• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.39 no.1
• /
• pp.105-115
• /
• 2016

Ensemble classification involves combining individually trained classifiers to yield more accurate prediction, compared with individual models. Ensemble techniques are very useful for improving the generalization ability of classifiers. The random subspace ensemble technique is a simple but effective method for constructing ensemble classifiers; it involves randomly drawing some of the features from each classifier in the ensemble. The instance selection technique involves selecting critical instances while deleting and removing irrelevant and noisy instances from the original dataset. The instance selection and random subspace methods are both well known in the field of data mining and have proven to be very effective in many applications. However, few studies have focused on integrating the instance selection and random subspace methods. Therefore, this study proposed a new hybrid ensemble model that integrates instance selection and random subspace techniques using genetic algorithms (GAs) to improve the performance of a random subspace ensemble model. GAs are used to select optimal (or near optimal) instances, which are used as input data for the random subspace ensemble model. The proposed model was applied to both Kaggle credit data and corporate credit data, and the results were compared with those of other models to investigate performance in terms of classification accuracy, levels of diversity, and average classification rates of base classifiers in the ensemble. The experimental results demonstrated that the proposed model outperformed other models including the single model, the instance selection model, and the original random subspace ensemble model.

• Tribology and Lubricants
• /
• v.28 no.6
• /
• pp.265-271
• /
• 2012

In the preceding Part I study, for improving the unbalance response vibration of a large PRT motor-generator rotor fundamentally by design, a series of design analyses were carried out for bearing improvement by retrofitting from original plain partial journal bearings, applied for operation at a rated speed of 1,800 rpm, to final tilting pad journal bearings. To satisfy evenly key basic lubrication performances such as the minimum lift-off speed and maximum oil-film temperature, a design solution of 5-pad tilting pad journal bearings and maximizing the direct stiffness by about two times has been achieved. In this Part II study, a detailed rotordynamic analysis of the large PRT motor-generator rotor-bearing system will be performed, applying both the original plain partial journal bearings and the retrofitted tilting pad journal bearings, to confirm the effect of rotordynamic vibration improvement after retrofitting. The results show that the rotor unbalance response vibrations with the tilting pad journal bearings are greatly reduced by as much as about one ninth of those with the plain partial journal bearings. In addition, for the tilting pad journal bearings there exist no critical speed up to the rated speed and just one instance of a concerned critical speed around the rated speed, whereas for the plain partial journal bearings there exist one instance of a critical speed up to the rated speed and two instances of concerned critical speeds around the rated speed.

• Journal of the Semiconductor & Display Technology
• /
• v.19 no.3
• /
• pp.12-18
• /
• 2020

In semiconductor manufacturing, defect detection is critical to maintain high yield. Currently, computer vision systems used in semiconductor photo lithography still have adopt to digital image processing algorithm, which often occur inspection faults due to sensitivity to external environment. Thus, we intend to handle this problem by means of using Mask R-CNN instead of digital image processing algorithm. Additionally, Mask R-CNN can be trained with image dataset pre-processed by means of the specific designed digital image filter to extract the enhanced feature map of Convolutional Neural Network (CNN). Our approach converged advantage of digital image processing and instance segmentation with deep learning yields more efficient semiconductor photo lithography inspection system than conventional system.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2021.05a
• /
• pp.40-41
• /
• 2021

The construction industry has the highest occupational accidents/injuries among all industries. Korean government installed surveillance camera systems at construction sites to reduce occupational accident rates. Construction safety managers are monitoring potential hazards at the sites through surveillance system; however, the human capability of monitoring surveillance system with their own eyes has critical issues. Therefore, this study proposed to build a deep learning-based safety monitoring system that can obtain information on the recognition, location, identification of workers and heavy equipment in the construction sites by applying multiple-object tracking with instance segmentation. To evaluate the system's performance, we utilized the MS COCO and MOT challenge metrics. These results present that it is optimal for efficiently automating monitoring surveillance system task at construction sites.

• Journal of Intelligence and Information Systems
• /
• v.20 no.4
• /
• pp.121-139
• /
• 2014

Predicting corporate failure has been an important topic in accounting and finance. The costs associated with bankruptcy are high, so the accuracy of bankruptcy prediction is greatly important for financial institutions. Lots of researchers have dealt with the topic associated with bankruptcy prediction in the past three decades. The current research attempts to use ensemble models for improving the performance of bankruptcy prediction. Ensemble classification is to combine individually trained classifiers in order to gain more accurate prediction than individual models. Ensemble techniques are shown to be very useful for improving the generalization ability of the classifier. Bagging is the most commonly used methods for constructing ensemble classifiers. In bagging, the different training data subsets are randomly drawn with replacement from the original training dataset. Base classifiers are trained on the different bootstrap samples. Instance selection is to select critical instances while deleting and removing irrelevant and harmful instances from the original set. Instance selection and bagging are quite well known in data mining. However, few studies have dealt with the integration of instance selection and bagging. This study proposes an improved bagging ensemble based on instance selection using genetic algorithms (GA) for improving the performance of SVM. GA is an efficient optimization procedure based on the theory of natural selection and evolution. GA uses the idea of survival of the fittest by progressively accepting better solutions to the problems. GA searches by maintaining a population of solutions from which better solutions are created rather than making incremental changes to a single solution to the problem. The initial solution population is generated randomly and evolves into the next generation by genetic operators such as selection, crossover and mutation. The solutions coded by strings are evaluated by the fitness function. The proposed model consists of two phases: GA based Instance Selection and Instance based Bagging. In the first phase, GA is used to select optimal instance subset that is used as input data of bagging model. In this study, the chromosome is encoded as a form of binary string for the instance subset. In this phase, the population size was set to 100 while maximum number of generations was set to 150. We set the crossover rate and mutation rate to 0.7 and 0.1 respectively. We used the prediction accuracy of model as the fitness function of GA. SVM model is trained on training data set using the selected instance subset. The prediction accuracy of SVM model over test data set is used as fitness value in order to avoid overfitting. In the second phase, we used the optimal instance subset selected in the first phase as input data of bagging model. We used SVM model as base classifier for bagging ensemble. The majority voting scheme was used as a combining method in this study. This study applies the proposed model to the bankruptcy prediction problem using a real data set from Korean companies. The research data used in this study contains 1832 externally non-audited firms which filed for bankruptcy (916 cases) and non-bankruptcy (916 cases). Financial ratios categorized as stability, profitability, growth, activity and cash flow were investigated through literature review and basic statistical methods and we selected 8 financial ratios as the final input variables. We separated the whole data into three subsets as training, test and validation data set. In this study, we compared the proposed model with several comparative models including the simple individual SVM model, the simple bagging model and the instance selection based SVM model. The McNemar tests were used to examine whether the proposed model significantly outperforms the other models. The experimental results show that the proposed model outperforms the other models.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2002.05a
• /
• pp.565-568
• /
• 2002

A lot of rotating machines are being used in the industrial world and electric motor and generator take the most part of it. When it comes to the electric motor and generator, we can not help thinking about the eddy current because it brings a loss of electric and can be a important reason of the heat generation. To attenuate eddy current. laminated silicon steel sheets are being used in general. Especially, laminated rotor is being used for rotating part of the electric motor and generator and it decreases electrical loss and heat generation but we can be faced with another problem. In general, most of the motor and generator can be normally operated under 3600rpm because they are designed to have the first critical speed more than that speed. But nowadays, they should be operated more than the first critical speed as usual with the trend of high speed. large scale and high precision in industrial world. The critical speed can be determined from the inertia and stiffness for the rotor and bearing of rotating systems. The laminated rotor stiffness can be hardly determined because it can be derived a lot factors for instance rotor material and shape. lamination material and shape. insulation material. lamination force and so on. In this paper, the change of the natural frequency of the motor was examined with the change of the lamination force as an experimental method.

• Earthquakes and Structures
• /
• v.1 no.2
• /
• pp.147-162
• /
• 2010

The ground acceleration measured at a point on the earth's surface is composed of several waves that have different phase velocities, arrival times, amplitudes, and frequency contents. For instance, body waves contain primary and secondary waves that have high frequency content and reach the site first. Surface waves are composed of Rayleigh and Love waves that have lower phase velocity, lower frequency content and reach the site next. Some of these waves could be of more damage to the structure depending on their frequency content and associated amplitude. This paper models critical earthquake loads for single-degree-of-freedom (SDOF) inelastic structures considering evolution of the seismic waves in time and frequency. The ground acceleration is represented as combination of seismic waves with different characteristics. Each seismic wave represents the energy of the ground motion in certain frequency band and time interval. The amplitudes and phase angles of these waves are optimized to produce the highest damage in the structure subject to explicit constraints on the energy and the peak ground acceleration and implicit constraints on the frequency content and the arrival time of the seismic waves. The material nonlinearity is modeled using bilinear inelastic law. The study explores also the influence of the properties of the seismic waves on the energy demand and damage state of the structure. Numerical illustrations on modeling critical earthquake excitations for one-storey inelastic frame structures are provided.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2003.06a
• /
• pp.788-792
• /
• 2003

Rotating machines are widely used in industrial world and especially motor and generator take up much part of it. As for this kind of motor and generator, electrical loss due to eddy current is the very important factor and that is also a primary factor causes heat generation. To solve this kind of problem like the above. insulated laminating silicon steel sheet is used to prevent eddy current effect. Laminated rotor is widely used as rotating shaft of motor and generator. Due to that, electrical loss and heat problem can be solved but designer meets another problem. In general. most of the motor and generator can be normally operated under 3,600 rpm because they are designed to have the first critical speed more than that speed. But nowadays, they should be operated more than the first critical speed as usual with the trend of high speed, large scale and high precision in industrial world. The critical speed can be determined from the inertia and stillness for the rotor and bearing of rotating systems. The laminated rotor stiffness can be hardly determined because it can be derived a lot factors for instance rotor material and shape, lamination material and shape, insulation material. lamination force and so on. In this paper, the change of the natural frequency of the motor was examined with the change of the lamination force as an experimental method and design criteria will be presented for motor & generator designer, who can apply the result of numerical analysis with equivalent diameter scheme with ease.

• Journal of Environmental Impact Assessment
• /
• v.16 no.1
• /
• pp.45-58
• /
• 2007

Despite the ubiquitous practice of environmental impact assessment around the world, social impact assessment has been alienated from assessment process from the beginning. Not until 1993 was 'Interorganizational Committee' established in the United States to prepare for 'Principles and Guidelines for Social Impact Assessment.' This study is an attempt to critically examine US 'Principles and Guidelines.' First, the study traces history of social impact assessment to reveal why the latter became "the orphan in the assessment process." Second, it critically reviews 'Principles and Guidelines' to find its merits and defects. For instance, a principle regarding environmental justice is perceived as necessary as society has become conscious of social justice and equity while putting too much emphasis on predictive traits of social impact assessments only fosters "checklist mentality." Third, the study reflects on 'Principles and Guidelines' in particular and social impact assessment in general in order to probe what is social impact assessment. To do so, it pays attention to scholars, who have criticized technocratic and procedural elements of 'Principles and Guidelines.' They show that social impact assessment is philosophically and methodologically teleological in that "fluid and contested meanings" between social impacts and the public are meaningful in itself. And simple procedural guarantee of the public involvement, they argue, is not enough to define social impacts. Lastly, from the critical analysis of 'Principles and Guideline,' the study looks for alternatives to improve how to assess social impacts in a Korean context.

• Ocean Systems Engineering
• /
• v.6 no.1
• /
• pp.99-115
• /
• 2016

In meeting the technical needs for deepwater conditions and overcoming the shortfalls of single-layer pipes for deepwater applications, pipe-in-pipe (PIP) systems have been developed. While, for PIP pipelines directly laid on the seabed or with partial embedment, one of the primary service risks is lateral buckling. The critical axial force is a key factor governing the global lateral buckling response that has been paid much more attention. It is influenced by global imperfections, submerged weight, stiffness, pipe-soil interaction characteristics, et al. In this study, Finite Element Models for imperfect PIP systems are established on the basis of 3D beam element and tube-to-tube element in Abaqus. A parameter study was conducted to investigate the effects of these parameters on the critical axial force and post-buckling forms. These parameters include structural parameters such as imperfections, clearance, and bulkhead spacing, pipe/soil interaction parameter, for instance, axial and lateral friction properties between pipeline and seabed, and load parameter submerged weight. Python as a programming language is been used to realize parametric modeling in Abaqus. Some conclusions are obtained which can provide a guide for the design of PIP pipelines.