cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
-
Contact Email
-
Phone
-
Journal Mail Official
-
Editorial Address
-
Location
Kota yogyakarta,
Daerah istimewa yogyakarta
INDONESIA
Jurnal Ilmiah Teknik Elektro Komputer dan Informatika (JITEKI)
Published by Universitas Ahmad Dahlan
ISSN : 23383070     EISSN : 23383062     DOI : -
Core Subject : Science, Engineering,
Computer Science & IT Electrical & Electronics Engineering
JITEKI (Jurnal Ilmiah Teknik Elektro Komputer dan Informatika) is a peer-reviewed, scientific journal published by Universitas Ahmad Dahlan (UAD) in collaboration with Institute of Advanced Engineering and Science (IAES). The aim of this journal scope is 1) Control and Automation, 2) Electrical (power), 3) Signal Processing, 4) Computing and Informatics, generally or on specific issues, etc.
Arjuna Subject : -
Articles 505 Documents
 37   38   39   40   41 
Design and Implementation of a Mobile Tourist Recommendation System for Sleman Using the Haversine Formula Faizy, Akmal; Sutopo, Joko
Jurnal Ilmiah Teknik Elektro Komputer dan Informatika Vol. 11 No. 2 (2025): June
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/jiteki.v11i2.30220

Abstract

Tourists often face challenges in accessing relevant information and navigating destinations due to the fragmented nature of existing mobile applications. This study addresses these issues by developing a mobile-based tourist recommendation system tailored to Sleman Regency, Yogyakarta. The system integrates features such as destination recommendations using the Haversine formula for proximity calculations, ticket purchasing, real-time navigation, a curated list of local culinary options, and a bug-reporting mechanism. Developed using the Agile methodology, the system underwent iterative enhancements based on user feedback. User preferences, including travel categories (e.g., nature, culture, family, temple) and culinary interests, are dynamically analyzed to provide personalized recommendations. The culinary feature highlights traditional dishes and popular dining spots, promoting local gastronomy. Black-box testing verified the system’s functionality, achieving a 100% success rate, while feedback mechanisms were incorporated to enable continuous improvement. The system ensures data privacy and security through robust encryption and regulatory compliance. It is scalable and adaptable, with potential applications beyond Sleman Regency. Furthermore, the system promotes sustainable tourism by encouraging eco-friendly destinations, respecting local cultural values, and supporting local culinary businesses. A comparative analysis highlights its advantage over conventional multi-application solutions by consolidating key features into a unified platform. This integrated solution streamlines tourist planning, navigation, and culinary exploration, enhancing convenience, user engagement, and satisfaction while contributing to the growth of Sleman’s tourism sector.
Handwritten Digits Detection Using Convolutional Neural Network Effendi , Doni Oktavian Ibnu; Saidah, Sofia; Putri , Yusnita
Jurnal Ilmiah Teknik Elektro Komputer dan Informatika Vol. 11 No. 2 (2025): June
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/jiteki.v11i2.30238

Abstract

Numbers are a collection of many lines and curves and play a vital role in everyday life. Each person has unique characteristics in handwriting, making handwritten digit detection a challenging task. This paper presents an approach for detecting handwritten digits using deep learning algorithms, particularly the Convolutional Neural Network (CNN)-based YOLOv8 family models. The main objective is to compare various YOLOv8 variants (YOLOv8n, YOLOv8s, YOLOv8m, YOLOv8l, and YOLOv8x) and determine the most optimal one in detecting handwritten digits.  Experimental results show that the YOLOv8x variant achieves the highest performance, with a mean Average Precision (mAP) of 96.9%, a recall of 100%, a precision of 99.8%, and an F1-score of 99.9%. The research contributions are achieving high accuracy in handwritten digit detection using the YOLOv8x model and utilizing a custom primary dataset of 3,000 handwritten digits for training and evaluation, which adds novelty and real-world relevance to the study.
Impact of Image Quality Enhancement Using Homomorphic Filtering on the Performance of Deep Learning-Based Facial Emotion Recognition Systems Bahri, Al; Oktiana, Maulisa; Fitria, Maya; Zulfikar
Jurnal Ilmiah Teknik Elektro Komputer dan Informatika Vol. 11 No. 2 (2025): June
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/jiteki.v11i2.30409

Abstract

Facial emotion recognition technology is crucial in understanding human expressions from images or videos by analyzing distinct facial features.  A common challenge in this technology is accurately detecting a person's facial expression, which can be hindered by unclear facial lines, often due to poor lighting conditions. To address these challenges, it is essential to improve image quality. This study investigates how enhancing image quality through homomorphic filtering and sharpening techniques can boost the accuracy and performance of deep learning-based facial emotion recognition systems. Improved image quality allows the classification model to focus on relevant expression features better.  Therefore, this research contributes to in facilitating more intuitive and responsive communications by enabling system to interpret and respond to human emotions effectively. The testing used three different architectures: MobileNet, InceptionV3, and DenseNet121. Evaluasi kinerja dilakukan menggunakan metrik akurasi, presisi, recall, dan F1-score. Experimental results indicated that image enhancement positively impacts the accuracy of the facial emotion recognition system. Specifically, the average accuracy increased by 1-2% for the MobileNet architecture, by 5-7% for InceptionV3, and by 1-3% for DenseNet121.
Impact of Cosine Similarity Function on SVM Algorithm for Public Opinion Mining About National Sports Week 2024 on X Mansyur, Abil; Karo Karo, Ichwanul Muslim; Firdaus, Muliawan; Simamora, Elmanani; Darari, Muhammad Badzlan; Habibi, Rizki; Panggabean, Suvriadi
Jurnal Ilmiah Teknik Elektro Komputer dan Informatika Vol. 11 No. 2 (2025): June
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/jiteki.v11i2.30605

Abstract

Public opinion on PON 2024 (National Sports Week in Indonesia) became a trending topic on X (formerly Twitter), reflecting both positive and negative sentiments. Understanding these sentiments is important for evaluating the event and preparing for the upcoming. However, baseline SVM algorithms using standard kernel functions are not optimized for text similarity and limit performance in sentiment analysis. This research proposes cosine similarity as a substitution for the kernel function on SVM, enhancing the sentiment analyzer's performance on public opinions about PON 2024. The approach leverages cosine similarity's strength in handling text-based data. The key contribution of this research is the integration of cosine similarity into the SVM algorithm as a replacement for kernel functions, improving performance in sentiment analysis. Additionally, this study offers a comprehensive comparison with baseline SVM and provides actionable insights for upcoming PON. The study collected 1,011 tweets related to PON 2024 using web scraping and the Twitter API, followed by labeling sentiments as positive, neutral, or negative. Several preprocessing techniques also were applied to prepare the data. Two models were developed: baseline SVM and another using SVM integrated with cosine similarity, both evaluated through accuracy, precision, recall, and F1-score. The baseline SVM achieved 85.1% accuracy, 85% precision, 83% recall, and 83.3% F1-score, struggling particularly with negative sentiment. Opposite, by integrating cosine similarity on SVM, the performance improved to 88.73% accuracy, 88.3% precision, 89.3% recall, and 88.3% F1-score—a boost of 3.3-6.3%. Additionally, the public opinion revealed that positive sentiments mostly focused on athlete achievements and medal awards, while negative sentiments highlighted issues like referee performance and specific sports (e.g., football). This approach can serve as a valuable tool for event organizers to identify public concerns and maintain positive aspects for the upcoming PON 2028.
A Machine Learning-Based Approach for Retail Demand Forecasting: The Impact of Spending Score and Algorithm Optimization Adriani, Ni Putu Erica Puspita; Huizen, Roy Rudolf; Hermawan, Dadang
Jurnal Ilmiah Teknik Elektro Komputer dan Informatika Vol. 11 No. 2 (2025): June
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/jiteki.v11i2.30630

Abstract

Demand forecasting in the retail industry remains a critical challenge, with inaccurate predictions leading to substantial inventory inefficiencies, financial losses, and reduced customer satisfaction. Traditional forecasting methods, primarily reliant on historical sales data, often lack the capacity to effectively model the complexities of dynamic consumer behavior and rapid market fluctuations. To address this, this study proposes a refined demand forecasting approach through the introduction of the Spending Score, a novel synthetic feature that synthesizes customer purchase frequency and total spending to augment predictive accuracy. We implement and optimize machine learning algorithms, specifically Random Forest, Decision Tree, and Support Vector Machine (SVM), using rigorous hyperparameter tuning techniques to determine the most effective model for retail demand prediction. Utilizing detailed customer transaction data, this research aims to identify key purchasing patterns that significantly influence demand variability. By integrating the Spending Score into our predictive models, we provide a data-driven framework enabling retailers to optimize inventory management, enhance targeted marketing strategies, and minimize operational inefficiencies. Empirical results demonstrate that the inclusion of the Spending Score leads to more stable and accurate demand forecasts, facilitating improved alignment between supply and market demand. While acknowledging potential limitations, including data scalability issues and the risk of feature-induced bias, future research will explore the integration of real-time data streams, advanced deep learning methodologies, and expanded datasets to further improve predictive capabilities and model adaptability in the continuously evolving retail landscape.
Performance Analysis of Prediction Methods on Tokyo Airbnb Data: A Comparative Study of Hyperparameter-Tuned XGBoost, ARIMA, and LSTM Nurfalah, Rizal Farhan Nabila; Hostiadi, Dandy Pramana; Triandini, Evi
Jurnal Ilmiah Teknik Elektro Komputer dan Informatika Vol. 11 No. 2 (2025): June
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/jiteki.v11i2.30631

Abstract

The rapid growth of the digital economy has increased the importance of accurately predicting Airbnb property occupancy rates, especially in dynamic and competitive markets such as Tokyo, Japan. Property owners face significant challenges in forecasting occupancy rates due to seasonal patterns, non-linear trends, and complex temporal dependencies within the data. Addressing these challenges, this study investigates the performance of ARIMA, XGBoost, and LSTM models in predicting Airbnb occupancy rates in Tokyo. The dataset is collected from Airbnb listings and includes relevant features such as location, price, customer reviews, and historical occupancy rates. The models were optimized using Grid Search for ARIMA and Random Search for XGBoost and LSTM to identify the best hyperparameter configurations. Evaluation metrics included Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Coefficient of Determination (R²), which are more appropriate for regression tasks. The results indicate that XGBoost achieves the highest R² (0.23), followed by LSTM (0.19) and ARIMA (0.03). However, the low R² values suggest that the models struggle to capture occupancy rate variations, indicating the potential influence of unmodeled external factors such as seasonality and policy changes. This study highlights the importance of hyperparameter tuning in improving prediction accuracy and contributes by providing an in-depth comparison of regression-based models for Airbnb occupancy forecasting.
A Hybrid Classification Model Based on BERT for Multi-Class Sentiment Analysis on Twitter Uyun, Shofwatul; Rosalin, Rizqi Praimadi; Sari, Luky Vianika; Sucinta, Hanny Handayani
Jurnal Ilmiah Teknik Elektro Komputer dan Informatika Vol. 11 No. 2 (2025): June
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/jiteki.v11i2.30665

Abstract

Social media is one of the media to convey opinions and sentiments. Sentiment analysis is an important tool for researchers and business people to understand user emotions efficiently and accurately. Choosing the right classification model has a significant impact on sentiment classification performance. However, the diversity of model architectures and training techniques poses its own challenges. In addition, relying on a single classification model often causes noise, bias, data imbalance, and limitations in handling data variations effectively. This study proposes a hybrid classification model where BERT is the baseline. Furthermore, BERT will be hybridized using LSTM, and BERT is hybridized with CNN to improve sentiment analysis on Twitter social media data. The hybrid approach aims to reduce the limitations of a single model classifier by increasing model effectiveness, reducing bias, and optimizing the model on imbalanced data. The following are the steps in this study, data preprocessing, data balancing, tokenization, model training, and performance evaluation. Three models were trained: the baseline BERT model, the BERT-CNN hybrid, and the BERT-LSTM hybrid. Model performance was assessed using accuracy, precision, recall, and F1 score. Experimental results show that the baseline BERT model achieves an accuracy of 91.45%, while BERT-LSTM achieves 91.60%, and BERT-CNN achieves the highest accuracy of 91.80%. However, further analysis is needed to determine whether these improvements are statistically significant and whether the hybrid model offers additional benefits beyond accuracy, such as remembering underrepresented sentiment categories.
Optimizing Machine Learning-Based Network Intrusion Detection System with Oversampling, Feature Selection and Extraction Shiddiq, Rama Wijaya; Karna, Nyoman; Irawati, Indrarini Dyah
Jurnal Ilmiah Teknik Elektro Komputer dan Informatika Vol. 11 No. 2 (2025): June
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/jiteki.v11i2.30675

Abstract

Network security is a global challenge that requires intelligent and efficient solutions. Machine Learning (ML)-based Network Intrusion Detection Systems (NIDS) have been proven to enhance accuracy in detecting cyberattacks. However, the main challenges in implementing ML-based IDS are dataset imbalance and large dataset size. This research addresses these challenges by applying oversampling techniques to balance the dataset, feature selection using random forest to identify the most relevant features, and feature extraction using Principal Component Analysis (PCA) to further reduce the selected important features. Additionally, K-fold cross-validation is used to test the features to minimize bias and ensure the model does not suffer from overfitting, while Optuna is implemented to automatically optimize model parameters for maximum accuracy. Since IDS performance deteriorates with high-dimensional features, the combination of methods used is evaluated based on feature selection applied to the model using datasets wtih 45 features selected from UNSW-NB15, 78 features from CIC-IDS-2017, and 80 features from CIC-IDS-2018 using various ML algorithms. The results demonstrate that the combination technique with feature selection, along with maximum optimization for each model significantly improves performance on large and imbalanced datasets reaching 99% accuracy compared to conventional methods in network traffic analysis.
Depression Detection on Social Media X Using Hybrid Deep Learning CNN-BiGRU with Attention Mechanism and FastText Feature Expansion Widiarta, I Wayan Abi; Setiawan, Erwin Budi
Jurnal Ilmiah Teknik Elektro Komputer dan Informatika Vol. 11 No. 2 (2025): June
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/jiteki.v11i2.30687

Abstract

Depression is a global mental health disorder affecting over 280 million people, with significant challenges in identifying sufferers due to societal stigma. In Indonesia, the National Adolescent Mental Health Survey in 2022 revealed that 17.95 million adolescents experience mental health disorders, with a portion of them suffering from depression. Social media platform X offers an alternative for individuals to share their mental health status anonymously, bypassing societal stigma. This study proposes a hybrid deep learning model combining CNN and BiGRU with an attention mechanism, TF-IDF for feature extraction, and FastText for feature expansion to detect depression in Indonesian tweets. The dataset comprises 50,523 Indonesian tweets, supplemented by a similarity corpus of 151,117 data. To optimize model performance, five experimental scenarios were conducted, focusing on split ratios, n-gram configurations, maximum features, feature expansion, and attention mechanisms. The main contribution of this research is the novel integration of FastText for feature expansion and the attention mechanism within a CNN-BiGRU hybrid model for depression detection. The results demonstrate the effectiveness of this combination, with the BiGRU-ATT-CNN-ATT model achieving an accuracy of 84.40%. However, challenges such as handling noisy, ambiguous social media data and addressing out-of-vocabulary words remain. Future research should explore additional feature expansion techniques, optimization algorithms, and approaches to handle noisy data, improving model robustness for real-world applications in mental health detection.
CFO-RetinaNet: Convolutional Feature Optimization for Oil Palm Ripeness Assessment in Precision Agriculture Alfariz, Muhammad Alkam; Santoso, Hadi
Jurnal Ilmiah Teknik Elektro Komputer dan Informatika Vol. 11 No. 2 (2025): June
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/jiteki.v11i2.30753

Abstract

Accurate ripeness assessment of oil palm fruit bunches (FFB) is critical for optimizing yield and quality in the palm oil industry, yet manual grading remains subjective and labor-intensive. This study proposes CFO-RetinaNet, an enhanced RetinaNet framework integrating deformable convolutions and hybrid attention mechanisms to optimize multi-scale convolutional features for robust ripeness classification under variable field conditions. Our key contribution is threefold: (1) a novel dataset of 4,728 high-resolution, expert-annotated FFB images spanning five ripeness stages (Immature to Decayed), collected under diverse lighting and occlusion scenarios in Central Kalimantan, Indonesia; (2) a feature optimization pipeline combining adaptive feature fusion and dynamic focal loss to improve discriminative capability for nuanced inter-class distinctions; and (3) a scalable deep learning solution validated through rigorous field testing. The model achieves a mean average precision (mAP) of 83.6% and an F1-score of 98.3%, outperforming YOLOv5 (82.5% mAP) and Faster R-CNN (76.4% mAP), with 18.5% fewer misclassifications than standard RetinaNet. It retains 99% accuracy in low-light conditions and reduces labor costs by automating error-prone grading tasks. By publicly releasing the dataset and framework, this work advances precision agriculture standards, offering a transferable solution for ordinal maturity classification in perennial crops while supporting sustainable palm oil production through optimized harvesting decisions.

Page 39 of 51 | Total Record : 505

 37   38   39   40   41 

Filter by Year

2015 2025


Reset
Filter By Issues
All Issue Vol. 11 No. 4 (2025): December Vol. 11 No. 3 (2025): September Vol. 11 No. 2 (2025): June Vol. 11 No. 1 (2025): March Vol. 10 No. 4 (2024): December Vol. 10 No. 3 (2024): September Vol. 10 No. 2 (2024): June Vol. 10 No. 1 (2024): March Vol. 9 No. 4 (2023): December Vol 9, No 4 (2023): December Vol 9, No 3 (2023): September Vol. 9 No. 3 (2023): September Vol 9, No 2 (2023): June Vol. 9 No. 2 (2023): June Vol. 9 No. 1 (2023): March Vol 9, No 1 (2023): March Vol 8, No 4 (2022): Desember Vol 8, No 3 (2022): September Vol 8, No 2 (2022): June Vol 8, No 1 (2022): March Vol 7, No 3 (2021): December Vol 7, No 2 (2021): August Vol. 7 No. 1 (2021): April Vol 7, No 1 (2021): April Vol 6, No 2 (2020): December Vol 6, No 1 (2020): June Vol. 6 No. 1 (2020): June Vol 5, No 2 (2019): December Vol 5, No 1 (2019): June Vol 4, No 2 (2018) Vol 4, No 2 (2018) Vol 4, No 1 (2018) Vol 4, No 1 (2018) Vol 3, No 2 (2017) Vol 3, No 2 (2017) Vol 3, No 1 (2017) Vol 3, No 1 (2017) Vol 2, No 2 (2016) Vol 2, No 2 (2016) Vol 2, No 1 (2016) Vol 2, No 1 (2016) Vol 1, No 1 (2015) Vol 1, No 1 (2015) More Issue