• Journal of the military operations research society of Korea
• /
• v.33 no.2
• /
• pp.61-73
• /
• 2007

We examined technological forecasting of extended TFDEA(Technological Forecasting with Data Envelopment Analysis) and thereby apply the extended method to the technological forecasting problem of main battle tank. The TFDEA has the possibility of using comparatively inefficient DMUs(Decision Making Units) because it is based on DEA(Data Envelopment Analysis), which usually leads to multiple efficient DMUs. Therefore, TFDEA may result in incorrect technological forecasting. Instead of using the simple DEA, we incorporated the concept of Super-efficiency, Cross-efficiency, and CCCA(Constrained Canonical Correlation Analysis) into the TFDEA respectively, and applied each method to the case study of main battle tank using verifiable practical data sets. The comparative analysis shows that the use of CCCA with TFDEA results in very comparable prediction accuracies with respect to MAE(Mean Absolute Error), MSE(Mean Squared Error), and RMSE(Root Mean Squared Error) than using the concept of Super-efficiency and Cross-efficiency.

• The Korean Journal of Applied Statistics
• /
• v.30 no.1
• /
• pp.25-39
• /
• 2017

Over 5 million people participate in cycle racing betting and its revenue is more than 2 trillion won. This study predicts the ranking of cycle racing using various statistical analyses and identifies important variables which have influence on ranking. We propose competitive ranking prediction models using various classification and regression methods. Our model can predict rankings with low misclassification rates most of the time. We found that the ranking increases as the grade of a racer decreases and as overall scores increase. Inversely, we can observe that the ranking decreases when the grade of a racer increases, race number four is given, and the ranking of the last race of a racer decreases. We also found that prediction accuracy can be improved when we use centered data per race instead of raw data. However, the real profit from the future data was not high when we applied our prediction model because our model can predict only low-return events well.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.38 no.4
• /
• pp.59-71
• /
• 2001

The Current CORBA shows some limitations for its successful deployment in real time system applications. Recently, OMG adopted Real-Time CORBA specification, which is defined as an extension to CORBA. The goal of the Real-Time CORBA is to provide a standard for CORBA ORB implementations that support 'end to end predictability'. In order to support 'end-to-end predictability', Real Time CORBA specifies many components such as priority model, communication protocol configuration, thread management, and etc. Among them, 'priority model' is the most important mechanism for avoiding or bounding priority inversion in CORBA invocations. In this paper, we present our efforts on a design ,and implementation of the Priority Model in Real-Time CORBA specification. The implementation is done as an extension of omniORB2(v.3.0.0), a popular open source non real time ORB. Experiment results demonstrate that our priority model implementation shows better performance and predictability than the non real-time ORB.

• The Korean Journal of Applied Statistics
• /
• v.31 no.3
• /
• pp.367-382
• /
• 2018

Prediction models for a corporate bond rating in existing studies have been developed using various models such as linear regression, ordered logit, and random forest. Financial characteristics help build prediction models that are expected to be contained in the assigning model of the bond rating agencies. However, the ranges of bond ratings in existing studies vary from 5 to 20 and the prediction models were developed with samples in which the target companies and the observation periods are different. Thus, a simple comparison of the prediction accuracies in each study cannot determine the best prediction model. In order to conduct a fair comparison, this study has collected corporate bond ratings and financial characteristics from 2013 to 2017 and applied prediction models to them. In addition, we applied the elastic-net penalty for the linear regression, the ordered logit, and the ordered probit. Our comparison shows that data-driven variable selection using the elastic-net improves prediction accuracy in each corresponding model, and that the random forest is the most appropriate model in terms of prediction accuracy, which obtains 69.6% accuracy of the exact rating prediction on average from the 5-fold cross validation.

• Journal of Korea Port Economic Association
• /
• v.26 no.3
• /
• pp.18-41
• /
• 2010

The purpose of this paper is to show the empirical measurement way for predicting the seaport efficiency by using Super SBM(Slack-based Measure) with Wilcoxson signed-rank test under CRS(constant returns to scale) condition for 20 Korean ports during 11 years(1997-2007) for 3 inputs(port investment amount, birthing capacity, and cargo handling capacity) and 5 outputs(Export and Import Quantity, Number of Ship Calls, Port Revenue, Customer Satisfaction Point for Port Service and Container Cargo Throughput). The main empirical results of this paper are as follows. First, Super SBM model has well reflected the real data according to the Wilcoxon signed rank test, because p values have exceeded the significance level. Second,Super-SBM has shown about 87% of predicting ratio for the ports efficiency and the optimal size of investment in domestic seaport. The policy implication to the Korean seaports and planner is that Korean seaports should introduce the new methods like Super-SBM method with Wilcoxon signed rank test for predicting the efficiency of port performance and the optimal size of investment as indicated by Panayides et al.(2009, pp.203-204).

• Journal of Korea Port Economic Association
• /
• v.24 no.4
• /
• pp.313-327
• /
• 2008

The purpose of this paper is to show the empirical measurement way for predicting the seaport efficiency by using SBM with Wilcoxson signed-rank test under CRS(constant returns to scale) condition for 20 Korean ports during 1994-2003 for 2 inputs(birthing capacity, cargo handling capacity) and 3 outputs(Export and Import Quantity, Number of Ship Calls, Port Revenue). The main empirical results of this paper are as follows. First, forecasting data have well reflected the real data according to the Wilcoxon signed rank test, because p values have exceeded the 0.05 significance level. Second, SBM has shown the effectiveness for predicting the ports efficiency even though the predicting powers are different according to the levels of p values. The policy implication to the Korean seaports and planner is that Korean seaports should introduce the new methods like SBM method with Wilcoxon signed rank test for predicting the port performance and enhancing the efficiency.

• The Journal of Bigdata
• /
• v.5 no.2
• /
• pp.77-84
• /
• 2020

In this study, we conducted a study on the win-loss predicton analysis of korean professional baseball by artificial intelligence models. Based on the model, we predicted the winner as well as each team's final rank in the league. Additionally, we developed a website for viewers' understanding. In each game's first, third, and fifth inning, we analyze to select the best model that performs the highest accuracy and minimizes errors. Based on the result, we generate the rankings. We used the predicted data started from May 5, the season's opening day, to August 30, 2020 to generate the rankings. In the games which Kia Tigers did not play, however, we used actual games' results in the data. KNN and AdaBoost selected the most optimized machine learning model. As a result, we observe a decreasing trend of the predicted results' ranking error as the season progresses. The deep learning model recorded 89% of the model accuracy. It provides the same result of decreasing ranking error trends of the predicted results that we observe in the machine learning model. We estimate that this study's result applies to future KBO predictions as well as other fields. We expect broadcasting enhancements by posting the predicted winning percentage per inning which is generated by AI algorism. We expect this will bring new interest to the KBO fans. Furthermore, the prediction generated at each inning would provide insights to teams so that they can analyze data and come up with successful strategies.

• Annual Conference on Human and Language Technology
• /
• 2020.10a
• /
• pp.184-187
• /
• 2020

다중 지문 기계독해는 질문과 여러 개의 지문을 입력받고 입력된 지문들에서 추출된 정답 중에 하나의 정답을 출력하는 문제이다. 다중 지문 기계독해에서는 정답이 있을 단락을 선택하는 순위화 방법에 따라서 성능이 크게 달라질 수 있다. 본 논문에서는 단락 안에 정답이 있을 확률을 예측하는 단락 재순위화 모델과 선택된 단락에서 서술형 정답을 위한 세부적인 정답의 경계를 예측하는 세부 단락 선별 기법을 제안한다. 단락 순위화 모델 학습의 경우 모델 학습을 위해 각 단락의 출력에 softmax와 cross-entroy를 이용한 손실 값과 sigmoid와 평균 제곱 오차의 손실 값을 함께 학습하고 키워드 매칭을 함께 적용했을 때 KorQuAD 2.0의 개발셋에서 상위 1개 단락, 3개 단락, 5개 단락에서 각각 82.3%, 94.5%, 97.0%의 재현율을 보였다. 세부 단락 선별 모델의 경우 입력된 두 단락을 비교하는 duoBERT를 이용했을 때 KorQuAD 2.0의 개발셋에서 F1 83.0%의 성능을 보였다.

• Applied Chemistry for Engineering
• /
• v.35 no.3
• /
• pp.200-208
• /
• 2024

Chung's equations-IX and Chung's equation-XII were utilized to predict the fire risk and evaluate fire risk ratings for four types of wood: camphor, cherry, rubber, and elm trees. The combustion tests were conducted using a cone calorimeter test method by ISO 5660-1 standards. The fire risk and fire risk rating (FRR) were compared for Fire Risk Index-IX (FRI-IX) and Fire Risk Index-XII (FRI-XII). The results yielded Fire Performance Index-XI (FPI-XI) ranging from 0.08 to 11.48 and Fire Growth Index-XI (FGI-XI) ranging from 0.67 to 111.89. The Fire Risk Index-XII (FRI-XII), indicating fire risk rating, exhibited an increasing order of cherry (0.45): Grade A (Ranking 5) < PMMA (1): Grade A (Ranking 4) < elm (1.23): Grade A (Ranking 3) < rubber (1.56): Grade A (Ranking 2) << camphor (148.23): Grade G (Ranking 1). Additionally, the fire risk index-IX (FRI-IX) was cherry (0): Grade A (Ranking 3) ≈ rubber (0): Grade A (Ranking 3) ≈ elm tree (0): Grade A (Ranking 3) < PMMA (1): Grade A (Ranking 2) << camphor tree (66.67): Grade G (Ranking 1). In general, camphor was found to have the highest fire risk. In conclusion, although the expression of the index is different as shown based on the standards of FRI-IX and FRI-XII, predictions based on fire risk assessment of combustible materials showed similar trends.

• The Korean Journal of Applied Statistics
• /
• v.27 no.5
• /
• pp.681-691
• /
• 2014

Ranking-based conjoint analysis(RBCA) and choice-based conjoint analysis(CBCA) have attracted significant interest in various fields such as marketing research. When conducting research, the researcher has to select one suitable approach in consideration of strengths and weaknesses. This article performs an empirical comparison of the predictability of RBCA and CBCA in order to provide criterion for the selection. A new concept of measurement set is developed by combining the ranking set and choice set. The measurement set enables us to apply two approaches separately on the same consumer group that allows a fair comparison of predictability. RBCA and CBCA are conducted on consumer preferences for RTD-coffee; subsequently, the predicted values of market shares and hit rates are compared. The study result reveals that their predictabilities are not significantly different. Further, the result indicates that RBCA is recommended if the researcher wants to improve data quality by filtering out poor responses or to implement the market segmentation. In contrast, CBCA is recommended if the researcher wants to lessen the burden on the respondents or to measure preferences under similar conditions with the actual marketplace.