• Journal of the Korean Society for Marine Environment & Energy
• /
• v.20 no.1
• /
• pp.45-54
• /
• 2017

The youth unemployment problem has become a consistent issue with the number recently surpassing 1million. The marine and fisheries sector, being no exception, is having problems attracting outstanding individuals to the sector on the one hand, and in providing high quality jobs on the other, resulting in an imbalance in the supply and demand of the marine and fisheries sector workforce. In order to supply a workforce that meets the future and on the ground demands, addressing of the qualitative rather than the quantitative aspects of the imbalance issue is more important. Thus, the following strategies are recommended: 1) focus on developing a highly skilled workforce that corresponds to future and on the ground demands; 2) improve educational infrastructures such as training equipment, and enhance the professional capacity of school teachers; 3) establish an integrated system for the management of the education and re-education of human resources.

• Journal of KIISE
• /
• v.44 no.12
• /
• pp.1275-1281
• /
• 2017

The supervised learning approach is suitable for classification of insulting sentences, but pre-decided training sentences are necessary. Since a Character-level Convolution Neural Network is robust for each character, so is appropriate for classifying abusive sentences, however, has a drawback that demanding a lot of training sentences. In this paper, we propose transfer learning method that reusing the trained filters in the real classification process after the filters get the characteristics of offensive words by generated abusive/normal pair of sentences. We got higher performances of the classifier by decreasing the effects of data shortage and class imbalance. We executed experiments and evaluations for three datasets and got higher F1-score of character-level CNN classifier when applying transfer learning in all datasets.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.66 no.11
• /
• pp.1584-1590
• /
• 2017

This paper presents a implementation method of reliable speed limitation controller considering motor parameter variation in high speed operation. In spinning process of drum washing machine, speed increase has to be limited when unallowable imbalance mass is detected. Otherwise, severe noise and vibration can happen because noise and vibration are proportional to imbalance mass. To detect imbalance mass, d-axis current magnitude is used. However, we have to compensate for back-emf and power supply variation by means of detecting them because d-axis current is affected by both of them. On the other hand, we have to carefully estimate back-emf because back-emf is affected by stator resistance variation and inverter voltage error. Stator resistance variation can happen by manufacturing process for mass production or temperature variation in running. And there are inverter voltage errors between command voltage from micro-computer to inverter and real voltage from inverter to motor because of rising and falling time delay and turn-on resistance of power semiconductor switch. To solve this problem, we propose 2-step align current injection method which is to inject step-wise current right before starting. By this method, we can simply obtain stator resistance by ratio of voltage without inverter voltage error and current, and we can measure inverter voltage error. So we can obtain more exact model current, and then by simple calculation with compensation gain, we can estimate more accurate motor back-emf. We show that this method works well. It is verified through experiments.

• Journal of Internet Computing and Services
• /
• v.22 no.1
• /
• pp.13-22
• /
• 2021

Recently network based attack technologies are rapidly advanced and intelligent, the limitations of existing signature-based intrusion detection systems are becoming clear. The reason is that signature-based detection methods lack generalization capabilities for new attacks such as APT attacks. To solve these problems, research on machine learning-based intrusion detection systems is being actively conducted. However, in the actual network environment, attack samples are collected very little compared to normal samples, resulting in class imbalance problems. When a supervised learning-based anomaly detection model is trained with such data, the result is biased to the normal sample. In this paper, we propose to overcome this imbalance problem through One-Class Anomaly Detection using an auto encoder. The experiment was conducted through the NSL-KDD data set and compares the performance with the supervised learning models for the performance evaluation of the proposed method.

• Convergence Security Journal
• /
• v.23 no.1
• /
• pp.69-77
• /
• 2023

Recently, ransomware damage has been increasing rapidly around the world, including Irish health authorities and U.S. oil pipelines, and is causing damage to all sectors of society. In particular, research using machine learning as well as existing detection methods is increasing for ransomware detection and response. However, traditional machine learning has a problem in that it is difficult to extract accurate predictions because the model tends to predict in the direction where there is a lot of data. Accordingly, in an imbalance class consisting of a large number of non-Ransomware (normal code or malware) and a small number of Ransomware, a technique for resolving the imbalance and improving ransomware detection performance is proposed. In this experiment, we use two scenarios (Binary, Multi Classification) to confirm that the sampling technique improves the detection performance of a small number of classes while maintaining the detection performance of a large number of classes. In particular, the proposed mixed sampling technique (SMOTE+ENN) resulted in a performance(G-mean, F1-score) improvement of more than 10%.

• Journal of the Korea Convergence Society
• /
• v.13 no.1
• /
• pp.23-30
• /
• 2022

The failure of the reefer container causes a great loss of cost, but the current reefer container alarm system is inefficient. Existing studies using simulation data of refrigeration systems exist, but studies using actual operation data of refrigeration containers are lacking. Therefore, this study classified the causes of failure using actual refrigerated container operation data. Data imbalance occurred in the actual data, and the data imbalance problem was solved by comparing the logistic regression analysis with ENN-SMOTE and class weight with the 2-stage algorithm developed in this study. The 2-stage algorithm uses XGboost, LGBoost, and DNN to classify faults and normalities in the first step, and to classify the causes of faults in the second step. The model using LGBoost in the 2-stage algorithm was the best with 99.16% accuracy. This study proposes a final model using a two-stage algorithm to solve data imbalance, which is thought to be applicable to other industries.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.2
• /
• pp.263-283
• /
• 2024

Users' comments after online shopping are critical to product reputation and business improvement. These comments, sometimes known as e-commerce reviews, influence other customers' purchasing decisions. To confront large amounts of e-commerce reviews, automatic analysis based on machine learning and deep learning draws more and more attention. A core task therein is sentiment analysis. However, the e-commerce reviews exhibit the following characteristics: (1) inconsistency between comment content and the star rating; (2) a large number of unlabeled data, i.e., comments without a star rating, and (3) the data imbalance caused by the sparse negative comments. This paper employs Bidirectional Encoder Representation from Transformers (BERT), one of the best natural language processing models, as the base model. According to the above data characteristics, we propose the F_MixBERT framework, to more effectively use inconsistently low-quality and unlabeled data and resolve the problem of data imbalance. In the framework, the proposed MixBERT incorporates the MixMatch approach into BERT's high-dimensional vectors to train the unlabeled and low-quality data with generated pseudo labels. Meanwhile, data imbalance is resolved by Focal loss, which penalizes the contribution of large-scale data and easily-identifiable data to total loss. Comparative experiments demonstrate that the proposed framework outperforms BERT and MixBERT for sentiment analysis of e-commerce comments.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.44 no.3
• /
• pp.106-116
• /
• 2021

This paper considers a joint problem for blood inventory planning at hospitals and blood delivery planning from blood centers to hospitals, in order to alleviate the blood service imbalance between big and small hospitals being occurred in practice. The joint problem is to determine delivery timing, delivery quantity, delivery means such as medical drones and legacy blood vehicles, and inventory level to minimize inventory and delivery costs while satisfying hospitals' blood demand over a planning horizon. This problem is formulated as a mixed integer programming model by considering practical constraints such as blood lifespan and drone specification. To solve the problem, this paper employs a Lagrangian relaxation technique and suggests a time efficient Lagrangian heuristic algorithm. The performance of the suggested heuristic is evaluated by conducting computational experiments on randomly-generated problem instances, which are generated by mimicking the real data of Korean Red Cross in Seoul and other reliable sources. The results of computational experiments show that the suggested heuristic obtains near-optimal solutions in a shorter amount of time. In addition, we discuss the effect of changes in the length of blood lifespan, the number of planning periods, the number of hospitals, and drone specifications on the performance of the suggested Lagrangian heuristic.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.20 no.1
• /
• pp.100-106
• /
• 2009

This paper deals with an estimation problem of the gain and phase imbalances between the in-phase and quadrature components in the quadrature receivers which are widely used in wireless communications. It is shown that the estimates derived from the suggested auto-calibration algorithm is asymptotically minimum-variance unbiased as a function of the sampling time. In order to show this characteristic, the probability density functions of the estimates for the gain and phase imbalances are derived first. Then the mean and variance functions are investigated analytically or numerically based on the density functions.

• Journal of Power Electronics
• /
• v.18 no.4
• /
• pp.1245-1254
• /
• 2018

The stability of a grid-connected system (GCS) has become a critical issue with the increasing utilization of renewable energy sources. Under grid faults, however, a grid-connected inverter cannot work efficiently by using only the traditional droop control. In addition, the unbalance factor of voltage/current at the common coupling point (PCC) may increase significantly. To ensure the stable operation of a GCS under grid faults, the capability to compensate for grid imbalance should be integrated. To solve the aforementioned problem, an improved voltage-type grid-connected control strategy is proposed in this study. A negative sequence conductance compensation loop based on a positive sequence power droop control is added to maintain PCC voltage balance and reduce grid current imbalance, thereby meeting PCC power quality requirements. Moreover, a stable analysis is presented based on the small signal model. Simulation and experimental results verify the aforementioned expectations, and consequently, the effectiveness of the proposed control scheme.