• Journal of IKEEE
• /
• v.24 no.2
• /
• pp.475-485
• /
• 2020

Recently, a crowdsensing system that provides a new sensing service with real-time sensing data provided from a user's device including a sensor without installing a separate sensor has attracted attention. In the crowdsensing system, meaningless data may be provided due to a user's operation error or communication problem, or false data may be provided to obtain compensation. Therefore, the detection and removal of the abnormal data determines the quality of the crowdsensing service. The proposed methods in the past to detect these anomalies are not efficient for the fast-changing environment of crowdsensing. This paper proposes an anomaly data detection method by extracting the characteristics of continuously and rapidly changing sensing data environment by using machine learning technology and modeling it with an appropriate algorithm. We show the performance and feasibility of the proposed system using deep learning binary classification model of supervised learning and autoencoder model of unsupervised learning.

• Journal of Information Processing Systems
• /
• v.20 no.3
• /
• pp.375-390
• /
• 2024

Edge computing architecture has effectively alleviated the computing pressure on cloud platforms, reduced network bandwidth consumption, and improved the quality of service for user experience; however, it has also introduced new security issues. Existing anomaly detection methods in big data scenarios with cloud-edge computing collaboration face several challenges, such as sample imbalance, difficulty in dealing with complex network traffic attacks, and difficulty in effectively training large-scale data or overly complex deep-learning network models. A lightweight deep-learning model was proposed to address these challenges. First, normalization on the user side was used to preprocess the traffic data. On the edge side, a trained Wasserstein generative adversarial network (WGAN) was used to supplement the data samples, which effectively alleviates the imbalance issue of a few types of samples while occupying a small amount of edge-computing resources. Finally, a trained lightweight deep learning network model is deployed on the edge side, and the preprocessed and expanded local data are used to fine-tune the trained model. This ensures that the data of each edge node are more consistent with the local characteristics, effectively improving the system's detection ability. In the designed lightweight deep learning network model, two sets of convolutional pooling layers of convolutional neural networks (CNN) were used to extract spatial features. The bidirectional long short-term memory network (BiLSTM) was used to collect time sequence features, and the weight of traffic features was adjusted through the attention mechanism, improving the model's ability to identify abnormal traffic features. The proposed model was experimentally demonstrated using the NSL-KDD, UNSW-NB15, and CIC-ISD2018 datasets. The accuracies of the proposed model on the three datasets were as high as 0.974, 0.925, and 0.953, respectively, showing superior accuracy to other comparative models. The proposed lightweight deep learning network model has good application prospects for anomaly traffic detection in cloud-edge collaborative computing architectures.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2014.10a
• /
• pp.97-98
• /
• 2014

Vibration signal has been widely utilized in the diagnostics of rotating mechanical system. Diagnostics systems in rotating machinery are depends on the vibration data which are acquired from the system. However, the characteristics of acquired data can be vary according to the position of anomaly installed or the position of data acquired. In this research, the coordinate transform of journal bearing vibration signal was proposed to overcome the limitation mentioned above. Journal bearing systems are equipped two gap sensors with ninety degree angles, and it can enable to generate coordinate transformed signals in arbitrary angle domain. More abundant information for each normal or anomaly conditions are obtained from coordinate transformation than only the data of the existing measuring position, then we have developed a reliable and robust diagnosis algorithm for journal bearing system.

• Economic and Environmental Geology
• /
• v.31 no.1
• /
• pp.59-68
• /
• 1998

Gravity, magnetic and VLF surveys were carried out to investigate the dimension, nature and stability of the waste materials filled in the Seokdae landfill, Pusan. The Seokdae landfill, which is located in a former valley, was used as a dump for mainly domestic-type waste materials for 6 years from 1987. The landfill site is classfied into A, B, C and D areas according to the sequence of dumping period. The Bouguer gravity anomaly map shows maximum variation of 3.1 mgals on the landfill and its general appearance has close relation with the thickness of waste filled. The local variation of anomaly, however, reflect the degree of compactness of waste materials which may be affected by the nature of waste and dumping time. In the case of area A, where dumping process was terminated at the very last stage, most part show negative anomaly compared to other areas. We think that the composition of the waste materials in the area A is high in leftover food and paper trash and they are still in uncompacted condition. In area B, the general trend of variation of gravity anomaly is appeared to be high anomaly in northern part and decrease to the southern part. This is well matched with the prelandfill topography of the landfill site. The southern part of area B is located in the center of valley and its present surface is comparatively rugged, which may be due to the differential settlement of deep burried waste. The thickness of waste in area C is relatively thin, but the gravity anomaly appears to be low. Considering the present condition of surface, it can be inferred that low density wastes such as leftover food were mainly filled in this area. Area D, as in the case of area B, shows gravity anomaly that has close relation with the prelandfill topography. Magnetic data show the variation of total field intensity varies in the range of 46600~51000 nT, and reach maximum anomaly of 4400 nT. The overall pattern of magnetic anomaly well reflects the distribution of magnetic materials in the landfill. The result of VLF survey reveals several low resistivity zones, which may serve as underground passages for contaminant flow, in the area C located near the small Village.

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.12
• /
• pp.411-418
• /
• 2020

In the field of information security, IDS(Intrusion Detection System) is normally classified in two different categories: signature-based IDS and anomaly-based IDS. Many studies in anomaly-based IDS have been conducted that analyze network traffic data generated in cyberspace by machine learning algorithms. In this paper, we studied pre-processing methods to overcome performance degradation problems cashed by rare classes. We experimented classification performance of a Machine Learning algorithm by reconstructing data set based on rare classes and semi rare classes. After reconstructing data into three different sets, wrapper and filter feature selection methods are applied continuously. Each data set is regularized by a quantile scaler. Depp neural network model is used for learning and validation. The evaluation results are compared by true positive values and false negative values. We acquired improved classification performances on all of three data sets.

• Bulletin of the Korean Space Science Society
• /
• 2011.04a
• /
• pp.26.2-26.2
• /
• 2011

Flyby anomaly (unexpected energy increase during Earth Gravity Assists) indicates existence of an unknown non-conservative perturbation which affects hyperbolic trajectories. This presentation focuses on first order position and velocity perturbation formulas derived in terms of classical orbital element variations for hyperbolic orbit. By using both the perturbation formulas and numerical approach, we analyze effects of hypothetical acceleration models proposed by Hasse (2009), Lewis (2009), Gerrad and Sumner (2008), and Busack (2007). Based on analysis of perturbation effect on low earth orbit, we find that typical position perturbation is about 10m which is much larger than current orbit determination accuracy. From this, we deduce that anomalous acceleration only affects hyperbolic orbit or behaves differently in bound orbit. On the other hand, based on analysis of perturbation effects on hyperbolic trajectories, we find that position and velocity perturbations are highly different from acceleration models, and all of proposed models fail to explain observed range and Doppler data. Thus, it can be concluded that not only energy variations but also kinematics gives us crucial clues on the flyby anomaly, and kinematical characteristic should be considered in modeling flyby anomaly.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.14 no.2
• /
• pp.205-212
• /
• 2011

In this paper, we have proposed a method for quantization of the stratification strength in the sea water and analysing the distributions of the maximum stratification depths calculated by the method at the seas near the Korean peninsular. For calculating the stratification strength, modified and applied the potential energy anomaly formular which was suggested by Simpson in 1977. The data had been collected by NFRDI from 1971 to 2008 were used to determine the maximum vertical density gradient depth and the relative potential energy anomaly at that depth. In the East Sea, the stratification depth has become deepened about 20m in February and April since 1971. In Yellow-South Sea, the maximum density gradient depth has been deepened about 10m only in December during the same period and the difference of the stratification depth between summer and winter has been enlarged. These trends of variation of stratification strength and depth near the Korean peninsular should be investigated more carefully and continuously. And the results of these studies could be adopted for the more efficient operation of underwater weapon and detection systems.

• ETRI Journal
• /
• v.39 no.5
• /
• pp.621-631
• /
• 2017

Abnormal samples are usually difficult to obtain in production systems, resulting in imbalanced training sample sets. Namely, the number of positive samples is far less than the number of negative samples. Traditional Support Vector Machine (SVM)-based anomaly detection algorithms perform poorly for highly imbalanced datasets: the learned classification hyperplane skews toward the positive samples, resulting in a high false-negative rate. This article proposes a new imbalanced SVM (termed ImSVM)-based anomaly detection algorithm, which assigns a different weight for each positive support vector in the decision function. ImSVM adjusts the learned classification hyperplane to make the decision function achieve a maximum GMean measure value on the dataset. The above problem is converted into an unconstrained optimization problem to search the optimal weight vector. Experiments are carried out on both Cloud datasets and Knowledge Discovery and Data Mining datasets to evaluate ImSVM. Highly imbalanced training sample sets are constructed. The experimental results show that ImSVM outperforms over-sampling techniques and several existing imbalanced SVM-based techniques.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.3
• /
• pp.127-136
• /
• 2024

This study explores the correlation between system anomalies and large-scale logs within the Spark cluster environment. While research on anomaly detection using logs is growing, there remains a limitation in adequately leveraging logs from various components of the cluster and considering the relationship between anomalies and the system. Therefore, this paper analyzes the distribution of normal and abnormal logs and explores the potential for anomaly detection based on the occurrence of log templates. By employing Hadoop and Spark, normal and abnormal log data are generated, and through t-SNE and K-means clustering, templates of abnormal logs in anomalous situations are identified to comprehend anomalies. Ultimately, unique log templates occurring only during abnormal situations are identified, thereby presenting the potential for anomaly detection.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.8
• /
• pp.2298-2315
• /
• 2024

This paper introduces a method aimed at diagnosing the presence or absence of lesions by detecting anomalies in Chest X-ray images. The proposed approach is based on the PatchCore anomaly detection method, which extracts a feature vector containing location information of an image patch from normal image data and calculates the anomaly distance from the normal vector. However, applying PatchCore directly to medical image processing presents challenges due to the possibility of diseases occurring only in specific organs and the presence of image noise unrelated to lesions. In this study, we present an image alignment method that utilizes affine transformation parameter prediction to standardize already captured X-ray images into a specific composition. Additionally, we introduce a region-specific abnormality detection method that requires affine-transformed chest X-ray images. Furthermore, we propose a method to enhance application efficiency and performance through feature map hard masking. The experimental results demonstrate that our proposed approach achieved a maximum AUROC (Area Under the Receiver Operating Characteristic) of 0.774. Compared to a previous study conducted on the same dataset, our method shows a 6.9% higher performance and improved accuracy.