22 research outputs found
Visualisasi Serangan Remote to Local (R2L) Dengan Clustering K-means
Get PDFVisualisasi merupakan salah satu teknik untuk meningkatkan akurasi deteksi serangan yang terjadi di jaringan. Visulisasi bertujuan untuk mempermudah dalam mengenali dan menyimpulkan serangan terjadi. Clustering k-means dapat digunakan untuk mendeteksi paket serngan dan paket normal. Serangan remote to local adalah serngan yang dilakukan oleh attacker untuk mendapatkan akses akun ke sebuah sistem yang sebelumnya tidak memiliki akun ke sistem tersebut. Pola serangan R2L pada dataset DARPA dapat dikenali dengan beberapa paramer seperti source address, destination address, flags, ip length, dan tcp length
Defence against adversarial attacks on IoT detection systems using deep belief network
Get PDFAn Adversarial attack is a technique used to deceive machine learning models to make incorrect predictions by providing slightly modified inputs from the original. Intrusion detection system (IDS) is a crucial tool in computer network security for the detection of adversarial attacks. Deep learning is a trending method in both research and industry, and this study proposes the use of a deep belief network (DBN). DBN can recognize data with small differences, but is also vulnerable to adversarial attacks. Therefore, this research suggests an internet of things-intrusion detection system (IoT-IDS) architecture using a DBN that can counter adversarial attacks. The chosen adversarial attack for this study is the fast gradient sign method (FGSM) used to evaluate the IoT IDS using the DBN model. Testing was conducted in two scenarios: first, the model was trained without adversarial attacks; second, the model was trained with adversarial attacks. The test results indicate that the DBN model struggles to detect FGSM attacks, achieving an accuracy of only 46% when it is not trained with adversarial attacks. However, after training with the FGSM dataset, the DBN model successfully detected adversarial attacks with an accuracy of 97%
Two-steps feature selection for detection variant distributed denial of services attack in cloud environment
Get PDFThe prevalence of cloud computing among organizations poses a significant problem in ensuring security. Specifically, distributed denial of services (DDoS) attacks targeting cloud computing networks can lead to financial losses for consumers of cloud computing services. This assault has the potential to render cloud services inaccessible. The detection system serves as a remedy to prevent more substantial losses. This research aims to enhance the efficacy of the system detection model by integrating feature selection with three machine learning algorithms: decision tree (DT), random forest (RF), and naïve Bayes (NB). Therefore, our study suggests combining two phases of feature selection into the DDoS attack detection procedure. The first phase uses the information gain (IG) feature selection technique approach, and the second phase uses the principal component analysis (PCA) feature extraction approach. The technique is referred to as two-step feature selection. The test findings indicate that the implementation of two-step feature selection can enhance the performance of the DT and RF detection models by around 9%
the ENHANCE OBJECT TRACKING ON AUGMENTED REALITY USING HYBRID CONVOLUTIONAL NEURAL NETWORK AND FAST CORNER DETECTION
No full textMarkerless augmented reality (AR) is utilized in applications that do not require anchoring to the real world and do not require the use of physical markers (fiducial markers). Augmented object displays not only float but also allow for the automatic placement of 3D augmented reality objects on flat surfaces to enhance realism in real time. There are two challenges that need to be addressed in Markerless AR systems: object tracking and registration, as well as the influence of light intensity. Therefore, the objective of this research is to propose the use of Convolutional Neural Networks (CNN) and Features from Accelerated Segment Test (FAST) corner detection for tracking or detecting objects in markerless augmented reality systems. Testing was conducted using three epoch schemes: 10, 50, and 100. The test results were measured using several parameters, including the execution time, testing loss, and testing accuracy. The test results indicated an improvement in the performance of the tested object detection. The accuracy testing results of using the CNN and FAST corner detection methods were superior to those of the CNN-only method and FAST corner detection alone, reaching 98%. However, this method increases the processing time for object detection. Thus, the processing time of the CNN without FAST corner detection was faster
Ensemble Method for Anomaly Detection On the Internet of Things
Get PDFThe internet of things generates various types of data traffic with a very large amount of data traffic which has an impact on security issues, one of which is an attack on the Internet of Things network. In the IoT data traffic flow, which contains various data, it turns out that the portion of attack data traffic is usually smaller than normal traffic. Therefore, the attack detection method must be able to recognize the type of attack on a very large data traffic flow and unbalanced data. High data dimensions and unbalanced data are one of the challenges in detecting attacks. To overcome the large data dimensions, Chi-square was chosen as a feature selection technique. In this study, the ensemble method is proposed to improve the ability to detect anomalies in unbalanced data. To produce an ideal detection method, a combination of several classification algorithms such as Bayes Network, Naive Bayes, REPtree and J48 is used. The CICIDS-2017 dataset is used as experimental data because it has a high data dimension which contains unbalanced data. The test results show that the proposed Ensemble method can improve the performance of anomaly detection for high-dimensional data containing unbalanced dat
PATTERN CLASSIFICATION SIGN LANGUAGE USING FEATURES DESCRIPTORS AND MACHINE LEARNING
Get PDFSign language is way of communication for the deaf and speech impaired. In Indonesia, the utilization of a standardized language involves the incorporation of American Sign Language (ASL). ASL is employed for various communication needs, ranging from basic alphanumeric fingerspelling (A-Z and numbers) to the more complex SIBI form (comprising gesture vocabulary) in everyday interactions as well as formal contexts. This surge in the digitization of sign language underscores the ongoing advancements in research and development. The challenge in this research lies in the ability to recognize American Sign Language (ASL) with diverse intensities and invariant backgrounds. Therefore, the study emphasis is on proposing a suitable segmentation method comparison for multi-intensity ASL cases. Subsequently, global feature descriptor methods, including Color Histogram, Hu Moments, and Haralick Texture techniques, are applied for feature extraction. The result of the Logistic Regression method versus the supervised Random Forest checks accuracy and suitability in identifying ASL fingerspelling. The findings of this research is predictive value of logistic regression is 48%, with class Y having the highest precision (0.86), class V having the lowest accuracy (0.16), and class L having the highest recall (0.73). The maximum precision in classes B, F, H, I, K, Y, and Z is 1.00, and the lowest in class U is 0.58, while the highest recall is in class G, which is 1.00. The lowest is in class V, while the predictive value from the random forest is 86 percent. Class H has the greatest f1 score (0.99), while class U has the lowest f1 score (0.64). The Random Forest method outperforms the two methods suggested in the paper, according to the comparison
Designing consensus algorithm for collaborative signature-based intrusion detection system
Get PDFSignature-based collaborative intrusion detection system (CIDS) is highly depends on the reliability of nodes to provide IDS attack signatures. Each node in the network is responsible to provide new attack signature to be shared with other node. There are two problems exist in CIDS highlighted in this paper, first is to provide data consistency and second is to maintain trust among the nodes while sharing the attack signatures. Recently, researcher find that blockchain has a great potential to solve those problems. Consensus algorithm in blockchain is able to increase trusts among the node and allows data to be inserted from a single source of truth. In this paper, we are investigating three blockchain consensus algorithms: proof of work (PoW), proof of stake (PoS), and hybrid PoW-PoS chain-based consensus algorithm which are possibly to be implemented in CIDS. Finally, we design an extension of hybrid PoW-PoS chain-based consensus algorithm to fulfill the requirement. This extension we name it as proof of attack signature (PoAS)
RNN-Based Intrusion Detection System for Internet of Vehicles with IG, PCA, and RF Feature Selection
Get PDFCyberattacks in the Internet of Vehicles (IoV) threaten road safety and data integrity, requiring intrusion detection systems (IDS) that capture temporal patterns in vehicular traffic. This study develops a Recurrent Neural Network (RNN)-based IDS and evaluates three feature-selection strategies—Information Gain (IG), Principal Component Analysis (PCA), and Random Forest (RF)—on the CICIoV2024 dataset. Features are normalized using Min–Max scaling before being fed into the RNN classifier. The models achieve perfect classification on held-out tests (accuracy/precision/recall/F1 = 1.00). However, probabilistic evaluation reveals low ROC–AUC scores (IG: 0.572, PCA: 0.429, RF: 0.415), indicating limited discriminative margins and potential overfitting or calibration issues despite flawless confusion matrices. PCA and RF further reduce computational overhead during inference compared to IG. These findings highlight that relying solely on accuracy can be misleading for IDS evaluation; temporal RNNs should be complemented with probability-aware training, calibration, or hybrid architectures. This work contributes a temporal-aware IDS framework for IoV and motivates future research on real-time deployment, hybrid RNN-CNN/LSTM models, and adversarial robustness to improve generalization and safety of connected vehicle
