• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제16권1호
• /
• pp.44-51
• /
• 2016

Recently, artificial intelligence has reached the level of top information technologies that will have significant influence over many aspects of our future lifestyles. In particular, in the fields of machine learning technologies for classification and decision-making, there have been a lot of research efforts for solving estimation and control problems that appear in the various kinds of portfolio management problems via data-driven approaches. Note that these modern data-driven approaches, which try to find solutions to the problems based on relevant empirical data rather than mathematical analyses, are useful particularly in practical application domains. In this paper, we consider some applications of modern data-driven machine learning methods for portfolio management problems. More precisely, we apply a simplified version of the sparse Gaussian process (GP) classification method for classifying users' sensitivity with respect to financial risk, and then present two portfolio management issues in which the GP application results can be useful. Experimental results show that the GP applications work well in handling simulated data sets.

• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2019년도 제60차 하계학술대회논문집 27권2호
• /
• pp.389-390
• /
• 2019

Tensor decomposition, proven to be an efficient data processing method, can be used to provide data-driven services. we propose a novel datadriven mutation strategy for parent individuals selection, namely tensor-based DE with parapatric and cross-generation(TPCDE).

• Advances in nano research
• /
• 제16권3호
• /
• pp.231-249
• /
• 2024

When the amplitude of the vibrations is equivalent to that clearance, the vibrations for small amplitudes will really be significantly nonlinear. Nonlinearities will not be significant for amplitudes that are rather modest. Finally, nonlinearities will become crucial once again for big amplitudes. Therefore, the concrete panel system may experience a big amplitude in this work as a result of the high temperature. Based on the 3D modeling of the shell theory, the current work shows the influences of the von Kármán strain-displacement kinematic nonlinearity on the constitutive laws of the structure. The system's governing Equations in the nonlinear form are solved using Kronecker and Hadamard products, the discretization of Equations on the space domain, and Duffing-type Equations. Thermo-elasticity Equations. are used to represent the system's temperature. The harmonic solution technique for the displacement domain and the multiple-scale approach for the time domain are both covered in the section on solution procedures for solving nonlinear Equations. An effective data-driven solution is often utilized to predict how different systems would behave. The number of hidden layers and the learning rate are two hyperparameters for the network that are often chosen manually when required. Additionally, the data-driven method is offered for addressing the nonlinear vibration issue in order to reduce the computing cost of the current study. The conclusions of the present study may be validated by contrasting them with those of data-driven solutions and other published articles. The findings show that certain physical and geometrical characteristics have a significant effect on the existing concrete panel structure's susceptibility to temperature change and GPL weight fraction. For building construction industries, several useful recommendations for improving the thermo-mechanics' behavior of structural concrete panels are presented.

• ETRI Journal
• /
• 제32권5호
• /
• pp.810-818
• /
• 2010

This paper presents a new framework for integrating untranscribed spoken content into the acoustic training of an automatic speech recognition system. Untranscribed spoken content plays a very important role for under-resourced languages because the production of manually transcribed speech databases still represents a very expensive and time-consuming task. We proposed two new methods as part of the training framework. The first method focuses on combining initial acoustic models using a data-driven metric. The second method proposes an improved acoustic training procedure based on unsupervised transcriptions, in which word endings were modified by broad phonetic classes. The training framework was applied to baseline acoustic models using untranscribed spoken content from parliamentary debates. We include three types of acoustic models in the evaluation: baseline, reference content, and framework content models. The best overall result of 18.02% word error rate was achieved with the third type. This result demonstrates statistically significant improvement over the baseline and reference acoustic models.

• 대한토목학회논문집
• /
• 제26권5C호
• /
• pp.343-350
• /
• 2006

국내 기초구조물에 대한 저항계수 산정 및 하중저항계수설계법(LRFD) 개발의 일환으로 항타강관말뚝에 대한 저항편향계수를 산정하였다. 정재하시험 및 지반조사 자료를 수집하여 국내외의 기준에 따른 극한지지력을 산정한 후 말뚝의 대표 극한지지력을 결정하였다. 자료의 통계분석 결과에 기초하면, Davisson 기준이 말뚝의 극한지지력을 가장 합리적으로 평가하는 것으로 나타났다. 정역학적 지지력 공식과 N치를 이용한 Meyerhof 경험식을 이용하여 설계 극한지지력을 산정하였다. 이들 자료의 비교 분석을 통해 항타강관말뚝의 저항편향계수는 정역학적 지지력공식과 Meyerhof 경험식에 대해 각각 0.98, 1.46으로 산정되었다. 또한 두 가지 지지력 산정방법 중 정역학적 지지력공식의 불확실성이 상대적으로 낮은 것으로 나타났다.

• 한국전산유체공학회:학술대회논문집
• /
• 한국전산유체공학회 1998년도 춘계 학술대회논문집
• /
• pp.15-28
• /
• 1998

As an alternative for solving the incompressible Navier-Stokes equations, we present a vorticity-based integro-differential formulation for vorticity, velocity and pressure variables. One of the most difficult problems encountered in the vorticity-based methods is the introduction of the proper value-value of vorticity or vorticity flux at the solid surface. A practical computational technique toward solving this problem is presented in connection with the coupling between the vorticity and the pressure boundary conditions. Numerical schemes based on an iterative procedure are employed to solve the governing equations with the boundary conditions for the three variables. A finite volume method is implemented to integrate the vorticity transport equation with the dynamic vorticity boundary condition . The velocity field is obtained by using the Biot-Savart integral derived from the mathematical vector identity. Green's scalar identity is used to solve the total pressure in an integral approach similar to the surface panel methods which have been well-established for potential flow analysis. The calculated results with the present mettled for two test problems are compared with data from the literature in order for its validation. The first test problem is one for the two-dimensional square cavity flow driven by shear on the top lid. Two cases are considered here: (i) one driven both by the specified non-uniform shear on the top lid and by the specified body forces acting through the cavity region, for which we find the exact solution, and (ii) one of the classical type (i.e., driven only by uniform shear). Secondly, the present mettled is applied to deal with the early development of the flow around an impulsively started circular cylinder.

• Smart Structures and Systems
• /
• 제24권4호
• /
• pp.507-524
• /
• 2019

A Bayesian dynamic linear model (BDLM) is presented for a data-driven analysis for response prediction and load effect separation of a revolving auditorium structure, where the main loads are self-weight and dead loads, temperature load, and audience load. Analyses are carried out based on the long-term monitoring data for static strains on several key members of the structure. Three improvements are introduced to the ordinary regression BDLM, which are a classificatory regression term to address the temporary audience load effect, improved inference for the variance of observation noise to be updated continuously, and component discount factors for effective load effect separation. The effects of those improvements are evaluated regarding the root mean square errors, standard deviations, and 95% confidence intervals of the predictions. Bayes factors are used for evaluating the probability distributions of the predictions, which are essential to structural condition assessments, such as outlier identification and reliability analysis. The performance of the present BDLM has been successfully verified based on the simulated data and the real data obtained from the structural health monitoring system installed on the revolving structure.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제32권5호
• /
• pp.396-405
• /
• 2005

기존의 주제 관련 연구들은 문서에 자주 등장하는 용어를 주제로 간주하는 등 문서에서 다루는 주제에 대한 정의가 모호하다. 또한 문서를 구성하는 기본 단위인 문장의 주제가 문서 요약 및 정보 추출 등의 연구 분야에 중요하게 활용될 수 있음에도 불구하고, 이에 대한 고려 없이 문서 전체의 주제를 추출하고 할당하는 연구가 대부분이다. 따라서 본 논문에서는 문장 단위의 주제 처리에 대한 기본 연구로서, 백과사전 영역에서 효과적인 중심어주도패턴에 기반한 문장주제 할당 기법을 제안하였다. 두산동아 백과사전 인물분야 2,381문서를 대상으로 성능을 분석해본 결과, 제안된 기법이 비교기준보다 향상된 성능을 보였으며, 특히 제안된 네 가지 중심어주도 패턴 중 술어를 기반으로 구성된 중심어주도패턴 유형 4가 학습집합에 대하여 평균 $98.96\%$, 실험집합에 대하여 $88.57\%$의 성능(F-score)으로 주제할당에 가장 효과적임을 알 수 있었다.

• 품질경영학회지
• /
• 제51권4호
• /
• pp.515-530
• /
• 2023

Purpose : This research aims to develop a data-driven inspection policy for radio stations utilizing the KS Q ISO 2859-1 sampling method, addressing potential regulatory relaxations and impending management challenges. Methods : Using radio station inspection big data from the past six years, we established a simulation model to evaluate the current policy. A new inspection sampling policy framework was designed based on the KS Q ISO 2859-1 method. The study compares the performance of the current and proposed inspection systems, offering insights for an improved inspection strategy. Results : This study introduced a simulation model for inspection system based on the KS Q ISO 2859-1 sampling method. Through various experimental designs, key performance indicators such as non-detection rate and sample proportion were derived, providing foundational data for the new inspection policy. Conclusion : Using big data from radio station inspections, we evaluated current inspection systems and quantitatively compared a new system across diverse scenarios. Our simulation model effectively verified the feasibility and efficiency of the proposed framework. For practical implementation, essential factors such as lot size, inspection cycle, and AQL standards need precise definition and consideration. Enhancing radio station inspections requires a policy-driven approach that factors in socio-economic impacts and solicits feedback from industry participants. Future study should also explore various perspectives related to legislative, institutional, and operational aspects of inspection organizations.

• Wind and Structures
• /
• 제5권5호
• /
• pp.441-462
• /
• 2002

This paper presents a practical numerical method to determine both the spatial and temporal distribution of driving rain on buildings. It is based on an existing numerical simulation technique and uses the building geometry and climatic data at the building site as input. The method is applied to determine the 3D spatial and temporal distribution of wind-driven rain on the facade a low-rise building of complex geometry. Distinct wetting patterns are found. The important causes giving rise to these particular patterns are identified : (1) sweeping of raindrops towards vertical building edges, (2) sweeping of raindrops towards top edges, (3) shelter effect by various roof overhang configurations. The comparison of the numerical results with full-scale measurements in both space and time for a number of on site recorded rain events shows the numerical method to yield accurate results.