• Journal of Information Processing Systems
• /
• v.16 no.2
• /
• pp.343-359
• /
• 2020

In this paper, a new algorithm is proposed for three-dimensional (3D) shape recognition using local features of model views and its sparse representation. The algorithm starts with the normalization of 3D models and the extraction of 2D views from uniformly distributed viewpoints. Consequently, the 2D views are stacked over each other to from view cubes. The algorithm employs the descriptors of 3D local features in the view cubes after applying Gabor filters in various directions as the initial features for 3D shape recognition. In the training stage, we store some 3D local features to build the prototype dictionary of local features. To extract an intermediate feature vector, we measure the similarity between the local descriptors of a shape model and the local features of the prototype dictionary. We represent the intermediate feature vectors of 3D models in the sparse domain to obtain the final descriptors of the models. Finally, support vector machine classifiers are used to recognize the 3D models. Experimental results using the Princeton Shape Benchmark database showed the average recognition rate of 89.7% using 20 views. We compared the proposed approach with state-of-the-art approaches and the results showed the effectiveness of the proposed algorithm.

• Advances in concrete construction
• /
• v.13 no.1
• /
• pp.11-23
• /
• 2022

Conventional concrete needs some improvement in the mechanical properties, which can be obtained by different admixtures. However, making concrete samples costume always time and money. In this paper, different types of hybrid algorithms are applied to develop predictive models for forecasting compressive strength (CS) of concretes containing metakaolin (MK) and fly ash (FA). In this regard, three different algorithms have been used, namely multilayer perceptron (MLP), radial basis function (RBF), and support vector machine (SVR), to predict CS of concretes by considering most influencers input variables. These algorithms integrated with the grey wolf optimization (GWO) algorithm to increase the model's accuracy in predicting (GWMLP, GWRBF, and GWSVR). The proposed MLP models were implemented and evaluated in three different layers, wherein each layer, GWO, fitted the best neuron number of the hidden layer. Correspondingly, the key parameters of the SVR model are identified using the GWO method. Also, the optimization algorithm determines the hidden neurons' number and the spread value to set the RBF structure. The results show that the developed models all provide accurate predictions of the CS of concrete incorporating MK and FA with R² larger than 0.9972 and 0.9976 in the learning and testing stage, respectively. Regarding GWMLP models, the GWMLP1 model outperforms other GWMLP networks. All in all, GWSVR has the worst performance with the lowest indices, while the highest score belongs to GWRBF.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.25 no.1
• /
• pp.1-10
• /
• 2001

As computer power is increased, refined finite element models are employed for structural analysis. However, it is difficult and expensive to use refined models in the design stage. The refined models especially cause problems in the preliminary design where the design is frequently changed. Therefore, simplified models are needed. The simplification process is regarded as an empirical technique. Simplified and equivalent finite element model of a structure has been studied and used in the preliminary design. A general approach to establish the simplified and equivalent model is presented. The generated simple model has satisfactory correlation with the corresponding refined finite element model. An optimization method, the Goal Programming algorithm is used to make the simple model. The simplified model is used for the design change and the changed design is recovered onto the original design. The presented method was verified with three examples.

• IE interfaces
• /
• v.18 no.3
• /
• pp.268-276
• /
• 2005

In recent years, the money Laundering crimes are increasing by means of foreign exchange transactions. Our study proposes four scoring models to provide early warning of the laundering in foreign exchange transactions for both inward and outward remittances: logistic regression model, decision tree, neural network, and ensemble model which combines the three models. In terms of accuracy of test data, decision tree model is selected for the inward remittance and an ensemble model for the outward remittance. From our study results, the accumulated number of transaction turns out to be the most important predictor variable. The proposed scoring models deal with the transaction level and is expected to help the bank teller to detect the laundering related transactions in the early stage.

• Journal of The Korean Association For Science Education
• /
• v.28 no.7
• /
• pp.759-767
• /
• 2008

This article arose from the previous studies, which suggested a synthetic list for the nature of science (NOS), discussed the relationship between the NOS and scientific inquiry and the development of the NOS in the context of scientific inquiry. In this article, for teaching scientific inquiry through the NOS, I proposed three teaching models - reflection, interaction, and the direct model -. Within these teaching models, understanding the NOS is viewed as a prerequisite condition for the improved performance of scientific inquiry. In the reflection model, the NOS is embedded and reflected in scientific inquiry without explicit introduction or direct explanation of the NOS. In the interaction model, concrete interaction between scientific inquiry and the NOS is encouraged during the process of scientific inquiry. In the direct model, subsequent to directly comprehending the NOS at the first stage of activity, students conduct scientific inquiry based on their understanding of the NOS. The intention of this present article is to facilitate the use of these models to develop teaching materials for more authentic scientific inquiry.

• Journal of Korea Water Resources Association
• /
• v.51 no.spc
• /
• pp.1079-1089
• /
• 2018

The majority of existing studies for quantifying uncertainties in climate change impact assessments suggest only the uncertainties of each stage, and not the total uncertainty and its propagation in the whole procedure. Therefore, this study has proposed a new method, the Uncertainty Delta Method (UDM), which can quantify uncertainties using the variances of projections (as the UDM is derived from the first-order Taylor series expansion), to allow for a comprehensive quantification of uncertainty at each stage and also to provide the levels of uncertainty propagation, as follows: total uncertainty, the level of uncertainty increase at each stage, and the percentage of uncertainty at each stage. For quantifying uncertainties at each stage as well as the total uncertainty, all the stages - two emission scenarios (ES), three Global Climate Models (GCMs), two downscaling techniques, and two hydrological models - of the climate change assessment for water resources are conducted. The total uncertainty took 5.45, and the ESs had the largest uncertainty (4.45). Additionally, uncertainties are propagated stage by stage because of their gradual increase: 5.45 in total uncertainty consisted of 4.45 in emission scenarios, 0.45 in climate models, 0.27 in downscaling techniques, and 0.28 in hydrological models. These results indicate the projection of future water resources can be very different depending on which emission scenarios are selected. Moreover, using Fractional Uncertainty Method (FUM) by Hawkins and Sutton (2009), the major uncertainty contributor (emission scenario: FUM uncertainty 0.52) matched with the results of UDM. Therefore, the UDM proposed by this study can support comprehension and appropriate analysis of the uncertainty surrounding the climate change impact assessment, and make possible a better understanding of the water resources projection for future climate change.

• Journal of Korea Water Resources Association
• /
• v.54 no.spc1
• /
• pp.1037-1051
• /
• 2021

Biochemical oxygen demand (BOD) concentration, one of important water quality indicators, is treated as the measuring item for the ecological chapter in lakes and rivers. This investigation employed novel two-stage hybrid paradigm (i.e., wavelet-based gated recurrent unit, wavelet-based generalized regression neural networks, and wavelet-based random forests) to predict BOD concentration in the Dosan and Hwangji stations, South Korea. These models were assessed with the corresponding independent models (i.e., gated recurrent unit, generalized regression neural networks, and random forests). Diverse water quality and quantity indicators were implemented for developing independent and two-stage hybrid models based on several input combinations (i.e., Divisions 1-5). The addressed models were evaluated using three statistical indices including the root mean square error (RMSE), Nash-Sutcliffe efficiency (NSE), and correlation coefficient (CC). It can be found from results that the two-stage hybrid models cannot always enhance the predictive precision of independent models confidently. Results showed that the DWT-RF5 (RMSE = 0.108 mg/L) model provided more accurate prediction of BOD concentration compared to other optimal models in Dosan station, and the DWT-GRNN4 (RMSE = 0.132 mg/L) model was the best for predicting BOD concentration in Hwangji station, South Korea.

• Journal of Crop Science and Biotechnology
• /
• v.10 no.1
• /
• pp.33-38
• /
• 2007

Nitrogen management at the panicle initiation stage(PI) should be fine-tuned for securing a concurrent high yield and high quality rice production. For calibration and testing of the recommendation models of N topdressing rates at PI for target grain yield and protein content of rice, three split-split-plot design experiments including five rice cultivars and various N rates were conducted at the experimental farm of Seoul National University, Korea from 2003 to 2005. Data from the first two years of experiments were used to calibrate models to predict grain yield and milled-rice protein content using shoot fresh weight(FW), chlorophyll meter value(SPAD), and the N topdressing rate(Npi) at PI by stepwise multiple regression. The calibrated models explained 85 and 87% of the variation in grain yield and protein content, respectively. The calibrated models were used to recommend Npi for the target protein content of 6.8%, with FW and SPAD measured for each plot in 2005. The recommended N rate treatment was characterized by an average protein content of 6.74%(similar to the target protein content), reduced the coefficient of variation in protein content to 2.5%(compared to 4.6% of the fixed rate treatment), and increased grain yield. In the recommended N rate treatments for the target protein content of 6.8%, grain yield was highly dependent on FW and SPAD at PI. In conclusion, the models for N topdressing rate recommendation at PI were successful under present experimental conditions. However, additional testing under more variable environmental conditions should be performed before universal application of such models.

• Journal of computational fluids engineering
• /
• v.21 no.3
• /
• pp.39-47
• /
• 2016

We conducted a multi-stage optimization to secure the desired performance of a centrifugal fan for home appliance in an early stage of product development. In optimization, the static pressure at the outlet of the fan is chosen as an objective function that is to be maximized, providing the required flow rate at the operating point of the fan. The optimization procedure begins with parameters for an initial baseline fan design. The baseline design is optimized by using a commercial optimization package. Accordingly, the corresponding blade models with a set of geometrical parameters are generated. Flow through a fan is simulated by solving the Reynolds-averaged Navier-Stokes equations. A multi-stage optimization scheme is employed to determine the family of optimum values for the parameters, leading to the pressure increase at the outlet of the fan. To validate the numerically obtained optimal design parameters, we fabricated the three types of fans using rapid prototyping and assessed the performance using a fan tester. Experimental results show that the design parameters at each stage satisfy the goal of optimization. The multi-stage optimization process turned out to be a useful tool in the development of a centrifugal fan.

• Journal of KIISE
• /
• v.41 no.9
• /
• pp.642-651
• /
• 2014

This paper proposes three probabilistic models for syllable-based Korean morphological analysis, and presents the performance of proposed probabilistic models. Probabilities for the models are acquired from POS-tagged corpus. The result of 10-fold cross-validation experiments shows that 98.3% answer inclusion rate is achieved when trained with Sejong POS-tagged corpus of 10 million eojeols. In our models, POS tags are assigned to each syllable before spelling recovery and morpheme generation, which enables more efficient morphological analysis than the previous probabilistic models where spelling recovery is performed at the first stage. This efficiency gains the speed-up of morphological analysis. Experiments show that morphological analysis is performed at the rate of 147K eojeols per second, which is almost 174 times faster than the previous probabilistic models for Korean morphology.