• Geomechanics and Engineering
• /
• 제33권3호
• /
• pp.231-242
• /
• 2023

This paper represents a comparative study in which Gene Expression Programming (GEP), Group Method of Data Handling (GMDH), and multiple linear regressions (MLR) were utilized to derive new equations for the prediction of time-dependent bearing capacity of pile foundations driven in cohesive soil, technically called pile set-up. This term means that many piles which are installed in cohesive soil experience a noticeable increase in bearing capacity after a specific time. Results of researches indicate that side resistance encounters more increase than toe resistance. The main reason leading to pile setup in saturated soil has been found to be the dissipation of excess pore water pressure generated in the process of pile installation, while in unsaturated conditions aging is the major justification. In this study, a comprehensive dataset containing information about 169 test piles was obtained from literature reviews used to develop the models. to prepare the data for further developments using intelligent algorithms, Data mining techniques were performed as a fundamental stage of the study. To verify the models, the data were randomly divided into training and testing datasets. The most striking difference between this study and the previous researches is that the dataset used in this study includes different piles driven in soil with varied geotechnical characterization; therefore, the proposed equations are more generalizable. According to the evaluation criteria, GEP was found to be the most effective method to predict set-up among the other approaches developed earlier for the pertinent research.

• 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.

• 한국통신학회논문지
• /
• 제35권6B호
• /
• pp.867-874
• /
• 2010

무선 센서 네트워크에서는 멀티 홉 전송동안 높은 패킷 손실률이 발생하기 때문에 신뢰성 있는 데이터 전송방안이 필요하다. 특히, 화재 경보 시스템과 같은 event-driven 데이터가 발생하는 경우, 신뢰성 있는 데이터 전송을 위해서는 손실된 패킷을 복원하기 위한 재전송 방안이 제공되어야 한다. 손실된 데이터의 재전송은 데이터를 캐쉬하고 있는 노드에 요청이 되기 때문에, 데이터를 캐쉬하고 있는 노드는 모든 데이터 패킷을 버퍼에서 유지하고 있어야 한다. 그러나 일반적으로 센서 네트워크의 노드들은 제한된 자원을 가지 있다. 따라서 신뢰성 있는 데이터 전송을 위해서는 손실 패킷의 재전송 방안과 노드의 버퍼 관리 기법이 함께 제공되어야 한다. 본 논문에서는 전송 데이터의 신뢰도에 따라 데이터의 캐쉬지점을 결정하여 손실된 데이터를 복원하는 손실 복원 기법을 사용하는 데이터 전송에서의 효율적인 버퍼 관리기법을 제안하고, 컴퓨터 시뮬레이션을 통하여 제안하는 방안의 우수성을 검증하였다.

• 공학교육연구
• /
• 제22권4호
• /
• pp.61-70
• /
• 2019

As product development process becomes complex, it becomes more important for engineering students to experience collaborative product development. Especially the collaboration experience based on Product Data Management (PDM) systems is useful, since participants are likely to use the same environment for their professional product development. However, instructors have difficulties to evaluate contribution of each participant to their projects during the practices, since it is hard to trace personal activities for collaborative design processes. To solve this problem, this study suggests a data-driven objective method that analyses product data accumulated in PDM databases to evaluate numerically calculated contributions of participants to their class projects. As a result, the quantitative measures provided by the data-driven analysis with qualitative measures for project results can improve the fairness and quality of evaluation of contributions of participants to collaborative projects. This study implemented the proposed evaluation method with an information system and discussed the result of the application of the system to product development practices.

• 제어로봇시스템학회논문지
• /
• 제17권11호
• /
• pp.1095-1100
• /
• 2011

In this paper, PHM (Prognostics and Health Management) techniques are briefly outlined. Prognostics, being a central step within the PHM, is explained in more detail, stating that there are three approaches - experience based, data-driven and model based approaches. Representative articles in the field of prognostics are also given in terms of the type of faults. Model based method is illustrated by introducing a case study that was conducted to the crack growth of the gear plate in UH-60A helicopter. The paper also addresses the comparison of the OBM (Overall Bayesian Method), which was developed by the authors with the PF (Particle Filtering) method, which draws great attention recently in prognostics, through the study on a simple crack growth problem. Their performances are examined by evaluating the metrics introduced by PHM society.

• Journal of the Korean Data and Information Science Society
• /
• 제9권2호
• /
• pp.179-188
• /
• 1998

커널형 회귀함수의 추정법 중에서 국소 다항회귀 추정법이 가장 우수한 것으로 알려져 있다. 국소다항회귀 추정법에서도 다른 종류의 커널추정량과 마찬가지로 평활량이 중요한 역할을 한다. 특히 회귀함수가 복잡한 구조를 가질 때 변수평활량(variable band-width)을 사용하는 것이 타당할 것이다. 본 연구에서는 완전자료기저(fully automatic, fully data-driven) 변수평활량 선택법을 제안한다. 이 선택법은 편향과 분산의 예비추정에 필요한 평활량을 교차타당성 방법으로 선택하여 MSE를 추정하고 그 값을 최소화하는 평활량을 택하는 것이다. 제안된 방법의 우수성을 모의실험을 통하여 확인하였다. 그리고 제안된 방법은 자료점이 성긴(sparse)부분에서 생길 수 있는 문제점 즉 X'X의 비정칙성(non-singularity)을 해결할 수 있는 방법이라는 데에도 큰 의미가 있다.

• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2022년도 제66차 하계학술대회논문집 30권2호
• /
• pp.585-586
• /
• 2022

사회현안을 해결하기 위한 데이터 분석을 위해서는 많은 양의 데이터 수집과 데이터 분석에 활용할 수 있도록 데이터 전처리가 필요하다. 많은 양의 데이터를 수집 및 처리를 위해 데이터 수집, 데이터 저장, 활용 시스템이 기능적으로 분리하여 시스템을 구성하고, 이에 따른 시스템간의 데이터 상호 연계가 필요하게 된다. 또한 외부 네트워크에 구성되어 있는 시스템간의 데이터 연계나, OpenAPI와 같이 다양한 데이터 서비스에서도 적용이 가능할 수 있도록 확장성과 유연성을 고려할 필요가 있다. 본 논문에서는 부산 지역현안 해결을 위한 시스템 구성에 있어, 확장성을 고려한 데이터 수집 시스템간의 효율적인 데이터 연계 방법을 제안하고자 한다.

• 한국해안해양공학회지
• /
• 제6권1호
• /
• pp.1-11
• /
• 1994

곡선형격자 삼차원 수치모델이 소개되며 바람에 의한 물의 순환을 계산하기 위하여 얕은 호수에서 적용되었다. 수치모델의 결과가 실측자료와 비교되었으며, 바람이 점차 증가할 때 물의 성층에 의한 효과가 좋은 계산 결과를 얻기 위하여 결정적임을 알 수 있었다. 기상자료가 불충분할 때 소위 Inverse Method가 물 표면에서 열흐름을 추정하기 위하여 사용되었다.

• 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.

• Wind and Structures
• /
• 제9권2호
• /
• pp.159-178
• /
• 2006

For the slender and flexible cable supported bridges, identification of all the flutter derivatives for the vertical, lateral and torsional motions is essential for its stability investigation. In all, eighteen flutter derivatives may have to be considered, the identification of which using a three degree-of-freedom elastic suspension system has been a challenging task. In this paper, a system identification technique, known as covariance-driven stochastic subspace identification (COV-SSI) technique, has been utilized to extract the flutter derivatives for a typical bridge deck. This method identifies the stochastic state-space model from the covariances of the output-only (stochastic) data. All the eighteen flutter derivatives have been simultaneously extracted from the output response data obtained from wind tunnel test on a 3-DOF elastically suspended bridge deck section-model. Simplicity in model suspension and measurements of only output responses are additional motivating factors for adopting COV-SSI technique. The identified discrete values of flutter derivatives have been approximated by rational functions.