• Journal of applied mathematics & informatics
• /
• v.3 no.2
• /
• pp.253-264
• /
• 1996

The generalized logit model of nominal type with random regressors is studied for bootstrapping. We assess the accuracy of some estimators for our generalized logit model using a Monte Carlo simu-lation. That is we study the finite sample properties containing the consistency and asymptotic normality of the maximum likelihood es-timators. Also we compare Newton Raphson algorithm with BHHH algorithm.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.10b
• /
• pp.308-310
• /
• 2000

관찰학습(OLA: Observational Learning Algorithm)은 앙상블 네트워크의 각 구성 모델들이 아른 모델들을 관찰함으로써 얻어진 가상 데이터와 초기에 bootstrap된 실제 데이터를 학습에 함께 이용하는 방법이다. 본 논문에서는, 초기 학습 데이터 셋을 분할하고 분할된 각 데이터 셋에 대하여 앙상블의 구성 모델들을 전문화(specialize)시키는 방법을 적용하여 기존의 관찰학습 알고리즘을 개선시켰다. 제안된 알고리즘은 bagging 및 boosting과의 비교 실험에 의하여, 보다 적은 수의 구성 모델로 동일 내지 보다 나은 성능을 나타냄이 실험적으로 검증되었다.

• MALSORI
• /
• no.58
• /
• pp.67-81
• /
• 2006

For a large corpus of time-aligned data, HMM based approaches are most widely used for automatic segmentation, providing a consistent and accurate phone labeling scheme. There are two methods for training in HMM. Flat starting method has a property that human interference is minimized but it has low accuracy. Bootstrap method has a high accuracy, but it has a defect that manual segmentation is required In this paper, a new algorithm is proposed to minimize manual work and to improve the performance of automatic segmentation. At first phase, voiced, unvoiced and silence classification is performed for each speech data frame. At second phase, the phoneme sequence is aligned dynamically to the voiced/unvoiced/silence sequence according to the acoustic phonetic rules. Finally, using these segmented speech data as a bootstrap, phoneme model parameters based on HMM are trained. For the performance test, hand labeled ETRI speech DB was used. The experiment results showed that our algorithm achieved 10% improvement of segmentation accuracy within 20 ms tolerable error range. Especially for the unvoiced consonants, it showed 30% improvement.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.1
• /
• pp.12-17
• /
• 2003

Optimal determination of cluster size has an effect on the result of clustering. In K-means algorithm, the difference of clustering performance is large by initial K. But the initial cluster size is determined by prior knowledge or subjectivity in most clustering process. This subjective determination may not be optimal. In this Paper, the genetic algorithm based optimal determination approach of cluster size is proposed for automatic determination of cluster size and performance upgrading of its result. The initial population based on attribution is generated for searching optimal cluster size. The fitness value is defined the inverse of dissimilarity summation. So this is converged to upgraded total performance. The mutation operation is used for local minima problem. Finally, the re-sampling of bootstrapping is used for computational time cost.

• Proceedings of the KSRS Conference
• /
• 1998.09a
• /
• pp.76-82
• /
• 1998

Increase of greenhouse gas due to $CO_2$ and CH$_4$ gases would cause the global warming in the atmosphere. According to the global circulation model, it is pointed out in the Okhotsk Sea that the large increase of atmospheric temperature might be occurredin this region by global warming due to the doubling of greenhouse effectgases. Therefore, it is very important to monitor the sea ice extents in the Okhotsk Sea. To improve the sea ice extents and concentration with more highly accuracy, the field experiments have begun to comparewith Airborne Microwave Radiometer (AMR) and video images installed on the aircraft (Beach-200). The sea ice concentration is generally proportional to the brightness temperature and accurate retrieval of sea ice concentration from the brightness temperature is important because of the sensitivity of multi-channel data with the amount of open water in the sea ice pack. During the field experiments of airborned AMR the multi-frequency data suggest that the sea ice concentration is slightly dependending on the sea ice types since the brightness temperature is different between the thin and small piece of sea ice floes, and a large ice flow with different surface signatures. On the basis of classification of two sea ice types, it is cleary distinguished between the thin ice and the large ice floe in the scatter plot of 36.5 and 89.0GHz, but it does not become to make clear of the scatter plot of 18.7 and 36.5GHz Two algorithms that have been used for deriving sea ice concentrations from airbomed multi-channel data are compared. One is the NASA Team Algorithm and the other is the Bootstrap Algorithm. Intrercomparison on both algorithms with the airborned data and sea ice concentration derived from video images bas shown that the Bootstrap Algorithm is more consistent with the binary maps of video images.

• Journal of Korean Society of Transportation
• /
• v.20 no.1
• /
• pp.39-52
• /
• 2002

Traffic data by vehicle classification is difficult for mutual exchange of data due to the different vehicle classification from each other by the data sources; as a result, application of the data is very limited. In Particular. in case of TCS vehicle classification in national highways, passenger car, van and truck are mixed in one category and the practical usage is very low. The research standardize the vehicle classification to convert other data and develop the model which can estimate national highway traffic data by the standardized vehicle classification from the raw traffic data obtained at the highway tollgates. The tollgates are categorized into several groups by their features and the model estimates traffic data by the standardized vehicle classification by using the point estimation and bootstrap algorithm. The result indicates that both of the two methods above have the significant level. When considering the bias of the extreme value by the sample size, the bootstrap algorithm is more sophisticated. Using result of this study, we is expect the usage improvement of TCS data and more specific comparison between the freeway traffic investigation and link volume on freeway using the TCS data.

• Journal of KIISE:Software and Applications
• /
• v.30 no.3_4
• /
• pp.251-258
• /
• 2003

A support vector machine (SVM) is supposed to provide a good generalization performance, but the actual performance of a actually implemented SVM is often far from the theoretically expected level. This is largely because the implementation is based on an approximated algorithm, due to the high complexity of time and space. To improve this limitation, we propose ensemble of SVMs by using Bagging (bootstrap aggregating) and Boosting. By a Bagging stage each individual SVM is trained independently using randomly chosen training samples via a bootstrap technique. By a Boosting stage an individual SVM is trained by choosing training samples according to their probability distribution. The probability distribution is updated by the error of independent classifiers, and the process is iterated. After the training stage, they are aggregated to make a collective decision in several ways, such ai majority voting, the LSE(least squares estimation) -based weighting, and double layer hierarchical combining. The simulation results for IRIS data classification, the hand-written digit recognition and Face detection show that the proposed SVM ensembles greatly outperforms a single SVM in terms of classification accuracy.

• Journal of Korean Institute of Industrial Engineers
• /
• v.39 no.2
• /
• pp.82-89
• /
• 2013

Novelty detection (ND) is an effective technique that can be used to determine whether a future observation is normal or not. In the present study we propose a novelty detection algorithm that can handle a situation where the distributions of target (normal) observations are inhomogeneous. A simulation study and a real case with the TFT-LCD process demonstrated the effectiveness and usefulness of the proposed algorithm.

• Journal of the Korean Data and Information Science Society
• /
• v.14 no.3
• /
• pp.441-450
• /
• 2003

Collaborative filtering is one of the most widely used methodologies for recommendation system. Collaborative filtering is based on a data matrix of each customer's preferences and frequently, there exits missing data problem. We introduced two imputation approach (multiple imputation via Markov Chain Monte Carlo method and multiple imputation via bootstrap method) to improve the prediction performance of collaborative filtering and evaluated the performance using EachMovie data.

• Journal of Internet of Things and Convergence
• /
• v.2 no.3
• /
• pp.17-24
• /
• 2016

In this paper, we developed web based Plossom system, an efficient and convenient project management system. The development of Plossom consists of a web page using HTML and Bootstrap framework, and the logic is built using the algorithm suitable for web page by using JQuery. Additionally, Plossom is implemented to support interoperability with external system by using Spring's Thymeleaf and MySQL efficiently. The Plossom system developed in this study is expected to be very useful in the development process of IOT and ICT related projects.