• 정보교육학회논문지
• /
• 제25권3호
• /
• pp.459-469
• /
• 2021

본 논문에서는 현장 교사 및 예비교사를 위한 기초 데이터과학 실습 교육 사례를 연구하였다. 본 논문에서는 기초 데이터과학 교육을 위해, 스프레드시트 SW를 데이터 수집 및 분석 도구로 사용하였다. 이후 데이터 가공, 예측 가설 및 예측 모델 검증을 위한 통계학을 교육하였다. 또한, 수천명 단위의 공공 빅데이터를 수집 및 가공하고, 모집단 예측 가설 및 예측 모델을 검증하는 교육 사례를 제안하였다. 이와 같은 데이터과학의 기초 교육내용을 담아, 스프레드시트 도구를 활용한 34시간 17주 교육 과정을 제시하였다. 데이터 수집, 가공 및 분석을 위한 도구로서, 스프레드시트는 파이썬과 달리, 프로그래밍 언어 및 자료구조에 대한 학습 부담이 없고, 질적 데이터와 양적 데이터에 대한 가공 및 분석 이론을 시각적으로 습득할 수 있는 장점이 있다. 본 교육 사례 연구의 결과물로서, 세가지 예측 가설 검증 사례들을 제시하고 분석하였다. 첫 번째로, 양적 공공데이터를 수집하여 모집단의 그룹별 평균값 차이 예측 가설을 검증하였다. 두 번째로, 질적 공공데이터를 수집하여 모집단의 질적 데이터 내 연관성 예측 가설을 검증하였다. 세 번째로, 양적 공공데이터를 수집하여 모집단의 양적 데이터 내 상관성 예측 가설 검증에 따른 회귀 예측 모델을 검증하였다. 그리고 본 연구에서 제안한 교육 사례의 효과성을 검증하기 위해, 예비교사와 현장교사의 만족도분석을 실시하였다.

• 한국전산유체공학회:학술대회논문집
• /
• 한국전산유체공학회 2001년도 추계 학술대회논문집
• /
• pp.118-125
• /
• 2001

The performance prediction of SMART MCP was performed using a computational fluid dynamics code. General capacity-head performance curve of MCP, which is provided to other design branches as design input, was obtained and it showed the typical type of axial pump performance curve. When four MCPs operate in parallel and one of them stops while the others continue to operate, SMART requires reduced power operation. A procedure for predicting the performance of SMART MCP for that case was developed and verified with available experimental data. An analysis based on the developed procedure was performed for two cases; the impeller of sloped MCP is fixed or free to rotate in reverse direction. According to the results, $73\%$ flow rate of normal operation enters the reactor core in the case of the locked impeller. In case of the impeller free rotation, the flow rate entering the reactor core is $62.8\%$.

• 전자공학회논문지S
• /
• 제34S권11호
• /
• pp.1-8
• /
• 1997

A prediction model is proposed to describe the path loss in propagation environment of indoor microcell. This model includes the lineal corridor for line--of-sight(LOS) and T-shaped corridor for non-line-of-sight(NLOS). In computation of receiving power the ray tracing technique based on image method is utilized and also reflected waves bounced on the walls and ceilings are considered. To check validity of the computed resuls cross checks between the predicted and measured are being made, which shows a close agreement for LOS case whereas somewhat disagreement for NLOS case. UTD technique is incorporated with propagation path determination algorithm in the treatment of NLOS case.

• Journal of Information Technology Applications and Management
• /
• 제13권1호
• /
• pp.77-94
• /
• 2006

Case-based Reasoning (CBR) is widely used in business and industry prediction. It is suitable to solve complex and unstructured business problems. Recently, the prediction accuracy of CBR has been enhanced by not only various machine learning algorithms such as genetic algorithms, relative weighting of Artificial Neural Network (ANN) input variable but also data mining technique such as feature selection, feature weighting, feature transformation, and instance selection As a result, CBR is even more widely used today in business area. In this study, we investigated the usefulness of the CBR method in forecasting success in implementing ERP systems. We used a CBR method based on the feature weighting technique to compare the performance of three different models : MDA (Multiple Discriminant Analysis), GECBR (GEneral CBR), FWCBR (CBR with Feature Weighting supported by Analytic Hierarchy Process). The study suggests that the FWCBR approach is a promising method for forecasting of successful ERP implementation in Small and Medium sized Enterprises.

• 재무관리논총
• /
• 제3권2호
• /
• pp.267-279
• /
• 1996

The prediction of future demand is a vital task in managing business operations. To this end, traditional approaches often focused on statistical techniques such as exponential smoothing and moving average. The need for better accuracy has led to nonlinear techniques such as neural networks and case based reasoning. In addition, experimental design techniques such as orthogonal arrays may be used to assist in the formulation of an effective methodology. This paper investigates a multistrategy approach involving neural nets, case based reasoning, and orthogonal arrays. Neural nets and case based reasoning are employed both separately and in combination, while orthoarrays are used to determine the best architecture for each approach. The comparative evaluation is performed in the context of an application relating to the prediction of Treasury notes.

• 한국컴퓨터정보학회논문지
• /
• 제14권10호
• /
• pp.1-10
• /
• 2009

현대의 프로세서 아키텍처에서 정확한 분기 예측은 시스템의 성능에 지대한 영향을 끼친다. 최근의 연구들은 예측 정확도뿐만 아니라, 예측 지연시간 또한 성능에 막대한 영향을 끼친다는 것을 보여준다. 하지만, 예측 지연시간은 간과되는 경향이 있다. 본 논문에서는 분기 예측지연시간을 극복하기 위한 분기 선예측 기법을 제안한다. 이 기법은 분기장치를 인출 단계에서 분리함으로써, 분기 예측기가 명령어 인출 장치로부터의 아무런 정보도 없이 스스로 분기 예측을 진행 가능하게 한다. 또한, 제안된 기법을 지원하기 위해, BTB의 구조를 새롭게 개선하였다. 실험 결과는 제안된 기법이 동일한수준의 분기 예측정확도를 유지하면서, 대부분의 예측지연시간을 은폐한다는 것을 보여준다. 더욱이 제안된 기법은 항상 1 싸이클의 예측 지연시간을 가지는 이상적인 분기 예측기를 사용한 경우보다도 더 나은 성능을 보여준다. 본 논문의 실험 결과에 따르면, 기존의 방식과 비교했을 때, 최대 11.92% 평균 5.15%의 IPC 향상을 가져온다.

• 한국경영정보학회:학술대회논문집
• /
• 한국경영정보학회 2007년도 International Conference
• /
• pp.608-613
• /
• 2007

Prediction of corporate bankruptcies has long been an important topic and has been studied extensively in the finance and management literature because it is an essential basis for the risk management of financial institutions. Recently, support vector machines (SVMs) are becoming popular as a tool for bankruptcy prediction because they use a risk function consisting of the empirical error and a regularized term which is derived from the structural risk minimization principle. In addition, they don't require huge training samples and have little possibility of overfitting. However. in order to Use SVM, a user should determine several factors such as the parameters ofa kernel function, appropriate feature subset, and proper instance subset by heuristics, which hinders accurate prediction results when using SVM In this study, we propose a novel hybrid SVM classifier with simultaneous optimization of feature subsets, instance subsets, and kernel parameters. This study introduces genetic algorithms (GAs) to optimize the feature selection, instance selection, and kernel parameters simultaneously. Our study applies the proposed model to the real-world case for bankruptcy prediction. Experimental results show that the prediction accuracy of conventional SVM may be improved significantly by using our model.

• 대기
• /
• 제17권1호
• /
• pp.27-44
• /
• 2007

The purpose of this study is to investigate predictability of the seasonal simulation by the METRI (Meteorological Research Institute) AGCM (Atmospheric General Circulation Model), which is a long-term prediction model for the METRI 3-month prediction system. We examine the performance skill of climate simulation and predictability by the analysis of variance of the METRI AGCM, focusing on the precipitation, 850 hPa temperature, and 500 hPa geopotential height. According to the result, the METRI AGCM shows systematic errors with seasonal march, and represents large errors over the equatorial region, compared to the observation. Also, the response of the METRI AGCM by the variation of the sea surface temperature is obvious for the wintertime and springtime. However, the METRI AGCM does not show the significant ENSO-related signal in autumn. In case of prediction over the east Asian region, errors between the prediction results and the observation are not quite large with the lead-time. However, in the predictability assessment using the analysis of variance method, longer lead-time makes the prediction better, and the predictability becomes better in the springtime.

• 제어로봇시스템학회논문지
• /
• 제20권7호
• /
• pp.730-735
• /
• 2014

This paper introduces a GP (Genetic Programming) based robust technique for the prediction of a heavy rain newsflash. The nature of prediction for precipitation is very complex, irregular and highly fluctuating. Especially, the prediction of heavy precipitation is very difficult. Because not only it depends on various elements, such as location, season, time and geographical features, but also the case data is rare. In order to provide a robust model for precipitation prediction, a nonlinear and symbolic regression method using GP is suggested. The remaining part of the study is to evaluate the performance of prediction for a heavy rain newsflash using a GP based nonlinear regression technique in Korean regions. Analysis of the feature selection is executed and various fitness functions are proposed to improve performances. The KLAPS data of 2006-2010 is used for training and the data of 2011 is adopted for verification.

• 한국가시화정보학회지
• /
• 제20권2호
• /
• pp.78-85
• /
• 2022

In these days, the rapid development in prediction technology using artificial intelligent is being applied in a variety of engineering fields. Especially, dimensionality reduction technologies such as autoencoder and convolutional neural network have enabled the classification and regression of high-dimensional data. In particular, pixel level prediction technology enables semantic segmentation (fine-grained classification), or physical value prediction for each pixel such as depth or surface normal estimation. In this study, the pressure distribution of the ship's surface was estimated at the pixel level based on the artificial neural network. First, a potential flow analysis was performed on the hull form data generated by transforming the baseline hull form data to construct 429 datasets for learning. Thereafter, a neural network with a U-shape structure was configured to learn the pressure value at the node position of the pretreated hull form. As a result, for the hull form included in training set, it was confirmed that the neural network can make a good prediction for pressure distribution. But in case of container ship, which is not included and have different characteristics, the network couldn't give a reasonable result.