• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.9 no.2
• /
• pp.1-11
• /
• 2010

The provision of warning information on upcoming hazards leading to potential crash occurrence is a significant countermeasure to prevent crashes on the highway. This study presents a methodology for selecting more effective warning messages using a multi-criteria value function. The understandability, preference level, and message reading time were used as measures of effectiveness (MOE) for messages. Expert judgements were incorporated into the value function by analytical hierarchical process (AHP) technique. Field experiments to evaluate the warning messages were conducted in a testbed section on the Jungboo-Naeryuk freeway. The proposed methodology would be a useful tool to support the design of various traffic information messages.

• Journal of KIISE
• /
• v.43 no.12
• /
• pp.1396-1403
• /
• 2016

In streaming data analysis where underlying data distribution may be changed or the concept of interest can drift with the progress of time, the ability to adapt to concept drift can be very powerful especially in the process of incremental learning. In this paper, we develop a general framework for an adaptive incremental classifier on data stream with concept drift. A distribution, representing the performance pattern of a classifier, is constructed by utilizing the distance between the confidence score of a classifier and a class indicator vector. A hypothesis test is then performed for concept drift detection. Based on the estimated p-value, the weight of outdated data is set automatically in updating the classifier. We apply our proposed method for two types of linear discriminant classifiers. The experimental results on streaming data with concept drift demonstrate that the proposed adaptive incremental learning method improves the prediction accuracy of an incremental classifier highly.

• KIPS Transactions on Computer and Communication Systems
• /
• v.9 no.4
• /
• pp.89-100
• /
• 2020

In order to reduce the time and cost used in insurance processing, studies have been actively carried out to apply blockchain smart contract technology to car insurance. However, by using traffic data that is insufficient to prove accidents, existing studies are being exposed to the risk of insurance fraud, such as forgery and overstated damage by malicious insurers. To solve this problem, we propose an accident data-based reliability estimation model by using both various types of data through sensors, RSUs, and IoT devices embedded in automobiles and smart contracts. In particular, the regression model was applied in consideration of the weight estimation according to the type of traffic accident data and the reliability estimation model trained according to various accident situations. The proposed model is expected to effectively reduce fraud and insurance litigation while providing transparency in the insurance process and streamlining it is well.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.5
• /
• pp.1125-1132
• /
• 2017

In a situation where the severity and frequency of drought events getting stronger and higher, many studies related to drought forecast have been conducted to improve the drought forecast accuracy. However it is difficult to predict drought events using a single model because of nonlinear and complicated characteristics of temporal behavior of drought events. In this study, in order to overcome the shortcomings of the single model approach, we first build various single models capable to explain the relationship between the meteorological drought index, Standardized Precipitation Index (SPI), and other independent variables such as world climate indices. Then, we developed a combined models using Stochastic Gradient Descent method among Ensemble Learnings.

• Korean Journal of Construction Engineering and Management
• /
• v.10 no.4
• /
• pp.3-13
• /
• 2009

When promoting BTL(Build Transfer Lease; below BTL) project, negotiation is a stage of examining observation and reflection of RFP(Request for Proposal below RFP) in terms with facilities, operating and financing. It keeps an important position in whole process. However, there is no consistent guideline or model which helps evaluating the result of negotiation. It is difficult to apprise the quantitative outcome after executing negotiation. Thus, this study presented the Value Engineering -based process and model of estimating the outcome of negotiation for the purpose of estimating and verifying the result of negotiation objectively, Evaluating factors of negotiation were classified into 6 fields, 38 divisions and 135 items, focused on cultural facilities on BTL project. Weight of every factor was estimated, and quantitative checklist was established. This study presented the model which could measure the outcome of negotiation. This result would be a critical checklist before negotiation on BTL project, an index of feedback during negotiation, and also a standard of estimating the outcome after negotiation.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2006.05a
• /
• pp.701-705
• /
• 2006

강우자료는 수문 모델링 작업에서 가장 기초적인 수문학적 입력자료로 시간과 공간에 따른 변동성이 크므로 규명하기 복잡한 수문현상 중의 하나이다. 산악지역이 많은 우리나라의 지형학적 특성과 태풍, 장마 및 특히, 최근의 게릴라성 집중호우 등으로 인하여 이러한 변동성이 더욱 커지고 있는 실정이다. 장기간 실측된 수문기상 기초 자료가 부족한 우리나라의 실정상 홍수예보 및 수공구조물 설계를 위해 정확한 강우량 자료의 취득이 선행돼야 한다. 따라서 적절한 장소에 수문관측소 설치 및 관리를 통해 양호한 강우량 자료를 획득해야 하지만, 현장 여건상 등의 이유로 미계측 및 결측, 이상자료가 발생하고 있다. 따라서 이러한 미계측 혹은 결측지점의 우량을 추정할 수 있는 방법을 비교, 분석하여 적절한 보정과정을 수행할 필요가 있다. 그간의 연구에서는 미계측 지점 혹은 산악지역에서의 점 강우량 보정방법에 대한 연구가 진행되었지만, 본 연구에서는 '도시홍수재해관리기술연구사업단'에서 운영 중인 도시하천 유역 특히 소배수구역에서의 결측 자료에 대해 여러 추정 방법을 비교, 분석하여 적절한 방안을 찾고자 한다. 이를 위하여 중랑천 유역의 3개 소배수 구역(월계1 배수구역, 군자 배수구역, 어린이대공원 배수구역)에 설치된 3개 우량관측소와 건설교통부 관할 우량관측소 2개소의 우량자료를 사용하였다. 본 연구에서는 결측치 보간을 위하여 널리 이용되고 있는 산술평균법(Arithmetic Average method), 역거리법(Reciprocal Distance Squared method), 거리고도비율법(Ratio of Distance and Elevation method), 인근관측소와의 관계식 이용, 크리깅방법(Simple Kriging method)을 비교, 검토 적용하였다. 중랑천 유역의 소배수구역을 대상으로 연중 발생하는 큰 호우사상에 대해 임의의 강우관측소를 결측지점으로 가정하고 주변의 강우관측소로부터 각각의 방법을 이용해 가중치들을 산정하여 결측지점의 강우량 값을 보정하고자 하였다. 또한 각각의 방법을 이용하여 얻어진 결과에 대해 실측값과 보정값의 오차정도를 평균절대오차법(Mean Absolute Error)과 제곱평균제곱근오차법(Root Mean Squared Error)에 의해 산정하여 보정 방법간의 효율성을 검토하고자 하였다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2009.05a
• /
• pp.198-202
• /
• 2009

기후변화는 강수와 기온을 변화시켜 수자원에 지대한 영향을 미칠 것으로 알려져 있다. 따라서 이에 대한 안정적인 수자원 관리를 위해서는 기후변화 영향을 정량적으로 평가하는 것이 필요하다. 기본적으로 기후변화에 대한 수자원의 영향을 연구할 때 '온실가스 배출시나리오, GCMs을 통한 기후모의, 시공간적 편차보정을 위한 상세화, 유출모형 적용을 통한 유출시나리오 생산'의 과정을 거친다. 그러나 유출시나리오를 얻기까지 과정에는 각각 불확실성을 가지고 있기 때문에 최종결과의 불확실성은 각 과정을 거치면서 매우 커진다고 할 수 있다. 다양한 배출시나리오, GCM 결과, 유출모형에 대해 단순평균 혹은 가중치를 주는 multi-model ensemble 기법은 각 경우에 따른 값의 범위를 제시할 수있다는 점 때문에 불확실성 평가에서 주로 이용되고 있다. 본 연구에서는 우리나라 5대강 유역 109개 중권역에 대해 multi-model ensemble을 적용하여 기후변화에 의한 수자원 영향을 평가하였다. 1971년에서 2100년까지 120년 기간에 대해 3개의 온실가스 배출시나리오, 13개의 GCMs 결과들을 수집하여 총 39개의 기후시나리오를 이용하였고, 이를 8개의 유출모형에 적용하여 총 312개의 유출시나리오를 생산하였다. 생산된 유출시나리오를 기준시간(1971${\sim}$2000)에 대한 미래의 세 기간(2020s, 2050s, 2080s)으로 나누어 변화율을 분석한 결과 여름철 유출량과 겨울철 유출량이 증가될것으로 나타났으나 겨울철 유출량 전망은 여름철에 비해 불확실성이 큰 것으로 나타났다. 공간적으로는 한강유역이 위치한 북쪽유역이 남쪽에 비해 불확실성이 큰 것으로 나타났다. 결과적으로 유출의 시공간적 편차에 의해 우리나라 수자원은 홍수피해 증가가 예상되었으며, 월별유출량의 변화로 인해 용수확보와 관리에 어려움이 증가할 것으로 전망되었다.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.17 no.1
• /
• pp.75-82
• /
• 2004

With an aim at eliminating the numerical dispersion error arising from the numerical simulation of stress wave propagation, numerical dispersion characteristics of the wave equation based one-dimensional finite element model are analyzed and some dispersion control scheme are proposed in this paper The dispersion analyses are carried out for two types of mass matrix, namely the consistent and the lumped mass matrices. Based on the finding of the analyses, dispersion correction techniques are developed for both the implicit and explicit schemes. For the implicit scheme, either the weighting factor for the spatial derivatives of each time level or the lumping coefficient for mass matrix is adjusted to minimize the numerical dispersion. In the case of the explicit scheme an artificial dispersion term is introduced in the governing equation. The validity of the dispersion correction techniques proposed in this study is demonstrated by comparing the numerical solutions obtained using the Present techniques with the analytical ones.

• The Korean Journal of Applied Statistics
• /
• v.21 no.2
• /
• pp.211-219
• /
• 2008

Customer Loyalty is very important for a company to be survived and to make profit for a long time. Especially, since the credit card company has to manage proper card members and merchants, the CRM(Customer Relationship Management) is much emphasized. A loyalty score is more essential to credit card companies which provide differential financial services based on card members and merchants than any other companies. In this paper, we discuss behavioral measures to define customer loyalty and suggest a method to make loyalty score with an example of a credit card company. The loyalty score developed is considered easy to understand and simple to apply in card industry. In the development of loyalty score, first, we define the loyal customers and non-loyal customers by measuring variables indicating loyalty. And we perform individual logistic regression by each exploratory measuring variable and obtain the weight of measure variables using Chi-square statistics which is used for model fitness. The loyalty score suggested shows very stable results in terms of PSI (Population Stability Index) as time goes.

• Journal of Korea Multimedia Society
• /
• v.3 no.3
• /
• pp.253-262
• /
• 2000

This paper proposes a new RBF neural network that determines the number and the center of hidden neurons based on the adaptive feature extraction for the input data. The principal component analysis is applied for extracting adaptively the features by reducing the dimension of the given input data. It can simultaneously achieve a superior property of both the principal component analysis by mapping input data into set of statistically independent features and the RBF neural networks. The proposed neural networks has been applied to classify the 200 breast cancer databases by 2-class. The simulation results shows that the proposed neural networks has better performances of the learning time and the classification for test data, in comparison with those using the k-means clustering algorithm. And it is affected less than the k-means clustering algorithm by the initial weight setting and the scope of the smoothing factor.