• ETRI Journal
• /
• v.27 no.6
• /
• pp.747-758
• /
• 2005

The problem of selection among competing models has been a fundamental issue in statistical data analysis. Good fits to data can be misleading since they can result from properties of the model that have nothing to do with it being a close approximation to the source distribution of interest (for example, overfitting). In this study we focus on the preference among models from a family of polynomial regressors. Three decades of research has spawned a number of plausible techniques for the selection of models, namely, Akaike's Finite Prediction Error (FPE) and Information Criterion (AIC), Schwartz's criterion (SCH), Generalized Cross Validation (GCV), Wallace's Minimum Message Length (MML), Minimum Description Length (MDL), and Vapnik's Structural Risk Minimization (SRM). The fundamental similarity between all these principles is their attempt to define an appropriate balance between the complexity of models and their ability to explain the data. This paper presents an empirical study of the above principles in the context of model selection, where the models under consideration are univariate polynomials. The paper includes a detailed empirical evaluation of the model selection methods on six target functions, with varying sample sizes and added Gaussian noise. The results from the study appear to provide strong evidence in support of the MML- and SRM- based methods over the other standard approaches (FPE, AIC, SCH and GCV).

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.3B
• /
• pp.213-224
• /
• 2009

Most of existing routing methods in wireless sensor networks to counter the local eavesdropping-based packet-tracing deal with a single asset and suffer from the packet-delivery latency as they prefer to take a separate path of many hops for each packet being sent. Recently, the author proposed a routing method, GSLP-w(GPSR-based Source-Location Privacy with crew size w), that enhances location privacy of the packet-originating node(i.e., active source) in the presence of multiple assets, yet taking a path of not too long. In this paper, we present a refined routing(i.e., next-hop selection) procedure of it and empirically study privacy strength and delivery latency with varying the crew size w(i.e., the number of packets being sent per path). It turns out that GSLP-w offers the best privacy strength when the number of packets being sent per path is randomly chosen from the range [$1,h_{s-b}/4$] and that further improvements on the privacy are achieved by increasing the random walk length TTLrw or the probability prw that goes into random walk(where, $h_{s-b}$ is the number of hops of the shortest path between packet-originating node s and sink b).

• The Research Journal of the Costume Culture
• /
• v.10 no.1
• /
• pp.89-101
• /
• 2002

The purpose of this study was to investigate the effect of demographics and general clothing buying behavior of college male students in China. A questionnaire was developed to measure clothing purchasing motives, fashion information sources, evaluation criteria of apparel product, stores selection criteria, and general clothing buying behavior. The questionnaire was administered to 300 college male students in Dandong of China. The data was analyzed using percentage, frequency, mean, factor analysis, and ANOVA, duncan test. The results of the study were as fellows: 1. The clothing purchasing motives of consumers were significantly different depending on demographics. 2. The fashion information sources of consumers were significantly different depending on subjects, pocket money, and level of life. 3. The evaluation criteria of apparel product of consumers were significantly different depending on parent's jobs. 4. The stores selection criteria of consumers were significantly different depending on parent's school career, mother's jobs, household income. 5. Apparel buying places of consumers were in order of ; department stores, speciality stores, natural market. The buying frequency of apparel showed the most three times a year. The consumers were influenced by friends for apparel purchasing, and the method of payment mainly cash. 6. About purchase experience of the imported clothing, the 67% consumers answered 'yes', and the country of origin of the imported clothing were in order of ; Japan, Italy, France, Korea. The imported clothing satisfaction level of consumers were significantly different depending on demographics.

• Journal of the Korean Society of Safety
• /
• v.27 no.6
• /
• pp.192-204
• /
• 2012

For a severe accident of nuclear power plant, an approach to estimation of the radiological source term using a severe accident code(MELCOR) has been proposed. Although the MELCOR code has a capability to estimate the radiological source term, it has been hardly utilized for the radiological consequence analysis mainly due to a lack of understanding on the relevant function employed in MELCOR and severe accident phenomena. In order to estimate the severe accident source term to be linked with the radiological consequence analysis, this study proposes 4-step procedure: (1) selection of plant condition leading to a severe accident(i.e., accident sequence), (2) analysis of the relevant severe accident code, (3) investigation of the code analysis results and post-processing, and (4) generation of radiological source term information for the consequence analysis. The feasibility study of the present approach to an early containment failure sequence caused by a fast station blackout(SBO) of a reference plant (OPR-1000), showed that while the MELCOR code has an integrated capability for severe accident and source term analysis, it has a large degree of uncertainty in quantifying the radiological source term. Key insights obtained from the present study were: (1) key parameters employed in a typical code for the consequence analysis(i.e., MACCS) could be generated by MELCOR code; (2) the MELOCR code simulation for an assessment of the selected accident sequence has a large degree of uncertainty in determining the accident scenario and severe accident phenomena; and (3) the generation of source term information for the consequence analysis relies on an expert opinion in both areas of severe accident analysis and consequence analysis. Nevertheless, the MELCOR code had a great advantage in estimating the radiological source term such as reflection of the current state of art in the area of severe accident and radiological source term.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.47 no.11
• /
• pp.24-29
• /
• 2010

Selection-type relaying protocol, which is one of cooperative relaying protocols, provides low decoding complexity and improved system performance due to selection diversity. In this paper, we deal with linear precoding technique that minimize the error probability of cooperative MIMO system. Under the assumption that full channel state information is available at whole nodes, linear source and relay precoders, which minimize mean squared error of the estimated symbol vector, are proposed. Moreover, unlikely to the conventional selection-type relaying protocol using a fixed threshold signal-to-noise-ratio, new transmission link selection algorithm selects direct link or relay link as a transmission link, is introduced. Simulation results show that the proposed linear precoder with the transmission link selection algorithm outperforms the conventional precoders for two-hop relaying protocols or selection-type relaying protocols.

• Journal of the Korean Society of Costume
• /
• v.49
• /
• pp.125-136
• /
• 1999

The present study was conducted to analyze(1) if there are any differences in jean's brand royalty according to the age·sex·educational background and income of consumers (2) the criteria for store choice by jean's brand royalty of consumers (3) the use of information souces by jean's brand royalty of consumers. The research method employed was normative-descriptive survey and questionnaires were main instrument to gather research materials. The survey was implemented from July 1st 1998 to the schools and two colleges and those who work at a comuter-related company and at a trade company. The data gathered was analyzed by mean standard deviation t-test ANOVA and Duncan -multiple range test. The results are summarized as follows: (1) 58.6% of the respondents were categorized as the brand loyalty group that was bigger than the non-brand loyalty group (41.4%) People in their late 2-'s were inclined to have the highest brand loyalty for jeans People in university and above had relatively higher brand royalty than to people with a high school or lower (2) Brand royalty group paid more attention on the quality on the quality of the fabric use; the trands of the products: brand name of the product: the distance of the store location : and the comfortable atmosphere of the store. Pleasant and kind services of the salesperson was considered as more important factor for store selection by the non-brand royalty group than by brand royalty group. (3) Almost all the information sources were used more often by brand royalty group than non-brand royalty group. For both groups their own experiences is most important information sources. The results shows that consumers have high brand loyalty for jeans product and also have many differences not only in the usage of information sources but also in the store selection criteria accorrding to the level of jeans brand loyalty . Therefore those industries carrying jeans should these points on their marketing strategies.

• Journal of the Korean Society of Radiology
• /
• v.13 no.6
• /
• pp.865-872
• /
• 2019

The purpose of this study was to analyze the correlation between radiation information sources and major selection satisfaction in radiologic science students of Korean and American. The subjects of this study were first-grade and second-grade at one university in Chungbuk, Korea and one university in Texas, USA. The survey study investigated from May 7, 2019 to May 28, 2019. As a result, first, these college students in both countries used the internet the most as them, but the percentage of information collected through textbooks was significantly higher among American college students. Second, Korean college students had a higher level of reliability in information sources than American college students in the order of internet, newspapers, and social network service (SNS), except government prints. Third, American college students had significantly higher satisfaction in major selection than Korean college students. Fourth, Korean college students were helped to select majors through various information sources, especially, the correlation coefficient between SNS was significantly higher.

• Korean Journal of Cognitive Science
• /
• v.13 no.2
• /
• pp.25-36
• /
• 2002

Two experiments were performed to explore the effects of the types of property and the tasks on the pattern of property inference in a source selection task in which the source objects were to be selected. animals that were globally similar to the target animal was selected as possible sources when the target properties were anatomical. However animals that were strongly associated with the target property were selected when the target properties were about ability In a passive inference task where premises were given. the global similarity between the source objects and the target object differently affected the confidence of the conclusion depending on the types of the target property: The similarity between the source and the target affected the degree of confidence when the target properties were anatomical ones, but did not affect when the target properties were about ability. The results suggested that participants seemed to have primitive understanding of the relevance of sources to the target properties, but did not. spontaneously seek or use the relevant information.

• The KIPS Transactions:PartD
• /
• v.15D no.6
• /
• pp.793-802
• /
• 2008

The development methods which utilize OSSs have been tried as new alternative to solve limits of the previous software developments such as the quality of software, and time and cost of developments. Especially, small and medium companies are suffering from difficulty in applying large-scale development methodology whenever they develop softwares. Therefore, in the current situation that demand for small-scale development methodology is increasing, the methods of utilizing OSSs can become an efficient way to save costs and reduce a development period. Accordingly, analysis of open-source is no fixed procedure or method to utilize open-source for software developments in the field. In this thesis, to solve such problems, we propose the procedures and methods for identifying and selecting suitable open-source, and effective methods for improvement and integration through least modification on the basis of synthesis of existing researches and experiences in development projects. For selection of OSSs, we did identify correct requirements for the software to be developed, investigate the open-source just matching with such requirements, draw a candidate index, establish assessment criteria, and the consequently present a method to select OSSs. And, we deduced the validity and improvement of each detailed activity from practical application to an actual project and assessment.

• Journal of Broadcast Engineering
• /
• v.21 no.1
• /
• pp.105-108
• /
• 2016

In this paper, a cooperative jamming network with multiple relays and multiple eavesdroppers is investigated. Among the relays, one best relay is selected to amplify and forward the signal to destination through two phases. To confuse eavesdroppers, the destination transmits a jamming signal in the first phase and the source transmits jamming signal in the second phase. Secrecy rate of this system is derived, and based on the available channel state information (CSI), relay selection schemes are proposed, respectively. Numerical results show that the performance of the proposed relay selection scheme outperforms than that of random relay selection scheme.