• Journal of Internet of Things and Convergence
• /
• v.9 no.3
• /
• pp.1-6
• /
• 2023

One of the economical ways to satisfy the quality of service desired by the user in a web service environment is to adjust the size of the object. To this end, this study finds the maximum size of objects that satisfy this constraint when the mean response time is given below an arbitrary threshold for quality of service. It can be inferred that in the steady state of system, the mean response time in the deterministic model by using the round-robin will be the same as that of the queueing model following the general distribution. Based on this, analytical formulas and procedures for finding the maximum object size are obtained. As a service distribution of web traffic, the Pareto distribution is appropriate, so the maximum object size is computed by applying the M/G(Pareto)/1 model and the M/G/1/PS model using exponential distribution as computational experience. Performance evaluation through numerical calculation shows that as the shape parameter in the Pareto distribution increases, the M/G(Pareto)/1 model and M/G/1/PS model have the same maximum object size. The results of this study can be used to environments where objects can be sized for economical web service control.

• Journal of the Korean Data and Information Science Society
• /
• v.22 no.2
• /
• pp.171-177
• /
• 2011

With the development of the actuarial and insurance industries, the distributions of the insurance payments data are deeply studied by many authors. It is known that theses types of distribution are very highly positively skewed and have a long thick upper tail such as Pareto or lognormal distribution. In 2005, Cooray and Ananda proposed a new model which is composed lognormal distribution and Pareto distribution. They said it as composite lognormal-Preto distribution. They showed that the proposed distribution was better fitted than lognormal or Pareto distribution. On the other hand many agreements about the insurance payment have some options for a trivially small payment or extremely large one because of the limits of total payment. Appling these cases, in this paper we consider the parameter estimation on the composite lognormal-Pareto distribution based on doubly censored samples.

• Journal of the Korean Society for information Management
• /
• v.24 no.4
• /
• pp.255-265
• /
• 2007

In this paper we analyzed regularity of daily patterns of distribution of Queries coming from internet search engine. And then, we proposed a Pareto distribution and Zipf law for identifying the query distribution and applied them to daily queries on the search engine during 2 week. We found that there is some evidence that Pareto and Zipf laws can be applied to evaluate the regularity of daily patterns of distribution of queries in search engine. Those results can be used to provide a better understanding of the social interests and trends using the query distribution patterns.

• Journal of Industrial Convergence
• /
• v.18 no.3
• /
• pp.19-25
• /
• 2020

In the software development process, software reliability evaluation is a very important issue. In particular, finding the optimal development model that satisfies high reliability is the more important task for software developers. For this, in this study, Pareto and Erlang life distributions were applied to the finite failure NHPP model to evaluate the reliability attributes. For this purpose, parametric estimation is applied to the maximum likelihood estimation method, and nonlinear equations are calculated using the bisection method. As a result, the Erlang model showed better performance than the Pareto model in the evaluation of the strength function and the mean value function. Also, as a result of inputting future mission time and evaluating reliability, the Erlang model showed an effectively high trend together with the Pareto model, while the Goel-Okumoto basic model showed a decreasing trend. In conclusion, the Erlang model is the best model among the proposed models. Through this study, it is expected that software developers will be able to use it as a basic guideline for exploring and evaluating the optimal software reliability model.

• The KIPS Transactions:PartB
• /
• v.11B no.7 s.96
• /
• pp.841-848
• /
• 2004

Evolutionary a1gorithms are well-suited for multi-objective optimization problems involving several, often conflicting objectives. Pareto-based evolutionary algorithms, in particular, have shown better performance than other multi-objective evolutionary algorithms in comparison. However, generalized evolutionary multi-objective optimization algorithms have a weak point, in which the distribution of solutions are not uni-formly distributed onto Pareto optimal front. In this paper, we propose an evolutionary a1gorithm for multi-objective optimization which uses seed individuals in order to overcome weakness of algorithms Published. Seed individual means a solution which is not located in the crowded region on Pareto front. And the idea of our algorithm uses seed individuals for reproducing individuals for next generation. Thus, proposed a1go-rithm takes advantage of local searching effect because new individuals are produced near the seed individual with high probability, and is able to produce comparatively uniform distributed pareto optimal solutions. Simulation results on five testbed problems show that the proposed algo-rithm could produce uniform distributed solutions onto pareto optimal front, and is able to show better convergence compared to NSGA-II on all testbed problems except multi-modal problem.

• KIPS Transactions on Computer and Communication Systems
• /
• v.8 no.10
• /
• pp.245-250
• /
• 2019

The use of NAND flash memory has increased rapidly due to the development of mobile fields. However, NAND flash memory has a limited lifespan, so studies are underway to predict its lifespan. Workload is one of the factors that significantly affect the life of NAND flash memory, and workload analysis studies in mobile environments are insufficient. In this paper, we analyze the distribution of workload in the mobile environment by collecting traces generated by using Android-based smartphones. The collected traces can be divided into three groups of hotness. Also they are distributed in the form of heavy tails. We fit this to the Pareto, Lognormal, and Weibull distributions, and Traces are closest to the Pareto distribution.

• The Korean Journal of Applied Statistics
• /
• v.27 no.6
• /
• pp.975-989
• /
• 2014

Lasso (Tibshirani, 1996) and Elastic Net (Zou and Hastie, 2005) have been widely used in various fields for simultaneous variable selection and coefficient estimation. Bayesian methods using a conditional Laplace and a double Pareto prior specification have been discussed in the form of hierarchical specification. Full conditional posterior distributions with each priors have been derived. We compare the performance of Bayesian lassos with Laplace prior and the performance with double Pareto prior using simulations. We also apply the proposed Bayesian hierarchical models to real data sets to predict the collapse of governments in Asia.

• Communications for Statistical Applications and Methods
• /
• v.16 no.3
• /
• pp.463-477
• /
• 2009

Parameter estimation methods such as maximum likelihood estimation method, probability weighted moments method, regression method have been popularly applied to various extreme value models in numerous literature. Among three methods above, the performance of regression method has not been rigorously investigated yet. In this paper the regression method is compared with the other methods via Monte Carlo simulation studies for estimation of parameters of the Generalized Extreme Value(GEV) distribution and the Generalized Pareto(GP) distribution. Our simulation results indicate that the regression method tends to outperform other methods under small samples by providing smaller biases and root mean square errors for estimation of location parameter of the GEV model. For the scale parameter estimation of the GP model under small samples, the regression method tends to report smaller biases than the other methods. The regression method tends to be superior to other methods for the shape parameter estimation of the GEV model and GP model when the shape parameter is -0.4 under small and moderately large samples.

• The Korean Journal of Applied Statistics
• /
• v.26 no.5
• /
• pp.713-723
• /
• 2013

An objective Bayesian estimation procedure of the two-parameter Pareto distribution is presented under the reference prior and the noninformative prior. Bayesian estimators are obtained by Gibbs sampling. The steps to generate parameters in the Gibbs sampler are from the shape parameter of the gamma distribution and then the scale parameter by the adaptive rejection sampling algorism. A numerical study shows that the proposed objective Bayesian estimation outperforms other estimations in simulated bias and mean squared error.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2012.05a
• /
• pp.123-123
• /
• 2012

일강수량은 수공구조물 설계 및 수자원계획을 수립하기 위한 입력 자료로 이용된다. 일반적으로 수자원계획은 장기적인 목적을 가지고 수행되어지며, 장기간의 일강수량 자료를 필요로 한다. 하지만 장기간의 일강수량 자료의 획득의 어려움으로 단기간의 일강수량자료를 이용하여 모의한 장기간 강수자료를 이용하게 된다. 이처럼 수자원계획의 수립에 있어서 일강수량 모의기법의 성능은 수자원계획의 신뢰성 및 결과에 큰 영향을 준다. 일강수량 모의기법은 국내외적으로 매우 활발하게 이루어지고 있으며, 수자원계획 및 수공구조물 설계 외에도 매우 다양한 목적으로 활용되어 지고 있다. 일강수량을 모의기법 중 강수계열의 단기간의 기억(memory)을 활용한 Markov Chain 모형이 가장 일반적이지만, 기존 Markov Chain 모형을 통한 일강수량 모의는 극치강수량을 재현하기 어렵다는 문제점이 있다. 또한, 일강수량 모의 기법의 목적인 수자원계획 및 수공구조물 설계 등의 입력자료로 활용되어지기 위해서는 모의 결과가 유역내 지점별 공간 상관성을 재현함으로써 모형의 우수성과 자료결과의 신뢰성을 확보할 수 있어야 하겠다. 이러한 점에서 본 연구에서는 내삽에서 우수한 재현능력을 갖는 핵 밀도함수와 극치강수량 재현에 유리한 GPD분포의 특징을 함께 고려할 수 있는 불연속 Kernel-Pareto Distribution 기반에 공간상관성 재현 알고리즘을 결합한 일강수량모의기법을 개발하였다. 한강유역의 18개 강수지점에 대해서 기존 Gamma분포를 사용한 Markov Chain 모형과 본 연구에서 제안한 방법을 적용하여 모형을 평가해 보고자 한다. Gamma 분포기반 Markov Chain 모형의 경우 일강수량 모의 시 1차모멘트인 평균과 2-3차 모멘트 모두 효과적으로 재현하지 못하는 문제점이 나타났다. 그러나 본 연구에서 적용한 다지점 불연속 Kernel-Pareto 분포 모형은 강수계열의 평균적인 특성뿐만 아니라 표준편차 및 왜곡도의 경우에도 관측치의 통계특성을 매우 효과적으로 재현하며, 100년빈도 강수량 모의결과 기존 모의모형의 문제점을 보완할 수 있는 개선된 결과를 보여주었다. 본 연구에서 제시한 방법론은 유역내의 공간상관성을 재현하며, 평균 및 중간값 등 낮은 차수의 모멘트 등 일강수량 분포특성을 더욱 효과적으로 모의할 수 장점을 확인하였다.