• Geomechanics and Engineering
• /
• v.37 no.3
• /
• pp.253-262
• /
• 2024

Dynamic properties are pivotal in soil analysis, yet their experimental determination is hampered by complex methodologies and the need for costly equipment. This study aims to predict dynamic soil properties using static properties that are relatively easier to obtain, employing machine learning techniques. The static properties considered include soil cohesion, friction angle, water content, specific gravity, and compressional strength. In contrast, the dynamic properties of interest are the velocities of compressional and shear waves. Data for this study are sourced from 26 boreholes, as detailed in a geotechnical investigation report database, comprising a total of 130 data points. An importance analysis, grounded in the random forest algorithm, is conducted to evaluate the significance of each dynamic property. This analysis informs the prediction of dynamic properties, prioritizing those static properties identified as most influential. The efficacy of these predictions is quantified using the coefficient of determination, which indicated exceptionally high reliability, with values reaching 0.99 in both training and testing phases when all input properties are considered. The conventional method is used for predicting dynamic properties through Standard Penetration Test (SPT) and compared the outcomes with this technique. The error ratio has decreased by approximately 0.95, thereby validating its reliability. This research marks a significant advancement in the indirect estimation of the relationship between static and dynamic soil properties through the application of machine learning techniques.

• The Bulletin of The Korean Astronomical Society
• /
• v.40 no.2
• /
• pp.31.3-32
• /
• 2015

We present the physical properties of star forming regions in IC 10 obtained from Korea VLBI Network (KVN) 22GHz, the Submillimeter Array (SMA) CO, Very Large Array (VLA) HI 21cm, optical (U, B, V and H-alpha), and Spitzer infrared observations. IC 10 is a nearby (~0.7Mpc) irregular blue compact dwarf (BCD) galaxy which is likely to be experiencing an intense and recent burst of star formation. This nearby infant system showing high star formation rate but low metallicity (<20% of that of the Sun) provides critical environment of interstellar medium (ISM) under which current galactic star formation models are challenged. To make quantitative analysis of the ISM in the galaxy, we apply 2D cross-correlation technique to the multi-wavelength data for the first time. By cross-correlating different tracers of star formation, dust and gas phases in IC 10 in a two dimensional way, we discuss the gas properties and star formation history of the galaxy.

• Computers and Concrete
• /
• v.33 no.3
• /
• pp.285-299
• /
• 2024

Recently, many engineering computations have realized their digital transformation to Machine Learning (ML)-based systems. Predicting the behavior of a structure, which is mainly computed with structural analysis software, is an essential step before construction for efficient structural analysis. Especially in the seismic-based design procedure of the structures, predicting the lateral load capacity of reinforced concrete (RC) columns is a vital factor. In this study, a novel ML-based model is proposed to predict the maximum lateral load capacity of RC columns under varying axial loads or cyclic loadings. The proposed model is generated with a Deep Neural Network (DNN) and compared with traditional ML techniques as well as a popular commercial structural analysis software. In the design and test phases of the proposed model, 319 columns with rectangular and square cross-sections are incorporated. In this study, 33 parameters are used to predict the maximum lateral load capacity of each RC column. While some traditional ML techniques perform better prediction than the compared commercial software, the proposed DNN model provides the best prediction results within the analysis. The experimental results reveal the fact that the performance of the proposed DNN model can definitely be used for other engineering purposes as well.

• IMMUNE NETWORK
• /
• v.23 no.1
• /
• pp.2.1-2.19
• /
• 2023

Immune diversification helps protect the host against a myriad of pathogens. CD8⁺ T cells are essential adaptive immune cells that inhibit the spread of pathogens by inducing apoptosis in infected host cells, ultimately ensuring complete elimination of infectious pathogens and suppressing disease development. Accordingly, numerous studies have been conducted to elucidate the mechanisms underlying CD8⁺ T cell activation, proliferation, and differentiation into effector and memory cells, and to identify various intrinsic and extrinsic factors regulating these processes. The current knowledge accumulated through these studies has led to a huge breakthrough in understanding the existence of heterogeneity in CD8⁺ T cell populations during immune response and the principles underlying this heterogeneity. As the heterogeneity in effector/memory phases has been extensively reviewed elsewhere, in the current review, we focus on CD8⁺ T cells in a "naive" state, introducing recent studies dealing with the heterogeneity of naive CD8⁺ T cells and discussing the factors that contribute to such heterogeneity. We also discuss how this heterogeneity contributes to establishing the immense complexity of antigen-specific CD8⁺ T cell response.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.6 s.38
• /
• pp.279-286
• /
• 2005

The objective of this study is to suggest a credit scoring model of a capital company's customers by integration of multiple classifiers using genetic algorithm. For this purpose , an integrated model is derived in two phases. In first phase, three types of classifiers MLP (Multi-Layered Perceptron), RBF (Radial Basis Function) and linear models - are trained, in which each type has three ones respectively so htat we have nine classifiers totally. In second phase, genetic algorithm is applied twice for integration of classifiers. That is, after htree models are derived from each group, a final one is from these three, In result, our suggested model shows a superior accuracy to any single ones.

• The Bulletin of The Korean Astronomical Society
• /
• v.41 no.1
• /
• pp.43.1-43.1
• /
• 2016

With the Korean VLBI Network (KVN), both single dish and VLBI monitoring observations of H2O and SiO masers were performed toward the semi-regular variable star R Crateris. In the case of 11 VLBI monitoring observations from Jan. 5, 2014 to Jan. 7, 2016, successful superposed maps of H2O and SiO masers were obtained at 7 epochs by adopting the Source Frequency Phase Referencing (SFPR) method. These results enable us to investigate the development of outflow and asymmetric motions from SiO maser to H2O maser regions according to stellar pulsation which are closely related with a mass-loss process. Single dish monitoring observations of H2O and SiO masers were also carried out from 2009 June to 2016 Feb. Intensity variations between H2O and SiO masers were investigated according to stellar optical phases together with peak velocity variations with respect to the stellar velocity. We will compare the VLBI results among different maser transitions with those of single dish.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.8 no.3
• /
• pp.9-14
• /
• 2008

In this paper, we propose the channel sounding scheme which is made for ideal communication between some application as well as the short distance of high speed data transmission in MIMO-OFDM system for Wireless PAN. This method is able to perceive the duration of the impulse response through the delaying of power delay profile, modeled a power delay profile which has an attenuate characteristic, and obtained the coefficient of channel response by ML (maximum likelihood). Through the amplitudes, phases and delays associated with each multipath component which were acquired from this channel sounding scheme, we can describe the wave propagation characteristics of channels between the transmitter and receiver so that the receiver could enhance not only the reliability but also the ability of communication link.

• The Bulletin of The Korean Astronomical Society
• /
• v.40 no.2
• /
• pp.58.4-59
• /
• 2015

We have performed single dish and VLBI monitoring observations of $H_2O$ and SiO masers toward the semi-regular variable star R Crateris using the Korean VLBI Network(KVN) 4 band receiving system. In the case of VLBI observations at 3 epochs, successful superposed maps of $H_2O$ and SiO masers were obtained on 2015 May by adopting the Source Frequency Phase Referencing(SFPR) method. These results enable us to investigate the development of outflow and asymmetric motions from SiO maser to $H_2O$ maser regions according to stellar pulsation which are closely related with a mass-loss process. Single dish monitoring observations were carried out from 2009 June to 2015 May. Intensity variations between $H_2O$ and SiO masers were investigated according to stellar phases together with peak velocity variations. We will compare the VLBI results with those of single dish.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.6
• /
• pp.764-770
• /
• 2004

This paper presents a hybrid data mining mechanism to extract expert knowledge from historical data and extend expert systems' reasoning capabilities by using fuzzy neural network (FNN)-based learning & rule extraction algorithm. Our hybrid data mining mechanism is based on association rule extraction mechanism, FNN learning and fuzzy rule extraction algorithm. Most of traditional data mining mechanisms are depended ()n association rule extraction algorithm. However, the basic association rule-based data mining systems has not the learning ability. Therefore, there is a problem to extend the knowledge base adaptively. In addition, sequential patterns of association rules can`t represent the complicate fuzzy logic in real-world. To resolve these problems, we suggest the hybrid data mining mechanism based on association rule-based data mining, FNN learning and fuzzy rule extraction algorithm. Our hybrid data mining mechanism is consisted of four phases. First, we use general association rule mining mechanism to develop an initial rule base. Then, in the second phase, we adopt the FNN learning algorithm to extract the hidden relationships or patterns embedded in the historical data. Third, after the learning of FNN, the fuzzy rule extraction algorithm will be used to extract the implicit knowledge from the FNN. Fourth, we will combine the association rules (initial rule base) and fuzzy rules. Implementation results show that the hybrid data mining mechanism can reflect both association rule-based knowledge extraction and FNN-based knowledge extension.

• Journal of Power Electronics
• /
• v.8 no.4
• /
• pp.354-362
• /
• 2008

This paper presents neural load torque compensation method which is composed of a deadbeat load torque observer and gains compensation by a parameter estimator. As a result, the response of the PMSM (permanent magnet synchronous motor) obtains better precision position control. To reduce the noise effect, the post-filter is implemented by a MA (moving average) process. The parameter compensator with an RLSM (recursive least square method) parameter estimator is adopted to increase the performance of the load torque observer and main controller. The parameter estimator is combined with a high performance neural load torque observer to resolve problems. The neural network is trained in online phases and it is composed by a feed forward recall and error back-propagation training. During normal operation, the input-output response is sampled and the weighting value is trained multi-times by the error back-propagation method at each sample period to accommodate the possible variations in the parameters or load torque. As a result, the proposed control system has a robust and precise system against load torque and parameter variation. Stability and usefulness are verified by computer simulation and experiment.