cover
Contact Name
Eko Fajar Cahyadi
Contact Email
ekofajarcahyadi@ittelkom-pwt.ac.id
Phone
+6285384848666
Journal Mail Official
infotel@ittelkom-pwt.ac.id
Editorial Address
Lembaga Penelitian dan Pengabdian Masyarakat (LPPM) Institut Teknologi Telkom Purwokerto Jl. D. I. Panjaitan, No. 128, Purwokerto 53147, Indonesia
Location
Kota bandung,
Jawa barat
INDONESIA
Jurnal INFOTEL
Published by Universitas Telkom
ISSN : 20853688     EISSN : 24600997     DOI : https://doi.org/10.20895/infotel.v15i2
Jurnal INFOTEL is a scientific journal published by Lembaga Penelitian dan Pengabdian Masyarakat (LPPM) of Institut Teknologi Telkom Purwokerto, Indonesia. Jurnal INFOTEL covers the field of informatics, telecommunication, and electronics. First published in 2009 for a printed version and published online in 2012. The aims of Jurnal INFOTEL are to disseminate research results and to improve the productivity of scientific publications. Jurnal INFOTEL is published quarterly in February, May, August, and November. Starting in 2018, Jurnal INFOTEL uses English as the primary language.
Articles 473 Documents
Metode Migrasi Lebah Madu Ratu untuk Meningkatkan Deteksi Fibrilasi Atrium dari Sinyal Detak Jantung Hafiizh, Muhammad; Aripriharta, Aripriharta; Elbaith Zaeni, Ilham Ari
JURNAL INFOTEL Vol 17 No 2 (2025): May
Publisher : LPPM INSTITUT TEKNOLOGI TELKOM PURWOKERTO

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20895/infotel.v17i2.1362

Abstract

Atrial Fibrillation (AF) is a common cardiac arrhythmia characterized by rapid and irregular electrical activity of the atrium. AF significantly increases the risk of ischemic stroke and mortality. With the increasing prevalence of cardiovascular risk factors, early detection of AF is crucial for effective intervention. Traditional electrocardiogram (ECG)-based detection methods face limitations, especially in asymptomatic patients or those with sporadic episodes of AF. This paper proposes a novel approach using the Queen Honey Bee Migration (QHBM) algorithm to detect AF from heartbeat signals. The dataset comprises both normal and AF heartbeat signals. The data undergoes preprocessing steps, including noise reduction and feature extraction. The system then classifies the signals using the QHBM algorithm. Key features such as heart rate variability (HRV), amplitude, and RR intervals are extracted for analysis. The QHBM algorithm achieved an accuracy of 95.2%, with a precision of 96.1%, a recall of 94%, and an F1 score of 95%. It outperformed traditional classifiers such as Random Forest, Support Vector Machine (SVM), and Naive Bayes across all performance metrics. In addition, QHBM demonstrated a superior ability to distinguish between normal sinus rhythm and AF, showing a significant improvement over the conventional method. Although the results are promising, challenges remain, including data imbalance and false positive and negative classifications. Oversampling techniques and further optimization of feature selection can enhance model performance. The QHBM algorithm presents a highly effective solution for automatic and real-time AF detection, offering a promising alternative to improve cardiac health monitoring systems.
TelUP Human Fall Dataset: A Motion Forecasting Study of Human Falls Widiyanto, Agung; Candraningtyas, Raphon Galuh; F.F, Andi Hisyam Helmi; Prameswari, Mayesq; Bashiran, Himam; Surahmat, Geugeut Nyarikawanti; Rahmah, Balqis Awaluna; Manika Dewi, Anak Agung Istri Candra; Yunus, Andi Prademon
JURNAL INFOTEL Vol 17 No 3 (2025): August
Publisher : LPPM INSTITUT TEKNOLOGI TELKOM PURWOKERTO

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20895/infotel.v17i3.1420

Abstract

This study investigates multitask learning approaches for human motion forecasting and fall classification using pose data extracted from video sequences. A custom dataset, the TelUP HumanFall Forecasting Dataset, was developed, containing annotated video frames representing fall and non-fall scenarios captured from six participants. Pose information was extracted using YOLOv11, producing 17 keypoints per frame, which were normalized and segmented into temporal sequences for training. Three deep learning architectures, Multilayer Perceptron (MLP), Recurrent Neural Network (RNN), and Long Short-Term Memory (LSTM), were implemented and evaluated. The models were assessed in a subject-independent test set consisting of two participants to ensure generalization. Quantitative evaluation measured the forecast error using the mean per joint position error (MPJPE) and classification accuracy. The MLP achieved the lowest MPJPE of 0.2630 (131.5 pixels), while the LSTM obtained the highest classification accuracy of 92.89%. Qualitative analysis revealed limitations in the capture of complex joint dynamics. Despite fast training convergence, the results emphasize a trade-off between forecast precision and classification accuracy. Future work will explore more expressive architectures and improved pose extraction methods to enhance forecast realism.
Penghitung Objek Karung di Konveyor Sabuk Berbasis Segmentasi dengan Teknik Thresholding Chaidir, Ali Rizal; Intyanto, Gramandha Wega; Setiabudi, Dodi; Wibowo, Dirgahayu Kusuma
JURNAL INFOTEL Vol 17 No 3 (2025): August
Publisher : LPPM INSTITUT TEKNOLOGI TELKOM PURWOKERTO

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20895/infotel.v17i3.1228

Abstract

Automation technology provides better outcomes from the perspective of time efficiency, material usage, and reducing error rates in a process. Sensors and visual sensing are components or methods frequently used in industrial automation systems. Visual sensing methods can replace simple tasks typically performed by an operator's vision in the industry, such as counting specific objects. Object counting algorithms are adapted to the type of object being counted; for example, counting fish objects and counting sack objects on a conveyor belt require different algorithms. Operators who are tired or unfocused can cause errors in counting objects like sacks on a conveyor belt, leading to financial losses. The main component used in this automatic counting system is a webcam. Each image frame is captured and processed in a computer to obtain parameters used as the basis for counting sack objects. The counting results are displayed on a monitor to facilitate the operator's view of the output. The method used is segmentation with a thresholding technique, which allows the separation of sack objects from the conveyor. The application of the segmentation method produces accurate counts; a total of 21 sack objects on the conveyor belt were counted without errors using this method. The use of filters did not affect the counting results, while the area size did. An area size of 50x50 provided the most accurate counting results and the best FPS (Frames Per Second) compared to other area sizes. This technique can ensure that the calculation process does not cause errors that result in losses.
Particle Size Detection of Palm Kernel Cake from Sieving Based on Images Using Convolutional Neural Network Irfansyah, Puput; Purwanto, Yohanes Aris; Wijaya, Sony Hartono; Nahrowi, Nahrowi
JURNAL INFOTEL Vol 17 No 3 (2025): August
Publisher : LPPM INSTITUT TEKNOLOGI TELKOM PURWOKERTO

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20895/infotel.v17i3.1246

Abstract

Palm kernel cake (PKC), a by-product of the palm oil industry, is widely used in animal feed due to its economic value. Its utilization reduces the reliance on costly conventional feed ingredients, reducing production expenses and improving livestock efficiency. However, contamination with palm kernel shells remains a key challenge, as it reduces quality and nutritional value. Identifying PKC particle sizes and addressing inconsistencies caused by contamination is complex, requiring advanced computational solutions. This study focuses on classifying the PKC particle sizes -fine, medium and coarse - using image processing combined with machine learning. A sieve shaker is applied to separate particles by size distribution, and a classification model is developed with Convolutional Neural Networks (CNN) under a transfer learning framework, which is effective for limited datasets. Six CNN architectures, MobileNet, Xception, InceptionV3, ResNet-152, VGG16, and NasNetMobile, are tested in four-layer configurations to identify the optimal setup. The results show that the proposed approach can classify PKC particle sizes with high accuracy. Among the models tested, MobileNet provides the best performance, achieving 0.99 accuracy and 0.98 F1 score in the second variation experiment. These findings present a practical and cost-effective method for assessing the quality of PKC, supporting scalable applications in feed production. This approach not only improves the accuracy of the evaluation, but also contributes to efficiency and sustainability in the livestock industry.
Digital Image Watermarking (DIW) yang kuat menggunakan SWT dan Watermark berbentuk data melingkar Vanda, Yoiceta; Dwiyanto, Dwiyanto; Nugroho, Agus
JURNAL INFOTEL Vol 17 No 3 (2025): August
Publisher : LPPM INSTITUT TEKNOLOGI TELKOM PURWOKERTO

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20895/infotel.v17i3.1284

Abstract

Digital images can be easily copied and distributed illegally using widely available software tools. A technique is needed that can be used to copyright protection. This technique can show the identity of the owner, so it can be used to track its authentication, namely the watermarking technique. Watermarking methods that embed side information into images with the aim of protecting copyright have been proposed. So far, these methods have been defeated by simple attacks such as rotation, translation and scaling. In this research, a novel digital image watermarking (DIW) system that takes advantage of the Wavelet Transform properties to embed a spread spectrum circular symmetric watermark in an image is proposed. Watermark is invisible, its mean imperceptible to the human eye but can be detected or extracted by software. Watermark robustness against translation, rotation, scaling, median filter, and adaptif filter attacks is achieved by the proposed method. Successful experiments showing the performance of the method to geometric transformation.
Bahasa Inggris Purnama, Dwi Adi
JURNAL INFOTEL Vol 17 No 3 (2025): August
Publisher : LPPM INSTITUT TEKNOLOGI TELKOM PURWOKERTO

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20895/infotel.v17i3.1289

Abstract

The industry is currently faced with rapid technological developments, including the challenges of industry 5.0. Therefore, it is necessary to develop advanced technology to improve automation and digitalization in the industrial sector. One of them is mining information from social media data, which produces large amounts of data storage (big data). Thus, there is potential to use social media data as a basis for policies to improve company performance. This study takes a case study of the telecommunications industry in Indonesia, using the Principal Component Analysis (PCA) and Principal Component Regression (PCR) methods. Big data is obtained from social media review data with a period of 33 weeks from unstructured data on telecommunications service products in Indonesia. The text mining stage produces 30 selected words for further analysis with PCA to produce the main components. Based on the evaluation results, the main components formed show a good correlation with the company's performance in the stock market based on five stock index indicators (price-open, high, low, close, and volume); at least there is one main component equation that shows a strong correlation. This shows the potential for using a data mining approach based on social media reviews as a basis for decision-making to improve company performance. Furthermore, the dominant variables formed from PCA are considered to obtain a simple mathematical model.
Data-Driven Product Segmentation for Shallot Commodities using PCA and K-Means Clustering Approach Winati, Famila Dwi; Arifin, Miftahol; Faturohman, Muhammad Iqbal; Mulyani, Enci
JURNAL INFOTEL Vol 17 No 3 (2025): August
Publisher : LPPM INSTITUT TEKNOLOGI TELKOM PURWOKERTO

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20895/infotel.v17i3.1307

Abstract

The shallot industry plays a strategic role in the Indonesian economy, especially in the Brebes Regency as the largest production center. However, challenges in the form of price fluctuations and low value-added products still burden farmers. Previous research tends to focus on individual products without considering a holistic product clustering strategy. This study aims to address the gap by applying the K-Means clustering method combined with Principal Component Analysis (PCA) to identify patterns in shallot and processed product sales data. The research data includes sales of 308 products from 2022-2024. The variables analyzed include product type, size, number of sold, and turnover. The results of the analysis formed three main groups, which are group 0 (small products with low performance), group 1 (large products with superior performance) and group 2 (medium products with stable performance). The findings indicate the importance of more targeted marketing strategies and product diversification. The implications of this study include optimizing superior products, revitalizing low-performing products, and developing stable products to expand the market. A customized e-Commerce-based strategy per cluster can improve the financial performance of the organization and the welfare of shallot farmers in a sustainable manner.
A Systematic Mapping Study On Multi-Algorithm Methods For Optimizing Transportation Systems Agustan, Agustan; Soeparyanto, Try Sugiyarto; Azikin, Thahir; Welendo, La; Mangidi, Uniadi; Isnawaty, Isnawaty
JURNAL INFOTEL Vol 17 No 3 (2025): August
Publisher : LPPM INSTITUT TEKNOLOGI TELKOM PURWOKERTO

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20895/infotel.v17i3.1328

Abstract

The integration of multi-algorithm methods has emerged as a transformative approach in addressing complex challenges within modern transportation systems. This study presents a systematic mapping review to explore the application, effectiveness, and potential advancements of multi-algorithm techniques across diverse transportation domains, including road, rail, air, and maritime transport. By synthesizing findings from 23 selected studies, this research identifies key algorithmic paradigms, such as machine learning (ML), genetic algorithms (GA), optimization models and hybrid frameworks, and their functional roles in enhancing decision making, resource allocation, and system efficiency. The analysis reveals that multi-algorithm systems offer significant advantages in managing uncertainty, processing large-scale datasets, and generating high-probability solutions for real-time operations. In particular, ML algorithms demonstrate robust capabilities in predictive maintenance and demand forecasting, while GA-based approaches excel in dynamic environments such as traffic signal optimization and UAV path planning. Despite these advances, critical challenges persist, including the need for high-quality data, scalable algorithm design, and seamless integration with existing infrastructure. Furthermore, certain promising methods such as the whale optimization algorithm (WOA) and graph neural networks (GNN) remain underutilized, highlighting opportunities for future exploration. This study underscores the necessity for interdisciplinary collaboration and methodological innovation to overcome deployment barriers and enhance the sustainability of intelligent transportation systems (ITS). Ultimately, multi-algorithm approaches have substantial potential to drive the evolution of transportation networks toward greater efficiency, resilience, and adaptability in an increasingly complex and dynamic mobility landscape.
A Novel Approach to Digital Image Security: Merging Cryptography and Steganography via LSB+TWO Technique Nugraha, Nur Budi; Santosa, Yaqutina Marjani
JURNAL INFOTEL Vol 17 No 3 (2025): August
Publisher : LPPM INSTITUT TEKNOLOGI TELKOM PURWOKERTO

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20895/infotel.v17i3.1338

Abstract

The increasingly rapid development of the digital era has resulted in information security becoming a crucial aspect in communication and data exchange. Digital images as a medium commonly used to store and transmit information are the main target in developing data security techniques. As the volume and sensitivity of information exchanged via digital images increases, the need for more sophisticated security mechanisms becomes increasingly pressing. This research combines cryptography and steganography using the LSB+TWO method. We integrate 256-bit AES in CBC mode for initial encryption, followed by improved LSB steganography techniques and a modified TWO algorithm. This method was tested on various datasets consisting of 1000 digital images with various resolutions. The results demonstrate significant improvements in the security and robustness of steganalysis compared to conventional methods, while maintaining high visual quality measured by PSNR and competitive embedding capacity. This research makes a significant contribution to the field of digital information security and paves the way for further development in image-based data protection techniques.
Semi-Supervised Sentiment Classification Using Self-Learning and Enhanced Co-Training Aribowo, Agus Sasmito; Khomsah, Siti; Saifullah, Shoffan
JURNAL INFOTEL Vol 17 No 3 (2025): August
Publisher : LPPM INSTITUT TEKNOLOGI TELKOM PURWOKERTO

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20895/infotel.v17i3.1344

Abstract

Sentiment classification is usually done manually by humans. Manual senti- ment labeling is ineffective. Therefore, automated labeling using machine learning is es- sential. Building a computerized labeling model presents challenges when labeled data is scarce, which can decrease model accuracy. This study proposes a semi-supervised learn- ing (SSL) framework for sentiment analysis with limited labeled data. The framework integrates self-learning and enhanced co-training. The co-training model combines three machine learning methods: Support Vector Machine (SVM), Random Forest (RF), and Lo- gistic Regression (LR). We use TF-IDF and FastText for feature extraction. The co-training model will generate pseudo-labels. Then, the pseudo-labels from models (SVM, RF, LR) are checked to choose the highest confidence — this is called self-learning. This framework is applied to English and Indonesian language datasets. We ran each dataset five times. The performance difference between the baseline model (without pseudo-labels) and SSL (with pseudo-labels) is not significant; the Wilcoxon Signed-Rank Test confirms it, obtaining a p- value < 0.05. Results show that SSL produces pseudo-labels on unlabeled data with quality close to the original labels on unlabeled data. Although the significance test performs well on four datasets, it has not yet surpassed the performance of the supervised classification (baseline). Labeling using SSL proves more efficient than manual labeling, as evidenced by the processing time of around 10-20 minutes to label thousands to tens of thousands of samples. In conclusion, self-learning in SSL with co-training can effectively label unla- beled data in multilingual and limited datasets, but it has not yet converged across various datasets.

Filter by Year

2009 2025


Filter By Issues
All Issue Vol 17 No 1 (2025): February 2025 Vol 17 No 3 (2025): August Vol 17 No 2 (2025): May Vol 16 No 4 (2024): November 2024 Vol 16 No 3 (2024): August 2024 Vol 16 No 2 (2024): May 2024 Vol 16 No 1 (2024): February 2024 Vol 15 No 4 (2023): November 2023 Vol 15 No 3 (2023): August 2023 Vol 15 No 2 (2023): May 2023 Vol 15 No 1 (2023): February 2023 Vol 14 No 4 (2022): November 2022 Vol 14 No 3 (2022): August 2022 Vol 14 No 2 (2022): May 2022 Vol 14 No 1 (2022): February 2022 Vol 13 No 4 (2021): November 2021 Vol 13 No 3 (2021): August 2021 Vol 13 No 2 (2021): May 2021 Vol 13 No 1 (2021): February 2021 Vol 12 No 4 (2020): November 2020 Vol 12 No 3 (2020): August 2020 Vol 12 No 2 (2020): May 2020 Vol 12 No 1 (2020): February 2020 Vol 11 No 4 (2019): November 2019 Vol 11 No 3 (2019): August 2019 Vol 11 No 2 (2019): May 2019 Vol 11 No 1 (2019): February 2019 Vol 10 No 4 (2018): November 2018 Vol 10 No 3 (2018): August 2018 Vol 10 No 2 (2018): May 2018 Vol 10 No 1 (2018): February 2018 Vol 9 No 4 (2017): November 2017 Vol 9 No 3 (2017): August 2017 Vol 9 No 2 (2017): May 2017 Vol 9 No 1 (2017): February 2017 Vol 8 No 2 (2016): November 2016 Vol 8 No 1 (2016): May 2016 Vol 7 No 2 (2015): November 2015 Vol 7 No 1 (2015): May 2015 Vol 6 No 2 (2014): November 2014 Vol 6 No 1 (2014): May 2014 Vol 5 No 2 (2013): November 2013 Vol 5 No 1 (2013): May 2013 Vol 4 No 2 (2012): November 2012 Vol 4 No 1 (2012): May 2012 Vol 3 No 2 (2011): November 2011 Vol 3 No 1 (2011): May 2011 Vol 2 No 2 (2010): November 2010 Vol 2 No 1 (2010): May 2010 Vol 1 No 2 (2009): November 2009 Vol 1 No 1 (2009): May 2009 More Issue