cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
M.Pd Asni Tafrikhatin
Contact Email
asni@politeknik-kebumen.ac.id
Phone
+6285643500965
Journal Mail Official
asni@politeknik-kebumen.ac.id
Editorial Address
Letnan Jenderal Suprapto No.73, Kranggan, Bumirejo, Kec. Kebumen, Kabupaten Kebumen, Jawa Tengah 54311
Location
Kab. kebumen,
Jawa tengah
INDONESIA
Jurnal E-Komtek
Published by Politeknik Piksi Ganesha Indonesia
ISSN : 25803719     EISSN : 26223066     DOI : https://doi.org/10.37339/e-komtek.v4i2.269
Core Subject : Science, Engineering,
Computer Science & IT Electrical & Electronics Engineering
Jurnal E-Komtek (Elektro-Komputer-Teknik) is a Journal that contains scientific articles in the form of research results, analytical studies, application of theory, and discussion of various problems relating to Electrical, Computer, and Automotive Mechanical Engineering.
Arjuna Subject : Ilmu Komputer (semua kategori) - Visi Komputer dan Pengenalan Pola
Articles 276 Documents
 24   25   26   27   28 
Comparative Analysis of the Accuracy of Multiple Linear Regression Method and Ridge Regression Method in Predicting Dengue Fever Cases in South Tangerang City Dina Aulia; Herman Bedi Agtriadi; Luqman
Jurnal E-Komtek (Elektro-Komputer-Teknik) Vol 9 No 1 (2025)
Publisher : Politeknik Piksi Ganesha Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37339/e-komtek.v9i1.2292

Abstract

One of the main health issues in South Tangerang City is dengue fever (DBD). This study aims to compare the accuracy of Multiple Linear Regression and Ridge Regression methods in predicting the number of DBD cases using weather data such as temperature, humidity, and average rainfall. The data used is monthly data from South Tangerang City. The analysis process includes preprocessing, splitting the dataset into training and testing data, and applying both regression methods. To determine the prediction error rate, model accuracy is evaluated using the Mean Absolute Percentage Error (MAPE) metric. The results indicate that Ridge Regression performs better for datasets with high multicollinearity, yielding a MAPE value of 20.12%, while Multiple Linear Regression is more effective for datasets with low feature correlation, showing a MAPE value of 44.6%. This study provides important insights into selecting predictive techniques based on the characteristics of the analyzed dataset. It is hoped that this research can improve mitigation and planning for DHF cases in South Tangerang City by choosing the appropriate approach.
Analysis of Long Short-Term Memory (LSTM) and Extreme Gradient Boosting (XGBoost) Algorithms to Predict the Number of Airplane Passengers at Makassar Sultan Hasanuddin International Airport : Systematic Literature Review Ainul Idham; Efy Yosrita
Jurnal E-Komtek (Elektro-Komputer-Teknik) Vol 9 No 1 (2025)
Publisher : Politeknik Piksi Ganesha Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37339/e-komtek.v9i1.2298

Abstract

This study compares the performance of Long Short-Term Memory (LSTM), Extreme Gradient Boosting (XGBoost), and hybrid techniques to forecast the number of aircraft passengers. This analysis was carried out utilizing the Systematic Literature Review (SLR) method and the PRISMA approach. Only 11 of the 44,564 items filtered during the initial round met the inclusion requirements. The LSTM model performed well in capturing time series patterns, however XGBoost was more robust when employed on data with noise and outliers. The hybrid model (LSTM + XGBoost) performed the best, with an average accuracy of 96%, RMSE of 0.015, and MAPE of 2.45%. This demonstrates that the hybrid technique is quite good in predicting the number of airplane passengers, particularly for complicated, dynamic, and seasonal time series data. These findings are recommended for the development of machine learning-based prediction systems for airports.
Literature Study: Prediction of the Type of Company where Students Work Using Naïve Bayes and Neural Network Algorithms Saputra, Angga; Luqman; Herman Bedi Agtriadi
Jurnal E-Komtek (Elektro-Komputer-Teknik) Vol 9 No 1 (2025)
Publisher : Politeknik Piksi Ganesha Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37339/e-komtek.v9i1.2314

Abstract

Research was conducted to evaluate the effectiveness of various machine learning algorithms, such as Naive Bayes, Support Vector Machine, Random Forest, and Artificial Neural Network (ANN), in predicting and classifying data. Naive Bayes proved to be efficient and accurate in structured data classification, such as predicting alumni's waiting time to get a job (94%) and vocational school students' job readiness (96.95%). On the other hand, neural network methods such as ANN and GRNN are superior in handling non-linear regression problems, such as house price prediction or college students' study period, although there is still room to improve accuracy. Random Forest is more suitable for complex data, while Naive Bayes is more effective for simple data. This research emphasizes the importance of selecting relevant variables, such as gender, major, and GPA, to improve model performance. Therefore, the selection of machine learning methods should be tailored to the type of data and the purpose of the analysis, as each algorithm has its own advantages and disadvantages.
Design and Implementation of a QR Code-Based Attendance Application at SMA Negeri 1 Cangkringan Muhammad Hilmiawan Sulthoni; Vikky Aprelia Windarni; Surya Tri Atmaja; Dewi Anisa Istiqomah; Fiyas Mahananing Puri
Jurnal E-Komtek (Elektro-Komputer-Teknik) Vol 9 No 1 (2025)
Publisher : Politeknik Piksi Ganesha Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37339/e-komtek.v9i1.2392

Abstract

The manual attendance system at SMA Negeri 1 Cangkringan still faces various obstacles, such as the time-consuming recording process, the risk of data loss, and the possibility of manipulating student attendance. Overcome these problems, a web-based student attendance application was developed using the Waterfall method. The method used in completing this system is the waterfall method, which consists of needs analysis, system planning, development, and testing. The system is designed using web technology, which allows better accessibility for teachers and administrative staff in recording attendance in real-time. The result of the user-friendly application design, allows teachers to record student attendance quickly and accurately. The system is also a report feature that can be accessed easily, making it easier to make decisions regarding student attendance. It is expected that the administration process at SMA N 1 Cangkringan can be more efficient, transparent, and reliable.
Optimization of Double Exponential Smoothing Model for Daily Earth Temperature Forecasting in Dayeuhluhur, Cilacap Ridzna Asep Purwanto; Hakim, Dimara Kusuma; Supriyono; Harjono
Jurnal E-Komtek (Elektro-Komputer-Teknik) Vol 9 No 1 (2025)
Publisher : Politeknik Piksi Ganesha Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37339/e-komtek.v9i1.2396

Abstract

Global warming has caused an increase in the Earth's surface temperature, which has a significant impact on the environment and human life. This study aims to predict the daily surface temperature in Dayeuhluhur District, Cilacap, for the next one year using the Double Exponential Smoothing (DES) method. The data used comes from the NASA POWER platform with a time span of 2015 to 2025, including three main variables: earth surface temperature (TS), solar radiation (ALLSKY_SFC_SW_DWN), and maximum 10-meter wind speed (WS10M_MAX). Preprocessing was done by removing February 29 in leap years and applying annual differencing (lag 365) to stabilize the seasonal pattern. Smoothing parameters α and β were optimized based on Mean Absolute Percentage Error (MAPE) values. Results show a moderate and consistent increasing trend in temperature, with the best accuracy in the temperature variable (MAPE 2.41%), followed by solar radiation (21.56%) and wind speed (30.18%). This method proves effective in forecasting temperature with clear seasonal patterns and contributes to supporting data-driven climate change mitigation policies.
Clustering of Earthquakes on The Island of Java Using K-Means Algorithm Based on Magnitude and Depth Viki Flendiansyah; Hakim, Dimara Kusuma; Feri Wibowo; Agung Purwo Wicaksono
Jurnal E-Komtek (Elektro-Komputer-Teknik) Vol 9 No 1 (2025)
Publisher : Politeknik Piksi Ganesha Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37339/e-komtek.v9i1.2397

Abstract

Indonesia is one of the countries with a high level of earthquake vulnerability because it is located in the Pacific Ring of Fire. Java Island, as the most populous region and the center of the national economy, has a great risk of earthquake impacts. This study aims to analyze earthquakes in Java Island during the 2019-2024 period using the K-Means algorithm. Clustering the data based on magnitude, depth, location, and time of occurrence resulted in three clusters that reflect the characteristics of earthquakes in the region. This clustering provides important insights into the distribution and intensity of earthquakes in Java. The information obtained can be used to support disaster mitigation efforts more strategically. The government and community are expected to be able to increase preparedness for disaster risks and design effective mitigation policies to minimize the impact of future earthquakes. This research shows the great potential of applying data-driven technology as a basis for decision-making in disaster mitigation in Indonesia.
A Herman Bedi Agtriadi; M Habibi; Zakiyah Misfazilah
Jurnal E-Komtek (Elektro-Komputer-Teknik) Vol 9 No 1 (2025)
Publisher : Politeknik Piksi Ganesha Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37339/e-komtek.v9i1.2501

Abstract

Breast cancer is the most common cancer globally with a malignant category that poses a serious and frightening threat to women. According to data from Globocan. In Indonesia alone in 2022 the number of new cases of breast cancer reached 66,271 cases, thus contributing (30,1.6%) of the total cancer cases in Indonesia. Of the cases with more than 22 thousand deaths, breast cancer is the second most deadly cancer. 70% of breast cancer cases are detected already at an advanced stage, where this case can occur due to delays in medical personnel who have not been able to detect breast cancer manually. This requires technology to help doctors and radiologists to evaluate Magnetic Resonance Imaging (MRI) images automatically. One of the deep learning methods useful for MRI image analysis is Convolutional Neural Network (CNN) using VGG19 and AlexNet architecture which has been proven in the classification process. This study uses data from Kaggle with a total of 1400 data. Through the use of the Convolutional Neural Network method, this study obtained a fairly optimal accuracy on the VGG19 architecture of 99% and on the AlexNet Architecture of 97%.
Implementation of YOLO V8 Algorithm in Organic and Anorganic Waste Detection Application for Waste to Energy Management Arvio, Yozika; Kusuma, Dine Tiara; Sangadji, Iriansyah BM
Jurnal E-Komtek (Elektro-Komputer-Teknik) Vol 9 No 1 (2025)
Publisher : Politeknik Piksi Ganesha Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37339/e-komtek.v9i1.2310

Abstract

Waste management in Indonesia is a major challenge, especially in the development of waste to energy (WTE). Accurate classification of organic and inorganic waste is required to optimise energy conversion. This research develops an automated waste detection system in temporary landfill sites (TPS) using the YOLOv8 algorithm, known for its high speed and accuracy. The research involved data collection, development of a YOLOv8-based computational model, and system construction and testing according to field requirements. The results show that YOLOv8 has high performance in detecting organic and inorganic waste, with 99.35% accuracy, 98.6% precision, 98.6% recall and 98.5% F1 score. This system can speed up the waste sorting process and has the potential to be used in domestic and public environments for the automatic detection of waste categories.
Application of Random Forest Classification Method in Determining the Best Quality Service in the Implementation of International Certification at ITCC ITPLN Hendra Jatnika; Luqman Luqman; Mochamad Farid Rifai; Nasya Miranda Umar
Jurnal E-Komtek (Elektro-Komputer-Teknik) Vol 9 No 1 (2025)
Publisher : Politeknik Piksi Ganesha Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37339/e-komtek.v9i1.2349

Abstract

Information Technology Certification Center or also known as ITCC is one of the work units owned by the Institute Technology of PLN which is a unit that organizes training and international certification. In order to improve the quality of service from the activities that have been organized by ITCC, the ITCC committee always prepares a link for participants to write feedback which will later become material for evaluation by ITCC. In this study, 2,720 data were used which were divided into 2 categories, namely 1,884 data with positive sentiment categories and 836 data with negative sentiment categories. The data is processed using the Random Forest method in order to find out the optimality of knowing the method. The final result obtained from the application of the Random Forest classification method is an accuracy percentage of 88.97% with a precision value of 0.92, recall of 0.91, and f1 score of 0.92
A Single Image Dehazing using U-Net and Lightweight Vision Transformer Aditya, Dion; Yosrita, Efy
Jurnal E-Komtek (Elektro-Komputer-Teknik) Vol 9 No 1 (2025)
Publisher : Politeknik Piksi Ganesha Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37339/e-komtek.v9i1.2352

Abstract

This research presents a single-image dehazing method that integrates a Lightweight Vision Transformer (LVT) and U-Net to capture both local and global features. LVT enhances resolution, U-Net extracts local features, and LVT refines global dependencies before fusion. Evaluations on O-Haze and HSTS datasets show PSNR scores of 27.88 (O-Haze, ResNet-50) and 28.22 (HSTS, no backbone), outperforming existing methods while maintaining competitive SSIM. The results demonstrate effectiveness in real-world haze scenarios, such as wildfire-induced haze in Indonesia.

Page 26 of 28 | Total Record : 276

 24   25   26   27   28 

Filter by Year

2017 2025


Reset
Filter By Issues
All Issue Vol 9 No 1 (2025) Vol 8 No 2 (2024) Vol 8 No 1 (2024) Vol 7 No 2 (2023) Vol 7 No 1 (2023) Vol 6 No 2 (2022) Vol 6 No 1 (2022) Vol 5 No 2 (2021) Vol 5 No 1 (2021) Vol 4 No 2 (2020) Vol 4 No 1 (2020) Vol 3 No 2 (2019) Vol 3 No 1 (2019) Vol 2 No 2 (2018) Vol 2 No 1 (2018) Vol 1 No 1 (2017) More Issue