• Journal of Broadcast Engineering
• /
• v.17 no.5
• /
• pp.781-788
• /
• 2012

In HEVC, the fast search method is used for reducing the complexity of the motion prediction procedure. It is consisted of the sub-sampled SAD which reduce the complexity of Sum of Absolute Differences(SAD) calculation and the simplified bi-prediction method which reduce the iterations of the uni-prediction for the bi-prediction. The computational complexity is largely decreased by the fast search method but the coding gain is also decreased. In this paper, the simplified bi-prediction is extended to compensate the performance loss and the prediction cost based complexity reduction methods are also proposed to reduce the complexity burden by the extended bi-prediction method. A prediction cost based complexity reduction method is consisted of early termination method for the extended bi-prediction and the bi-prediction skipping method. Compare with HM 6.0 references S/W, the average 0.42% of BD-bitrate is decreased by both the extended bi-prediction method and the prediction cost based complexity reduction methods with negligible increasement of the complexity.

• Journal of information and communication convergence engineering
• /
• v.21 no.1
• /
• pp.82-89
• /
• 2023

Various machine-learning models may yield high predictive power for massive time series for time series prediction. However, these models are prone to instability in terms of computational cost because of the high dimensionality of the feature space and nonoptimized hyperparameter settings. Considering the potential risk that model training with a high-dimensional feature set can be time-consuming, we evaluate a feature-importance-based feature selection method to derive a tradeoff between predictive power and computational cost for time series prediction. We used two machine learning techniques for performance evaluation to generate prediction models from a retail sales dataset. First, we ranked the features using impurity- and Local Interpretable Model-agnostic Explanations (LIME) -based feature importance measures in the prediction models. Then, the recursive feature elimination method was applied to eliminate unimportant features sequentially. Consequently, we obtained a subset of features that could lead to reduced model training time while preserving acceptable model performance.

• Journal of the Korea Institute of Building Construction
• /
• v.4 no.1
• /
• pp.105-110
• /
• 2004

Resource procurement is an important management area because cost of resource covers 40％ of total construction project cost and resource delivery has direct relationship with project performance. Integration of cost provides various potentials for effective and efficient project control. This study investigates the usefulness of time in resource procurement management focused on materials. These days, construction projects have characterized manufacture because of industrialization and component. Therefore, application of systematic resource planning has been requested in the construction. There are many companies conducting procurement of resource on the web by applying MRP, ERP etc. in the construction. However, in applying them in the construction yet, there is obstruction. MRP has the character doing its function under accurate cost prediction of resource. But prediction of resource is difficult in industry mechanism of the construction. If accurate cost prediction of resource is possible in the construction, it will be expected to reduce cost of procurement of resource substantially by applying successful resource planning model in the manufacture. On the basis of recent current, the purpose of study is to present procurement of resource system that period observance of construction and minimization of stock is possible by reflecting accurate lead-time to apply proactive thought to be able to cope with alteration of construction schedule efficiently in analyzing resource planning of the construction site.

• Proceedings of the KSR Conference
• /
• 2007.11a
• /
• pp.777-783
• /
• 2007

Life cycle costing is one of the most effective cost approaches when we choose a solution from series of alternative so the least long-term cost ownership is achieved. Life cycle costing in railway industry has been focused on the prediction of investment for railway vehicles. But in today, the life cycle cost, LCC, prediction on the aspect of operation and maintenance cost through whole life cycle is highly necessary. In this paper, we present a strategy for the development of life cycle cost estimation software on the aspect of maintenance strategies of railway vehicle. For this purpose, we suggested a structure of LCC software based on the UNIFE LCC model. And we developed a pilot version of software to evaluate the LCC model that we suggested for railway vehicle. We performed LCC analysis on the brake module of metro vehicle in case study and concluded that the software and model developed in this research could enough to support engineers in choosing better cost effective solutions from many alternatives.

• Journal of Intelligence and Information Systems
• /
• v.25 no.1
• /
• pp.127-137
• /
• 2019

Recidivism prediction has been a subject of constant research by experts since the early 1970s. But it has become more important as committed crimes by recidivist steadily increase. Especially, in the 1990s, after the US and Canada adopted the 'Recidivism Risk Assessment Report' as a decisive criterion during trial and parole screening, research on recidivism prediction became more active. And in the same period, empirical studies on 'Recidivism Factors' were started even at Korea. Even though most recidivism prediction studies have so far focused on factors of recidivism or the accuracy of recidivism prediction, it is important to minimize the prediction misclassification cost, because recidivism prediction has an asymmetric error cost structure. In general, the cost of misrecognizing people who do not cause recidivism to cause recidivism is lower than the cost of incorrectly classifying people who would cause recidivism. Because the former increases only the additional monitoring costs, while the latter increases the amount of social, and economic costs. Therefore, in this paper, we propose an XGBoost(eXtream Gradient Boosting; XGB) based recidivism prediction model considering asymmetric error cost. In the first step of the model, XGB, being recognized as high performance ensemble method in the field of data mining, was applied. And the results of XGB were compared with various prediction models such as LOGIT(logistic regression analysis), DT(decision trees), ANN(artificial neural networks), and SVM(support vector machines). In the next step, the threshold is optimized to minimize the total misclassification cost, which is the weighted average of FNE(False Negative Error) and FPE(False Positive Error). To verify the usefulness of the model, the model was applied to a real recidivism prediction dataset. As a result, it was confirmed that the XGB model not only showed better prediction accuracy than other prediction models but also reduced the cost of misclassification most effectively.

• Nuclear Engineering and Technology
• /
• v.52 no.7
• /
• pp.1429-1435
• /
• 2020

Applying an accurate parametric prediction model to identify abnormal or false pressurizer water levels (PWLs) is critical to the safe operation of marine pressurized water reactors (PWRs). Recently, deep-learning-based models have proved to be a powerful feature extractor to perform high-accuracy prediction. However, the effectiveness of models still suffers from two issues in PWL prediction: the correlations shifting over time between PWL and other feature parameters, and the example imbalance between fluctuation examples (minority) and stable examples (majority). To address these problems, we propose a cost-sensitive mechanism to facilitate the model to learn the feature representation of later examples and fluctuation examples. By weighting the standard mean square error loss with a cost-sensitive factor, we develop a Cost-Sensitive Long Short-Term Memory (CSLSTM) model to predict the PWL of PWRs. The overall performance of the CSLSTM is assessed by a variety of evaluation metrics with the experimental data collected from a marine PWR simulator. The comparisons with the Long Short-Term Memory (LSTM) model and the Support Vector Regression (SVR) model demonstrate the effectiveness of the CSLSTM.

• Journal of the Korea Institute of Building Construction
• /
• v.21 no.5
• /
• pp.495-506
• /
• 2021

A Conceptual cost estimate, which is computed in the preliminary step of a project, is important for decision-making by a contractor in terms of the project budget, economic feasibility and validity analysis, and alternative comparisons. Therefore, a high error rate of a prediction model for a conceptual cost estimate can lead to various problems including excessive project expenditures and a delayed break-even point. this study proposed optimal impact factors by configuring quantitative impact factors computable in a preliminary step in various cases(combinations of impact factors). subsequently, the accuracy of different cases was comparatively analyzed by using the cases as input values of a prediction model using regression analysis. when the optimal combination of impact factors proposed in this study and other combination of impact factors were applied to the prediction model, the regression analysis-based prediction model exhibited 0.2-4.7% improvements in accuracy, respectively. the optimal combination of impact factors proposed in this study improved the accuracy of the prediction model of a conceptual cost estimate by removing unnecessary impact factor.

• Korean Journal of Construction Engineering and Management
• /
• v.21 no.3
• /
• pp.48-56
• /
• 2020

Prediction of construction cost in the planning stage that provides basic information for feasibility study, budgeting, and planning is an important factor for successful project execution. In this study, a prediction model was developed for the purpose of improving the accuracy of estimating the construction cost of Hanok in the planning stage. The cost of this model is estimated by two methods. First, the cost of wood work, which accounts for the largest portion of the total construction cost, is estimated by calculating the approximate quantity under various conditions (structure type, roof type, plane type, etc.). Second, the cost of the rest work sections except the wood work is estimated by using the unit cost model. The predictive model was verified by two case projects, and the error rate of total construction cost was -4%(case 1) and -6%(case 2). These results showed an error rate in the range that can be applied to practice in the planning stage.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1996.04a
• /
• pp.147-150
• /
• 1996

Human posture prediction and motion simulation methods try to solve inverse kinematic problems using the optimization technique based on the concept of minimum principle. It is very important to select a cost function which relfects the human posture acurately. In this study, lifting postures were predicted using the five biomechanical cost functions and compared with real human postures in order to evaluate the predictivities of the cost functions. The result showed that all the biomechanical cost functions used in this study could not predict lifting postures accurately. The cost function which minimizes the sum of joint moments showed the smallest mean prediction error, while the one which minimizes the MUR showed statistically better performance.

• Journal of the Korean Society for Railway
• /
• v.8 no.3
• /
• pp.260-266
• /
• 2005

The deterioration of a shunting locomotive was characterized for the lifetime assessment. The locomotive has been used for shunting works in steel making processes, and in this investigation, various types of technical evaluation methods for the locomotive parts were employed to assess the current deterioration status and to provide important clue for lifetime prediction. Unlike other rolling stocks in railway applications, the diesel shunting locomotive is composed of major components such as diesel engine, transmission, gear box, brake system, electronic devices, etc., which cover more than 70 percent of the total price of the locomotive. Therefore, in this paper, each part of major components in the diesel locomotive was analyzed in terms of the degree of deterioration. The lift-cycle-cost (LCC) analysis was performed based on the maintenance and repair history as compared with economical cost to provide the cost-effective prediction, i.e., to assess either repair for reuse or putting the locomotive out of service based on cost-effective calculation.