the calculation of values in the macular class with precision 91%, recall 100%, f-1 score 95%, urticaria class with precision 100%, recall 91%, f-1 score 95%, papule class with precision 98%, recall 93%, f-1 score 96% and vesicular class with precision 93%, recall 99%, f-1 score 96%.
Analysis of The Effectiveness of Covid-19 Prevention Measures in Manado City Using the K-means Method Wikarsa, Liza; Pandelaki, Steven; Kurnia, Henrico
Jurnal Pekommas Vol 7 No 1 (2022): June 2022
Publisher : Sekolah Tinggi Multi Media “MMTC” Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56873/jpkm.v7i1.4390

Abstract

Covid-19 is an infectious disease caused by the Coronavirus, spreading quickly and resulting in death. Thus, the government of Manado city recommends all the residents always comply with the   Covid-19 health protocols. The study aimed to investigate the public's response to the effectiveness of the Covid-19 prevention measures imposed by the local government. The prevention measures include using masks, healthy lifestyles, clean lifestyles, and social distancing. K-means clustering was used to determine the level of effectiveness of prevention measures into four clusters, namely very effective (C1), moderately effective (C2), ineffective (C3), and very ineffective (C4). This algorithm yields consistent results despite the difficulty of predicting the K-values or the number of clusters at the beginning of the algorithm. As a result, 58% of respondents consider the prevention measures very effective. They have a high level of Covid-19 awareness and the consequences of violating the health protocols, even though about 10% of the respondents doubted the health protocols in Manado were effective due to the high number of death cases daily. The results can be used as insights for the local government to stop the spread of the Coronavirus in Manado city.
An Internet of Things-Based Trash Can Monitoring System Felicia Kusuma; Lanny Sitanayah; Steven Pandelaki
EPI International Journal of Engineering Vol 9 No 1 (2026): Volume 9 Number 1, February 2026
Publisher : Center of Techonolgy (COT), Engineering Faculty, Hasanuddin University

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

Abstract

Everyday people throw trash in trash cans, both in private and public cans. Hence, trash cans become one of the objects needed by community. When a trash can is full or produces bad odor, it may disturb the surrounding community. When people want to throw trash but the can is full, they usually throw it everywhere. When trash produces bad odor, it can disturb sense of smell and may cause diseases to surrounding community. Naïve Bayes is one of the supervised data mining algorithms for classification. To predict a class, Naïve Bayes requires historical data to find the posterior of each class. The Internet of Things is a technology that involves a large number of processes and devices to improve the quality of life of people who use it and facilitate access to information and services provided. The trash can monitoring system is built using C and Python languages. C language is used to program the sensor device, while Python language is used to program the application. The database used in this system is MySQL. The system takes data from an MQ-4 gas sensor and an HC-SR04 ultrasonic sensor in the sensor device, which is installed in a trash can, to classify the condition of the trash can. Based on the results of accuracy performance evaluation, the accuracy with 60:40 data partition is 79%, 70:30 data partition is 78.9%, and 80:20 data partition is 79.1%.
Retinocare: A Web-Based Intelligent System for Early Detection of Diabetic Retinopathy Using CNN Angelia Melani Adrian; Steven Pandelaki; Gladys Ratuliu; Jonathan Kamagi
Journal of Electrical Engineering and Computer (JEECOM) Vol 8, No 1 (2026)
Publisher : Universitas Nurul Jadid

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33650/jeecom.v8i1.13568

Abstract

Diabetic retinopathy (DR) is a leading cause of preventable blindness worldwide and is becoming a significant public health concern in Indonesia due to the rising prevalence of diabetes. Early detection is critical, yet access to ophthalmologists and conventional fundus cameras remains limited in many primary healthcare facilities. To address these challenges, this study proposes a cost-effective, web-based intelligent system for early detection of DR using smartphone-based fundus adapters and deep learning.A hybrid dataset was employed, combining publicly available fundus image repositories with locally collected retinal images from Indonesian healthcare facilities, annotated by ophthalmologists. Images were preprocessed through normalization, cropping, artifact removal, and augmentation to address variability, particularly from smartphone acquisitions. A DenseNet-121 convolutional neural network was fine-tuned on this hybrid dataset to classify DR into five severity levels according to the International Clinical Diabetic Retinopathy Disease Severity Scale. Model performance was evaluated using accuracy as the primary metric, with results compared against ophthalmologist annotations.The proposed system demonstrated promising performance in classifying DR severity levels, showing that combining public and local datasets improves contextual relevance and model robustness. Furthermore, integration into a web-based platform enables healthcare workers in primary care to upload fundus images, obtain real-time classification results, and facilitate referral decisions for severe cases.This study contributes to the development of an accessible and scalable screening tool for DR in Indonesia by integrating affordable imaging hardware, locally relevant datasets, and an AI-powered classification system. The approach has the potential to reduce reliance on expensive equipment and specialists, supporting national efforts to prevent diabetes-related blindness.
Co-Authors Angelia Melani Adrian Anthoni Pulakian Dame, Kristian Debby Paseru Felicia Kusuma Gladys Ratuliu Hendri Hermawan Adinugraha Henrico Kurnia Jonathan Kamagi Kondo, Kristian Y. Kristian Dame Kristian Dame Kristian Kondo Kurnia, Henrico Lahama, Artquito Beltsazar Lanny Sitanayah Micael Liem Michael George Sumampouw Polly, Victori Sengkey, Blessynta Christesa Steven Koloay Sumajouw, Karen Verna Y.P. Bokau Victory Polly Victory Polly Wikarsa, Liza