• Journal of Information Processing Systems
• /
• v.2 no.1
• /
• pp.23-27
• /
• 2006

This paper presents a feature selection technique based on distributional differences for efficient machine learning. Initial training data consists of data including many features and a target value. We classified them into positive and negative data based on the target value. We then divided the range of the feature values into 10 intervals and calculated the distribution of the intervals in each positive and negative data. Then, we selected the features and the intervals of the features for which the distributional differences are over a certain threshold. Using the selected intervals and features, we could obtain the reduced training data. In the experiments, we will show that the reduced training data can reduce the training time of the neural network by about 40%, and we can obtain more profit on simulated stock trading using the trained functions as well.

• Journal of Distribution Science
• /
• v.19 no.10
• /
• pp.75-86
• /
• 2021

Purpose: The purpose of this study is to address policy preparation and amendments on regulations in accommodation sharing for resource distribution by fostering better adjustment in a society, since previous studies are rarely investigated in those issues. After conduct exploratory research about laws and regulations of accommodation sharing, this study investigates how effective policy instruments improve trust in accommodation sharing and potential growth by investigating the perceptions of individuals and by applying policymaking procedures. Research design, data and methodology: The data is collected via online survey. Structural equation modeling with confirmatory factor analysis and non-recursive model with multiple regression analysis were applied. Results: The results of this study found that among proposed policy instruments, individuals perceive local ordinances, government publicizing and campaign, trust marks, taxation, penalties, and government controls are effective to build trust in accommodation sharing. Policies geared toward the majority of the public are more effective, while governments should establish a strategic approach as to which policies are introduced in public and which role the government plays in the departments. Conclusions: The results provide policy and managerial implications how to enhance distributional values of accommodation sharing economy with proper preparations and amendments of laws and regulations.

• Journal of Distribution Science
• /
• v.21 no.7
• /
• pp.21-31
• /
• 2023

Purpose: The purpose of this study is to investigate how to perceive the role of the slow city and provide policy and managerial implications on the necessity of the slow city management and distributional values based on perspectives of millennials and generation Z. This study examined i) how do millennials and generation Z perceive economic, environment, cultural, community, and quality of life factors on attitude toward the slow city? ii) how does attitude affect overall satisfaction and intention to recommend the slow city to other cities' residents? iii) how do millennials perspectives on proposed factors differ from generation Z? Research design, data and methodology: This study conducted an online survey and applied t-test, factor, ANOVA, and regression analysis. Results: This study found that effects of economic and quality of life factors on attitude toward the slow city showed significance in cases of millennials and generation Z, while effects of environment factor on overall attitude showed significance in the case of generation Z. Conclusions: Governments should foster how millennials and generation Z understand the meanings of the slow city to form a better attitude in a society and put efforts to build a better image and management of the slow city.

• Bulletin of the Korean Mathematical Society
• /
• v.48 no.2
• /
• pp.387-396
• /
• 2011

We consider a general functional equation with time variable which arises when we investigate regularity problems of some general functional equations. As a result we prove the regularity of the initial values of the solutions. Also as an application we prove the regularity of solutions of some classical functional equations and their distributional versions.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.28 no.4
• /
• pp.145-153
• /
• 2003

In this Paper, we examine the limiting distributional behaviour of extreme values of mixed Erlang random variables. We show that, in the finite mixture of Erlang distributions, the component distribution with an asymptotically dominant tail has a critical effect on the asymptotic extreme behavior of the mixture distribution and it converges to the Gumbel extreme-value distribution. Normalizing constants are also established. We apply this result to characterize the asymptotic distribution of maxima of sojourn times in M/M/s queuing system. We also show that Erlang mixtures with continuous mixing may converge to the Gumbel or Type II extreme-value distribution depending on their mixing distributions, considering two special cases of uniform mixing and exponential mixing.

• Communications for Statistical Applications and Methods
• /
• v.13 no.2
• /
• pp.343-358
• /
• 2006

The distributional properties of forecasts in an integer-valued time series model have not been discovered yet mainly because of the complexity arising from the binomial thinning operator. We propose two bootstrap methods to obtain nonparametric prediction intervals for an integer-valued autoregressive model : one accommodates the variation of estimating parameters and the other does not. Contrary to the results of the continuous ARMA model, we show that the latter is better than the former in forecasting the future values of the integer-valued autoregressive model.

• Journal of Plant Biology
• /
• v.13 no.1
• /
• pp.33-46
• /
• 1970

Pinus densiflora stands are common secondary forest communities on infertile soils in Korea. The stands are results of long severe past biotic pressure such as cutting, burning and grazing. These could be regarded as biotic climax in Korea. Because of their prevalent occurrence, relatively simple species and age composition, and their domestic economic importance, study of their distributional patterns may give some basic knowledge for better utilization of land resources in Korea. To detect distributional patterns and interspecific associations ten pine stands, each of which was homogenious with respect to topography and physiognomy, were subjectively selected from pine stands in Kyunggi Province near Seoul in 1969 and were made object of this study. Four contiguous systematic samples of count for trees, shrubs and seedlings from belt transects were collected from homogeneous areas within ten natural pine stands. The belt transect was 64m or 128m in length, and 1m, 2m or 4m in width. Basic units within the transect ranged from 64 to 256. The data from the contiguous transects were analysed in terms of multiple split-plot experiment. Departure from randomness of stem distribution, i.e., pattern, was tested in terms of variance mean ratio. For the detection of association between species, correlation coefficient was calculated for different block sizes. The values of ${\gamma}$ were tested by the usual t-test. Fine trees within one of the stands showed significant regular distribution through out the blocks. Within other eight stands pines were randomly distributed at basic unit with 4$\times$4m, 2$\times$2m, 2$\times$1m and 1$\times$1m. One significantly clumped distribution at basic unit 2$\times$2m, however, was observed from one of the pine stands. These randomly distributed groups were themselves significantly regularly distributed throughout the blocks for four pine stands. For the other four pine stands, in addition to the random distribution at the basie unit(the primary random group), randomly distributed groups with 32m dimension(the secondary random groups) were also observed. Both the primary and the secondary random groups were significantly regularly distributed at the rest of blocks. Pine seedlings were not distributed randomly thoughout the blocks. Within three of the ten stands they were contagiously distributed. Important shrub species underneath pines such as Querus serrata, Q. acutissima, Leapedeza intermedia, Rhododendron Yedoense var. poukhanenae, Juniperus utilis, Rhododendron mucronulatum var. ciliatum shnwed consistently similar distributional pattern with the pine at each stand. The shrub species pairs; Rhododendron Yedoense var. poukhanenae/Quercus serrata, Rhododendron mucronulatum var. ciliatum/Lespedeza intermedia were significantly negatively associated from 1m to 4m dimensional block sizes but became significantly positively associated from 8m sized blocks on. On the other hand the shrub species pairs; Lespedeza intermedia/Robinia Pseudoacacia, and Lespedeza bicolor var, japonica/Lespedeza intermedia were also significnalty negatively associated from 1m to 8m sized blocks but became significantly positively associated from 16m sized blocks on. The associational pattern between Rhododendron mucronul tun var. poukhanenae and Lespedeza intermedia was not consistent throughout the stands. In some stands negative associations were observed throughout the blocks except NS 32. From these observatons micro-edaphic variation within the pine stands seems not to be great enough to cause distributinal difference of pine trees within the ten pine stands. Among each species and pine seedings, however, the edaphic variation within the pine stands may be great enought to cause distributional variation.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.4
• /
• pp.959-967
• /
• 2016

Distributional assumptions on equity returns play a key role in valuation theories for derivative securities. Elberlein and Keller (1995) investigated the distributional form of compound returns and found that some of standard assumptions can not be justified. Instead, Generalized Hyperbolic (GH) distribution fit the empirical returns with high accuracy. Hu and Kercheval (2007) also show that the normal distribution leads to VaR (Value at Risk) estimate that significantly underestimate the realized empirical values, while the GH distributions do not. We consider saddlepoint approximations to estimate the VaR and the ES (Expected Shortfall) which frequently encountered in finance and insurance as measures of risk management. We supposed GH distributions instead of normal ones, as underlying distribution of linear portfolios. Simulation results show the saddlepoint approximations are very accurate than normal ones.

• The Journal of the Acoustical Society of Korea
• /
• v.35 no.2
• /
• pp.83-91
• /
• 2016

Music-similarity computation is crucial in developing music information retrieval systems for browsing and classification. This paper overviews the recently-proposed centroid-model based music retrieval method and applies the distributional similarity measures to the model for retrieval-performance evaluation. Probabilistic distance measures (also called divergence) compute the distance between two probability distributions in a certain sense. In this paper, we consider the alpha divergence in computing distance between two centroid models for music retrieval. The alpha divergence includes the widely-used Kullback-Leibler divergence and Bhattacharyya distance depending on the values of alpha. Experiments were conducted on both genre and singer datasets. We compare the music-retrieval performance of the distributional similarity with that of the vector distances. The experimental results show that the alpha divergence improves the performance of the centroid-model based music retrieval.

• Korean Journal of Plant Resources
• /
• v.13 no.3
• /
• pp.179-187
• /
• 2000

The achene morphology about 16 taxa of Korean Aster L. sensu lato were investigated to estimate its taxonomic values. The achene shapes were divided into four types; oblanceolate-oblong, obovate, oblong and obovate-oblong. The trichome shape on achene six types; uniseriate-conical, filiform, cylindrical, capitate type, long stalk capitate and globular. Their distributional features on upper part of achene four types; absent, sparse distribution of conical trichome, dense distribution of conical trichome and mixed distribution of conical and capitate trichome. The achene shapes and trichome characteristics were regarded to be a good characters in delimiting taxa because these did not differ among individuals in same taxa, but differ among the taxa. If Korean Aster L. sensu late were divided into Kalimeris, Heteropappus, Aster, Cymnaster, the capitate forms and mixed distribution of conical and capitate trichome were recognized as the good characters in delimting above section such as genus Kalimeris and Heteropappus, section Pseudocalimeris of Aster L. sensu stricto.