• The Journal of the Acoustical Society of Korea
• /
• v.43 no.2
• /
• pp.225-233
• /
• 2024

Active sonar transmits sound waves to detect covertly maneuvering underwater objects and detects the signals reflected back from the target. However, in addition to the target's echo, the active sonar's received signal is mixed with seafloor, sea surface reverberation, biological noise, and other noise, making target recognition difficult. Conventional techniques for detecting signals above a threshold not only cause false detections or miss targets depending on the set threshold, but also have the problem of having to set an appropriate threshold for various underwater environments. To overcome this, research has been conducted on automatic calculation of threshold values through techniques such as Constant False Alarm Rate (CFAR) and application of advanced tracking filters and association techniques, but there are limitations in environments where a significant number of detections occur. As deep learning technology has recently developed, efforts have been made to apply it in the field of underwater target detection, but it is very difficult to acquire active sonar data for discriminator learning, so not only is the data rare, but there are only a very small number of targets and a relatively large number of non-targets. There are difficulties due to the imbalance of data. In this paper, the image of the energy distribution of the detection signal is used, and a classifier is learned in a way that takes into account the imbalance of the data to distinguish between targets and non-targets and added to the existing technique. Through the proposed technique, target misclassification was minimized and non-targets were eliminated, making target recognition easier for active sonar operators. And the effectiveness of the proposed technique was verified through sea experiment data obtained in the East Sea.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.108-111
• /
• 2022

Applying deep learning to machine vision systems for defect detection of products requires vast amounts of training data about various defect cases. However, since data imbalance occurs according to the type of defect in the actual manufacturing industry, it takes a lot of time to collect product images enough to generalize defect cases. In this paper, we apply a Siamese neural network that can be learned with even a small amount of data to product defect detection, and modify the image pairing method and contrastive loss function by properties the situation of product defect image data. We indirectly evaluated the embedding performance of Siamese neural networks using AUC-ROC, and it showed good performance when the images only paired among same products, not paired among defective products, and learned with exponential contrastive loss.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2005.05a
• /
• pp.599-606
• /
• 2005

Novelty detection involves identifying novel patterns. They are not usually available during training. Even if they are, the data quantity imbalance leads to a low classification accuracy when a supervised learning scheme is employed. Thus, an unsupervised learning scheme is often employed ignoring those few novel patterns. In this paper, we propose two ways to make use of the few available novel patterns. First, a scheme to determine local thresholds for the Self Organizing Map boundary is proposed. Second, a modification of the Learning Vector Quantization learning rule is proposed so that allows one to keep codebook vectors as far from novel patterns as possible. Experimental results are quite promising.

• Journal of Astronomy and Space Sciences
• /
• v.1 no.1
• /
• pp.23-34
• /
• 1984

A bubble trapped in the liquid manifold of INTELSAT IV F-7 spacecraft caused a mass imbalance between the System 1 propellant tanks and a wobble half angle of 0.38 degree to 0.48 degree. A maneuver on May 14, 1980 passed the bubble through the axial jet and allowed propellant to redistribute. A 0.2 rpm change in sin rate was observed with an exponential decay time constant of 6 minutes. In this paper, moment of inertia, tank geometry and hydrodynamic models are derived to match the observed spin rate data. The values of the total mass of propellant considered were 16, 19 and 20 kgs with corresponding mass imbalances of 14.3, 15 and 15.1 Kgs, respectively. The result shows excellent agreement with observed spin rate data but it was necessary to assume a greater mass of hydrazine in the tanks than propellant accounting indicated.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.61 no.3
• /
• pp.355-360
• /
• 2012

It is difficult to converge power flow for the power system planning data. The main cause of power flow diverse is reactive power imbalance. A active power could be adjust by ELD or merit order but a reactive power couldn't dispatch before power flow analysis. The lack of reactive power of power system is cause a inadequate voltage drop This paper suggest new reactive power dispatch algorithm using switched shunt admittance. This algorithm uses reactive power sensitivity called switch shunt jacobian. When proposed algorithm applies to real system data that couldn't be conversed in PSS/E the power flow analysis is converged.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.9
• /
• pp.3539-3558
• /
• 2015

This paper investigates and analyzes the default block placement strategy of HDFS. HDFS is a typical representative distributed file system to stream vast amount of data effectively at high bandwidth to user applications. However, the default HDFS block placement policy assumes that all nodes in the cluster are homogeneous, and places blocks with a simple RoundRobin strategy without considering any nodes' resource characteristics, which decreases self-adaptability of the system. The primary contribution of this paper is the proposition of a network load sensitive block placement strategy. We have implemented our algorithm and justify it through extensive simulations and comparison with similar existing studies. The results indicate that our work not only performs much better in the data distribution but also improves write performance more significantly than the others.

• Korean Journal of Health Education and Promotion
• /
• v.19 no.1
• /
• pp.199-212
• /
• 2002

Stress not only brings disorders in social, family life but brings also changes in eating behaviors so that the imbalance of food intake is induced. Our study was carried out to find out the association between the degree of stress and eating behaviors for the subjects of 262 female college students in the Suwon area. The study instrument was a structured questionnaire composed with questions about demographic data, food preferences, questions related to life styles, and health. All of the collected data was analyzed by the degree of stress score, grouping into high(25〈) and low stress group(〈=25). High stress group showed unhealthy lifestyles, higher percentage of irregular menstruation, smoking, dizziness, higher amount of alcohol consumption. They also showed higher preference rate of spicy foods including hot foods, oil based foods and showed percentage of removing visible fat when eating meats. They more wanted to control weight and prefers snacking compared to low stress group. These results show that stress changes in eating behaviors in a way of unhealthy life styles.

• KDI Journal of Economic Policy
• /
• v.40 no.2
• /
• pp.1-21
• /
• 2018

The private brands (PB) of corporate retailers are booming in Korea. This paper examines the effect of the rise of PB on Korean retail and manufacturing. By utilizing both store-level data and firm-level data, I find that the expansion of PB elevates the profits of corporate retailers but does not significantly affect, and in some cases even reduces, those of subcontracting manufacturers. This occurs not only because sales of national brands (NB) decline due to the launch of similar PBs but also because the imbalance in the bargaining positions of the two parties has caused retail margins to be set high while manufacturers' operating profits are set low. The paper provides policy recommendations for fair contracts and cooperative development between retail and manufacturing companies.

• Journal of Internet Computing and Services
• /
• v.4 no.5
• /
• pp.51-60
• /
• 2003

Asynchronous iteration is a way to reduce performance degradation of some parallel algorithms due to load imbalance or transmission delay between computing nodes, which requires asymmetric communication between the nodes of different speeds. To implement such asynchronous communication on distributed memory systems, we suggest an MPMD method that creates an additional separate server process on each computing node, and compare it with an SPMD method that creates a single process per node.

• International Journal of Computer Science & Network Security
• /
• v.24 no.4
• /
• pp.1-10
• /
• 2024

In the era of big data, the growth of e-commerce transactions brings forth both opportunities and risks, including the threat of data theft and fraud. To address these challenges, an automated real-time fraud detection system leveraging machine learning was developed. Four algorithms (Decision Tree, Naïve Bayes, XGBoost, and Neural Network) underwent comparison using a dataset from a clothing website that encompassed both legitimate and fraudulent transactions. The dataset exhibited an imbalance, with 9.3% representing fraud and 90.07% legitimate transactions. Performance evaluation metrics, including Recall, Precision, F1 Score, and AUC ROC, were employed to assess the effectiveness of each algorithm. XGBoost emerged as the top-performing model, achieving an impressive accuracy score of 95.85%. The proposed system proves to be a robust defense mechanism against fraudulent activities in e-commerce, thereby enhancing security and instilling trust in online transactions.