IJCCS (Indonesian Journal of Computing and Cybernetics Systems)
Not a member yet
480 research outputs found
Sort by
Optimal Feature Selection in Diabetes Classification Using the MLP Algorithm
Full text linkIn 2021, approximately 531 million people worldwide were affected by diabetes, with 90% diagnosed as type 2. Diabetes often coexists as a comorbidity with other conditions such as kidney and heart disease. The research aims to employ machine learning for diabetes classification, with the Multilayer Perceptron (MLP) algorithm being a key component in the early detection process. The experiments utilized data from the UCI database of Sylhet hospitals, featuring 16 attributes and 2 classes indicating positive and negative diabetes cases. Performance testing using the MLP algorithm involved varying the number of neurons in the hidden layer. The research architecture is denoted as n:p:m, where n represents 16 neurons based on the attributes, m signifies 2 neurons based on the number of classes, and p undergoes variations. The machine learning tool employed in this research is Weka. Within the Weka tool, MLP offers types of hidden layer neuron configurations: 'a', 't', 'i', and 'o'. The test results, conducted with 520 training data and testing on the same dataset, yielded accuracies of 98.85%, 98.85%, 99.42%, and 98.46% for types 'a', 't', 'i', and 'o', respectively
Object Detection Based on You Look Only Once Version 8 for Real-Time Applications
Full text linkThis research focus to involves human detection in crowded situations, especially in the lecturer's room. The lecturer's room is very vulnerable because it can be accessed by anyone with only one entry and exit to the lecturer's room, so it would be perfect to place this Yolo camera in front of the lecturer's room so that incoming and outgoing activities can be monitored during work days on campus. The main challenge is how the system can distinguish individuals in dense crowds and identify their relative locations to each other. In this context, it is necessary to find a solution that can overcome the uncertainty of recognizing individuals in a group and accurately understand the location and distance between them. One proposed solution is to use the YOLO algorithm on video recordings to detect human objects in the lecturer's room during working hours. This research introduces the YOLOv8 model, a real-time detection system with high speed and accuracy in detecting and classifying objects in video recordings. YOLOv8 can accurately detect object movement, making it an efficient real-time framework for dealing with complex objects. This research experiment involved using eight different smartphone devices to collect datasets. Using various smartphone devices aims to test object detection performance under various shooting conditions, including variations in image quality, lighting, shooting angle, and camera resolution. The research results show that using multiple smartphone devices in dataset collection can improve the robustness and accuracy of object detection models. By integrating datasets from various sources and shooting conditions, the YOLOv8 model was successfully trained to better recognize objects in different situations, even in campus environments that often have challenges such as weather variations and lighting fluctuations. The test results show an accuracy rate of 93.33% in human object detectio
Optimizing Coral Fish Detection: Faster R-CNN, SSD MobileNet, YOLOv5 Comparison
Full text linkThis study underscores the critical role of accurate Chaetodontidae fish abundance observations, particularly in assessing coral reef health. By integrating deep learning algorithms (Faster R-CNN, SSD-MobileNet, and YOLOv5) into Autonomous Underwater Vehicles (AUVs), the research aims to expedite fish identification in aquatic environments. Evaluating the algorithms, YOLOv5 emerges with the highest accuracy, followed by Faster R-CNN and SSD-MobileNet. Despite this, SSD-MobileNet showcases superior computational speed with a mean average precision (mAP) of around 92.21% and a framerate of about 1.24 fps. Furthermore, employing the Coral USB Accelerator enhances computational speed on the Raspberry Pi 4, enabling real-time detection capabilities. This study incorporates centroid tracking, facilitating accurate counting by assigning unique IDs to identified objects per class. Ultimately, the real-time implementation of the system achieves 87.18% accuracy and 87.54% precision at 30 fps, empowering AUVs to conduct real-time fish detection and tracking, thereby significantly contributing to underwater research and conservation efforts
Electroencephalogram-Based Emotion Classification Using Machine Learning and Deep Learning Techniques
Full text link Electroencephalogram (EEG) records brain activity as electrical currents to discern emotions. As interest in human-computer emotional connections rises, reliable and implementable emotion recognition algorithms are essential. This study classifies EEG waves using machine and deep learning. A four-channel Muse EEG headband recorded neutral, negative, and positive emotions for the publicly available Feeling Emotions EEG dataset. Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), and Gated Recurrent Unit (GRU) were utilized for deep learning, while SVM, K-NN, and MLP were used for machine learning. The models were assessed for accuracy, precision, recall, and F1-Score. SVM, K-NN, and MLP have accuracy scores of 0.98, 0.95, and 0.97. Deep learning methods CNN, LSTM, and GRU had 0.98, 0.82, and 0.97 accuracy. SVM and CNN surpassed other approaches in accuracy, precision, recall, and F1-Score. The research shows that machine learning and deep learning can classify EEG signals to identify emotions. High accuracy results, especially from SVM and CNN, suggest these models could be used in emotion-aware human-computer interaction systems. This study adds to EEG-based emotion classification research by revealing model selection and parameter tweaking strategies for better categorization
Offensive Language and Hate Speech Detection using BERT Model
Full text linkHate speech detection is an important issue in sentiment analysis and natural language processing. This study aims to improve the effectiveness of hate speech detection in English text using the BERT model, along with modified preprocessing techniques to enhance the F1-score. The dataset, sourced from Kaggle, contains English text with hate speech content. Evaluation results show a significant improvement in the model's accuracy and overall text classification performance. The BERT model achieved 89.11% accuracy on test data, correctly predicting 85 out of 95 samples. While the model excels at classifying offensive text with around 95% accuracy, it struggles to distinguish between hate and offensive text, with some confusion between neither and offensive categories. The classification report shows F1-scores of 0.43 for the hate class, 0.94 for the offensive class, and 0.84 for the neither class, with a weighted average F1-score of 0.89 and a macro average of 0.73. These results indicate that the BERT model delivers solid performance in detecting hate speech, though there is room for improvement, particularly in distinguishing certain classes
HOSPITAL MANAGEMENT INFORMATION SYSTEM EVALUATION AT GRHA PERMATA IBU DEPOK
Full text linkThe GRHA Permata Ibu Hospital in Depok has been implementing the Hospital Management Information System (HMIS) since 2013 to support all hospital service processes. An evaluation of the HMIS is necessary to understand the actual state of the information system implementation. The objective is to examine and assess the HMIS at GRHA Permata Ibu Hospital to achieve results that are comparable using specific benchmarks. The goal is to obtain performance outcomes that support better, effective, and efficient services, and to identify the system's current condition for further action planning to improve its performance. The research follows a quantitative method with an online survey approach using Google Forms. The HOT-Fit evaluation model is used to assess the readiness level for utilizing an information system, focusing on the crucial components of Human, Organization, Technology, and Net Benefits. The study's results reveal that out of the 13 developed hypotheses, 6 hypotheses were accepted, while 7 hypotheses were rejected. Therefore, the research proves that not all proposed hypotheses are empirically supported. Based on the test results, several recommendations are provided to enhance the success rate of the HMIS implementation at GRHA Permata Ibu Hospital in Depok
DEVELOPMENTS AND TRENDS IN CYBERSECURITY AGAINST HUMAN FACTORS AND TIME PRESSURE USING BIBLIOMETRIC ANALYSIS
Full text linkMemahami keamanan siber sangat penting di era digital saat ini, dan penelitian telah dilakukan untuk memahami faktor-faktor yang mempengaruhi keberhasilan atau kegagalannya. Faktor manusia berperan penting dalam keamanan siber, dan lebih dari 95% serangan yang berhasil disebabkan oleh kesalahan manusia. Tekanan waktu adalah faktor lain yang tidak boleh diabaikan, karena organisasi sering kali menghadapi tekanan waktu yang tinggi dalam lingkungan bisnis yang kompetitif dan dinamis. Penelitian mengenai faktor manusia dalam keamanan siber menunjukkan bahwa faktor manusia masih menjadi perhatian utama dibandingkan dengan teknologi. Penelitian ini bertujuan untuk menganalisis perkembangan dan tren keamanan siber mengenai faktor manusia dan tekanan waktu dari tahun 2014 hingga 2023 menggunakan Analisis Bibliometrik dari software R studio. Metodologi penelitian meliputi perencanaan, identifikasi kata kunci, pencarian data Scopus, dan pembatasan pencarian pada "semua bidang" untuk memperoleh data yang sesuai dengan tema penelitian. Penelitian dibatasi sebanyak 110 jurnal yang diambil dari database Scopus. Kesimpulannya, memahami faktor manusia dan tekanan waktu dalam keamanan siber sangat penting bagi organisasi untuk meningkatkan langkah-langkah keamanan siber mereka. Dengan menganalisis perkembangan dan tren faktor-faktor ini, para peneliti dapat lebih memahami masa depan keamanan siber dan mengambil keputusan yang tepat untuk melindungi informasi dan infrastruktur penting.
Significant Wave Height Forecasting using Long-Short Term Memory (LSTM) in Seribu Island Waters
Full text linkWind waves are natural phenomena primarily generated by the wind. Information about wave height and period is highly crucial in various marine fields such as coastal engineering, fisheries, and maritime transportation. However, accurately predicting wave height remains a challenge due to the stochastic nature of ocean waves themselves. Several approaches to predicting wave height have been developed, including numerical models and machine learning methods, such as the Long-Short Term Memory (LSTM) algorithm, which has currently garnered significant attention from researchers. The objective of this research is to develop a forecast model for wind wave height using the LSTM algorithm in Seibu Island Waters, DKI Jakarta. The ERA5 dataset comprises zonal and meridional wind components and significant wave height, along with wind measurement data using the Automatic Weather System (AWS) instrument, are used to train and test to train and test the LSTM model. The research results show that the LSTM model can predict significant wave height effectively. Predictions using the ERA5 significant height dataset are observed to be closer to field data, with RMSE, MAE, and MAPE values of 0.1535 m, 0.1181 m, and 37.11% respectively. Thus, the model evaluation results indicate good performance, with relatively low RMSE and MAE values, and a good MAPE value. The highest accuracy in significant wave height prediction is found for forecasts one week (7 days) ahea
The Application of the Rabin-Karp Algorithm with the Synonym Recognition Approach to Detect Plagiarism in Student Assignments
Full text linkKemajuan teknologi yang pesat telah mempermudah segala hal, termasuk dalam bidang pendidikan. Namun, kecanggihan tersebut juga mengakibatkan penyalahgunaan teknologi, terutama dalam hal duplikasi atau plagiarisme. Masalah ini tidak hanya terjadi pada tugas esai tetapi juga pada kode program. Untuk mengatasi hal tersebut, telah dilakukan penelitian untuk mendeteksi plagiarisme pada tugas mahasiswa dengan menggunakan metode Rabin-Karp dan pendekatan Synonym Recognition. Penelitian ini menemukan bahwa tingkat kesamaan terkecil adalah 20%, sedangkan yang terbesar adalah 76%. Penelitian ini bertujuan untuk memberikan solusi yang cepat dan akurat untuk mencegah maraknya aktivitas plagiarisme di bidang akademik
Smart Product Recommendations in Web E-Commerce: Leveraging Apriori Algorithm for Market Basket Analysis
No full text The world of online commerce is becoming increasingly competitive, and to succeed in this field, it is not enough to showcase products to potential buyers. It is crucial to offer various products and keep product recommendations up-to-date, especially for customers who buy multiple items. To address this challenge, an intelligent system is needed that can automatically generate trending product recommendations based on sales data. In this research, the Market Basket Analysis (MBA) method analyzes consumer transaction data and identifies products often purchased together. The apriori algorithm is applied to generate association rules, and the Lift Ratio parameter is used to evaluate the strength of these rules. This research is implemented on an e-commerce website, and the generated association rules will be applied to provide automatic product recommendations based on recent sales trends. The results show that the automatic product recommendation system developed for the e-commerce website significantly helps users enhance their online shopping experience. Using the Lift Ratio parameter in validating association rules provides strong evidence of the relevance and accuracy of the generated product recommendations, which can increase customer satisfaction and sales potential