• Journal of the Korean Data and Information Science Society
• /
• v.23 no.4
• /
• pp.739-748
• /
• 2012

In this paper we propose a mixed effects least squares support vector machine (LS-SVM) for the censored data which are observed from different groups. We use weights by which the randomly right censoring is taken into account in the nonlinear regression. The weights are formed with Kaplan-Meier estimates of censoring distribution. In the proposed model a random effects term representing inter-group variation is included. Furthermore generalized cross validation function is proposed for the selection of the optimal values of hyper-parameters. Experimental results are then presented which indicate the performance of the proposed LS-SVM by comparing with a standard LS-SVM for the censored data.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2011.10a
• /
• pp.831-834
• /
• 2011

LEACH(One of the leading algorithms in the field of WSN) for the life of the system, even by the number of all nodes to ensure that the cluster header. However, each round does not guarantee a certain number of cluster header. So sometimes cluster header is elected of small number or not elected. If cluster header number is to small, takes a heavy load on cluster header. And empty cluster is occur depending on the location of the cluster header. The algorithm proposed in this paper, the area of interest is divided into sectors. And randomly, cluster header be elected one the in each sector. When clustering the sensor nodes will belong to the nearest cluster header. So clustering is independent of the sector. This algorithm is guarantee a certain number of cluster header in each round. And has prevent occurrence of empty cluster.

• The KIPS Transactions:PartC
• /
• v.11C no.5
• /
• pp.621-632
• /
• 2004

Recent measurement studies of real teletraffic data in modern telecommunication networks have shown that self-similar (or fractal) processes may provide better models of teletraffic in modern telecommunication networks than Poisson processes. If this is not taken into account, it can lead to inaccurate conclusions about performance of telecommunication networks. Thus, an important requirement for conducting simulation studies of telecommunication networks is the ability to generate long synthetic stochastic self-similar sequences. A new generator of pseu-do-random self-similar sequences, based on the fractional Gaussian nois and a wavelet transform, is proposed and analysed in this paper. Specifically, this generator uses Daubechies wavelets. The motivation behind this selection of wavelets is that Daubechies wavelets lead to more accurate results by better matching the self-similar structure of long range dependent processes, than other types of wavelets. The statistical accuracy and time required to produce sequences of a given (long) length are experimentally studied. This generator shows a high level of accuracy of the output data (in the sense of the Hurst parameter) and is fast. Its theoretical algorithmic complexity is 0(n).

• Proceedings of the Korea Water Resources Association Conference
• /
• 2019.05a
• /
• pp.101-101
• /
• 2019

전 지구적 기온상승으로 인한 기후변화는 사회적, 수문학적, 다양한 분야에 영향을 미친다. 또한 IPCC(Intergovernmental Panel on Climate Change)의 보고서에 따르면 미래에도 지속적으로 기온상승이 예상되며, 이러한 현상은 인류의 삶에 큰 영향을 미칠것으로 예상된다. 또한 수자원 및 관련 분야에서도 기온 상승에 따른 강수량, 강수의 주기 변동, 극한 기후사상의 심도(severity)와 빈도 변화에 따른 다양한 연구가 진행되고 있으며, 미래의 강우량과 온도를 예측하는 기후변화연구에서는 다양한 기후모형을 고려하여 분석한다. 하지만 모든 기후모형이 우리나라에 적합한 것은 아니므로 과거 기후를 모의한 결과를 토대로 성능이 뛰어난 모형의 결과에 더 높은 가중치를 주고 미래를 예측하는 연구가 활발히 진행되고 있다. 일반적으로 기후모형으로 GCM (General Circulation Model) 모의 결과가 이용되는데 우리나라에 대한 GCM 결과의 정확성을 분석하는 연구는 부족한 실정이다. 따라서 본 연구에서는 21개의 GCM을 대상으로 과거 모의 자료(1970년~2005년)를 실제 관측소에서 관측된 강수량과 비교하여 각 GCM들의 성능을 평가하고 이를 토대로, GCM들의 우선순위를 선정하였다. 또한 격자 기반 GCM 결과를 IDW (Inverse Distance Weighted) 방법을 사용하여 기상관측소로 지역적 상세화를 수행하였으며, GCM과 관측자료 사이의 편이를 보정하기 위해 6가지의 Quantile Mapping 방법과 Random Forest 기법을 사용하였다. 또한 편이 보정 기법 중 성능이 좋은 기법을 선택하여 관측소에 적용하였다. 편이 보정된 GCM 모의결과에 대한 성능을 토대로 우수한 GCM 순위를 도출하기 위해 다기준의사결정기법 중 하나인 TOPSIS (Technique for Order of Preference by Similarity to Ideal Solution)를 이용하였다. 그리고 GCM의 전망기간인 2010년부터 2018년까지의 Machine learning 방법과 Quantile mapping의 기법을 비교 및 성능이 우수한 편이 보정 방법을 선택한 후 전망기간 동안의 GCM 성능의 우선순위를 선정하였다.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.29 no.4
• /
• pp.729-738
• /
• 2019

In this paper, we propose two-way authentication protocol between unmanned systems in tactical wireless networks in which long distance communications are not guaranteed due to a poor channel conditions. It is assumed that every unmanned systems have same random data set before they put into combat. The proposed protocol generates authentication code(AC) using random data that causes hash collision. The requester for authentication encrypts the materials such as their identifier, time-stamp, authentication code with the secret key. After then the requester transmits the encrypted message to the receiver. The receiver authenticates the requester by verifying the authentication code included in the request message. The performance analysis of the proposed protocol shows that it guarantees the security for various attack scenarios and efficiency in terms of communication overhead and computational cost. Furthermore, we analyzed the effect of the parameter values of the proposed protocol on the performance and suggest appropriate parameter value selection guide according to the level of security requirement.

• The Journal of Bigdata
• /
• v.5 no.2
• /
• pp.155-170
• /
• 2020

The purpose of this study was to apply machine learning techniques (Decision Tree, Random Forest, XGBoost) to data from the 4th~6th year of the Seoul Education Longitudinal Study to find the factors predicting the career maturity of middle school students in Seoul city. In order to evaluate the machine learning application result, the performance of the model according to the indicators was checked. In addition, the model was analyzed using the XGBoostExplainer package, and R and R Studio tools were used for this study. As a result, there was a slight difference in the ranking of variable importance by each model, but the rankings were high in 'Achievement goal awareness', 'Creativity', 'Self-concept', 'Relationship with parents and children', and 'Resilience'. In addition, using the XGBoostExplainer package, it was found that the factors that protect and deteriorate career maturity by panel and 'Achievement goal awareness' is the top priority factor for predicting career maturity. Based on the results of this study, it was suggested that a comparative study of machine learning and variable selection methods and a comparative study of each cohort of the Seoul Education Termination Study should be conducted.

• Journal of the Korea Convergence Society
• /
• v.13 no.3
• /
• pp.33-41
• /
• 2022

In this article, we proposed to predict natural gas (NG) leakage levels through feature selection based on a factor analysis (FA) of the integrating the Korean Meteorological Agency data and natural gas leakage data for considering complex factors. The paper has been divided into three modules. First, we filled missing data based on the linear interpolation method on the integrated data set, and selected essential features using FA with OrdinalEncoder (OE)-based normalization. The dataset is labeled by K-means clustering. The final module uses four algorithms, K-nearest neighbors (KNN), decision tree (DT), random forest (RF), Naive Bayes (NB), to predict gas leakage levels. The proposed method is evaluated by the accuracy, area under the ROC curve (AUC), and mean standard error (MSE). The test results indicate that the OrdinalEncoder-Factor analysis (OE-F)-based classification method has improved successfully. Moreover, OE-F-based KNN (OE-F-KNN) showed the best performance by giving 95.20% accuracy, an AUC of 96.13%, and an MSE of 0.031.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.42 no.6
• /
• pp.843-852
• /
• 2022

Despite continuous safety management, the death rate of construction workers is not decreasing every year. Accordingly, various studies are in progress to prevent construction site accidents. In this paper, we developed an AI-based priority inspection target selection model that preferentially selects sites are expected to cause construction accidents among construction sites with construction costs of less than 5 billion won (KRW). In particular, Random Forest (90.48 % of accident prediction AUC-ROC) showed the best performance among applied AI algorithms (Classification analysis). The main factors causing construction accidents were construction costs, total number of construction days and the number of construction performance evaluations. In this study an ROI (return of investment) of about 917.7 % can be predicted over 8 years as a result of better efficiency of manual inspections human resource and a preemptive response to construction accidents.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.85-85
• /
• 2021

CDOM(Colored or Chromophoric Dissolved Organic Matter)은 바다, 호수 및 강에서 담수, 오수, 퇴적물 등으로부터 공급된 유기물질의 일종으로 가시광선에서 빛을 흡수하는 성질을 가지며, 2016년부터 환경부에서 선정한 하천, 호수 등 방류수의 수질오염 표준인 TOC(Total Organic Carbon)를 간접 추정할 수 있는 매개변수가 될 수 있다. 따라서, 본 연구에서는 백제보 상류 23 km 구간을 대상으로 2개년(2016~2017) 중 7일의 초분광영상 자료를 활용하여 내륙지역의 CDOM에 대한 적정 반사도 밴드값(R_rs)과 CDOM을 추정하는 알고리즘을 개발하고자 한다. CDOM은 흡수계수(α_CDOM)를 통해 간접 추정되며, 흡수계수의 기준 파장값(λ)은 연구별로 350 nm, 375 nm, 400 nm, 412 nm 및 440 nm 등 다르게 나타난다. 초분광영상은 AsaFENIX 초분광 센서에서 관측된 380~970 nm까지 4 nm 간격, 127개 대역의 분광해상도와 2 m의 공간해상도를 가진 영상을 활용하였으며, 자료의 연속성을 위해 smoothing 기법을 활용하여 가공하였다. 추정 알고리즘은 Random forest를 활용하였으며, 70%의 trainning과 30%의 test로 구분하여 적용하였다. 산출된 CDOM은 결정계수(R²), Nash-Sutcliffe efficiency(NSE)를 이용하여 실측 CDOM과 비교하였다. 흡수계수별 CDOM의 산정 결과 α_CDOM(350 nm)의 trainning, test에서 각각 R²가 0.71, 0.74, NSE가 0.25, 0.49로 가장 높았으며, 적정 반사도 밴드값은 R_rs(466), R_rs(493), R_rs(548), R_rs(641)를 사용하였을 때 trainning, test에서 각각 R²가 0.93, 0.90, NSE가 0.85, 0.69로 가장 높게 나타났다.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.5
• /
• pp.325-330
• /
• 2023

This paper proposes to apply machine learning to the process of predicting the slab thickness based on the structural analysis results or experience and subjectivity of engineers in the design of bridge data construction to enable digital-based decision-making. This study aims to build a reliable design environment by utilizing machine learning techniques to provide guide values to engineers in addition to structural analysis for slab thickness selection. Based on girder bridges, which account for the largest proportion of bridge data, a prediction model process for predicting slab thickness among superstructures was defined. Various machine learning models (Linear Regress, Decision Tree, Random Forest, and Muliti-layer Perceptron) were competed for each process to produce the prediction value for each process, and the optimal model was derived. Through this study, the applicability of machine learning techniques was confirmed in areas where slab thickness was predicted only through existing structural analysis, and an accuracy of 95.4% was also obtained. models can be utilized in a more reliable construction environment if the accuracy of the prediction model is improved by expanding the process