• Proceedings of the Acoustical Society of Korea Conference
• /
• 1987.11a
• /
• pp.89-92
• /
• 1987

In a template-matchig-based speech recognition syste, excessive weight given to perceptually unimportant spectral variations is undesirable for discriminating among acoustically similar words. By introducing a simple threshold-type nonlinearity applied to the distance metric, the word recognition performance can be improved for a vocabulary with similar sounding words, without modifying the system structure.

• Bulletin of the Korean Mathematical Society
• /
• v.49 no.5
• /
• pp.1041-1055
• /
• 2012

The parameter lower (upper) distribution set corresponds to the cylindrical lower or upper local dimension set for a self-similarmeasure on a self-similar set satisfying the open set condition.

• Journal of applied mathematics & informatics
• /
• v.14 no.1_2
• /
• pp.447-454
• /
• 2004

We study the Hausdorff and packing dimensions of subsets composing multifractal spectrum generated by a quasi-self-similar probability measure on a perturbed Cantor set.

• Communications of the Korean Mathematical Society
• /
• v.19 no.3
• /
• pp.469-475
• /
• 2004

We generalize S. Ikeda's results for perturbed cantor sets showing how we get the dimensions for deformed self-similar sets.

• Journal of applied mathematics & informatics
• /
• v.12 no.1_2
• /
• pp.385-393
• /
• 2003

Let Y(t) be a stochastic integral process represented by Brownian motion. We show that YHt (t) is continuous in t with probability one for Molder function Ht of exponent ${\beta}$ and finally we derive asymptotic self-similar process YM (t) which converges to Yw (t).

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.10e
• /
• pp.640-642
• /
• 2002

본 논문에서는 실제 네트워크에서 측정되고 있는 시간축상으로 매우 불안정한 상태의 트래픽 즉, ON／OFF 소스 모델로 가정하여 발생시킨 self-similar 트래픽을 이용하여 다단으로 이루어진 스위치 모델의 처리시간(processing time)에 따른 출력단에서self-similarity의 변화와 이용도(utilization) 등을 사용하여 네트워크의 성능을 분석한다.

• Bulletin of the Korean Mathematical Society
• /
• v.58 no.1
• /
• pp.133-146
• /
• 2021

For pseudo-free and recurrent self-similar groups, we show that continuous orbit equivalence of inverse semigroup partial actions implies continuous orbit equivalence of group actions. Conversely, if group actions are continuous orbit equivalent, and the induced homeomorphism commutes with the shift maps on their groupoids, we obtain continuous orbit equivalence of inverse semigroup partial actions.

• Proceedings of the KSME Conference
• /
• 2008.11a
• /
• pp.110-114
• /
• 2008

Determination of weld-induced residual stress has been an important issue in nuclear power industry because several failures were reported in dissimilar metal weld parts due to primary water stress corrosion cracking. In this context, a couple of remarkable round robin analyses were conducted to quantify the welding simulation variables and to establish optimized numerical analysis process. The purpose of the present research is to introduce welding simulation results for a safety and relief nozzle, which has a dissimilar metal weld part as well as a similar metal weld part. First, finite element analyses are carried out to calculate residual stresses at the inside of nozzle considering only dissimilar metal welding. Subsequently, residual stresses taking into account both the dissimilar and similar metal welding are computed. The similar metal weld effect is evaluated by compa

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2004.11a
• /
• pp.211-220
• /
• 2004

This paper suggests a collaborative business process between the companies that each has a restricted physical branch in its own area and wants to extend globally sales and delivery service. The companies integrate their business processes for sales and delivery using a shared product taxonomy table. We also suggest a similar product finding algorithm to make the product taxonomy table that defines product relationships to exchange them between the companies. The main idea of the proposed algorithm is using a multi-attribute decision making (MADM) to find the utility values of products in a same product class of the companies. Using the values we determine what products are similar. It helps the product manager to register the similar products into a same product sub-category. The companies then allow consumer to shop and purchase the products at their own residence site and deliver them or similar products to another sites.

• Journal of Information Processing Systems
• /
• v.18 no.2
• /
• pp.268-281
• /
• 2022

The development of information technology is bringing many changes to everyday life, and machine learning can be used as a technique to solve a wide range of real-world problems. Analysis and utilization of data are essential processes in applying machine learning to real-world problems. As a method of processing data in machine learning, we propose an approach based on applying multiple linear regression models by interlacing data to the task of classifying similar software. Linear regression is widely used in estimation problems to model the relationship between input and output data. In our approach, multiple linear regression models are generated by training on interlaced feature data. A combination of these multiple models is then used as the prediction model for classifying similar software. Experiments are performed to evaluate the proposed approach as compared to conventional linear regression, and the experimental results show that the proposed method classifies similar software more accurately than the conventional model. We anticipate the proposed approach to be applied to various kinds of classification problems to improve the accuracy of conventional linear regression.