• Journal of the Korean Data and Information Science Society
• /
• v.26 no.6
• /
• pp.1547-1555
• /
• 2015

We study the problem of nonignorable nonresponse in a two-way contingency table and there may be one or two missing categories. We describe a nonignorable nonresponse model for the analysis of two-way categorical table. One approach to analyze these data is to construct several tables (one complete and the others incomplete). There are nonidentifiable parameters in incomplete tables. We describe a hierarchical Bayesian model to analyze two-way categorical data. We use a nonignorable nonresponse model with Bayesian uncertainty analysis by placing priors in nonidentifiable parameters instead of a sensitivity analysis for nonidentifiable parameters. To reduce the effects of nonidentifiable parameters, we project the parameters to a lower dimensional space and we allow the reduced set of parameters to share a common distribution. We use the griddy Gibbs sampler to fit our models and compute DIC and BPP for model diagnostics. We illustrate our method using data from NHANES III data to obtain the finite population proportions.

• The Journal of Information Systems
• /
• v.28 no.4
• /
• pp.359-375
• /
• 2019

Purpose The purpose of this study is to investigate the nature of temporal non-use, namely, the type of temporal non-use and the dynamic relationships between temporary non-use and use by expanding the model of dynamic state transition of non-use which is proposed by the previous study. Design/methodology/approach This study adopted grounded theory methodology as a research methodology. Fifteen university students were interviewed for open coding and for selected coding which leads the research model ten students were interviewed. Findings Based on the coding, the types of use were divided into interactive use and one-way use, and the types of temporal use were classified into temporal non-use in which use is intended and temporal non-use in which use is not intended. The dynamic transition model is presented for each type of use and temporalnon-use, and an integrated model that combines them is presented. This study found the following four things: First, all non-use can be temporal Second, there are different types of temporal non-use. Third, the pattern of use is not just one, but new patterns will continue to emerge. Fourth, we found a change in the function of SNS.

• Journal of Korea Multimedia Society
• /
• v.10 no.12
• /
• pp.1663-1670
• /
• 2007

One-way hash chains are important cryptographic primitives and have been used as building blocks of various cryptographic applications. Advantages of one-way hash chains are their simplicity and efficiency for generation based on low-powered processors with short time. However, a drawback of one-way hash chains is their difficulty of control to compute interval values of one-way hash chains. That is, when hash values in one-way hash chain are used as encryption keys, if one hash value is compromised, then the attacker can compute other encryption keys from the compromised hash value. Therefore, direct use of one-way hash chains as encryption keys is limited to many cryptographic applications, such as pay per view system and DRM system. In this paper, we propose a new concept which is called interval hash chain using a hash function. In particular, proposed hash chains are made for only computing interval hash values by using two different one-way hash chains. The proposed scheme can be applied to contents encryption scheme for grading and partially usable contents in DRM system.

• Communications for Statistical Applications and Methods
• /
• v.14 no.3
• /
• pp.551-560
• /
• 2007

Kruskal and Wallis (1952) proposed typical nonparametric method in one-way layout problem. A special feature of this procedure is use of rank in mixed samples. In this paper, the new procedure based on placement as extension of the two sample placement tests described in Orban and Wolfe (1982) was proposed. Some critical values in small sample cases and comparative results of a Monte Carlo power study are presented.

• Proceedings of the KSME Conference
• /
• 2000.04a
• /
• pp.741-744
• /
• 2000

In general there occur warpage in one way long part. Warpage is caused by differential shrinkage-Orientation Effect, Volumetric Shrinkage Effect, Differential Cooling Effect -over the part. Deco-Top is located at the top of 29"TV set and it's shape is one way long$(626{\times}130mm)$. Material is used transparency ABS resin. So we can't design ribs in this part. And we use film gate to avoid weld line. In these reasons we must develop no ribs and no warpage product. In this study we use MOLDFLOW's software-MF/FLOW, MF/COOL, MF/WARP. Using MF/FLOW, set the flow balance and gate positioning. And we can set cooling channel layout and the optimum processing condition through MF/COOL and MF/WARP. In result we reduced trials and obtained good product.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.1
• /
• pp.245-254
• /
• 2016

Many surveys provide categorical data and there may be one or more missing categories. We describe a nonignorable nonresponse model for the analysis of two-way contingency tables from small areas. There are both item and unit nonresponse. One approach to analyze these data is to construct several tables corresponding to missing categories. We describe a hierarchical Bayesian model to analyze two-way categorical data from different areas. This allows a "borrowing of strength" of the data from larger areas to improve the reliability in the estimates of the model parameters corresponding to the small areas. Also we use a nonignorable nonresponse model with Bayesian uncertainty analysis by placing priors in nonidentifiable parameters instead of a sensitivity analysis for nonidentifiable parameters. We use the griddy Gibbs sampler to fit our models and compute DIC and BPP for model diagnostics. We illustrate our method using data from NHANES III data on thirteen states to obtain the finite population proportions.

• Journal of Environmental Policy
• /
• v.3 no.2
• /
• pp.89-112
• /
• 2004

This Study aims to demonstrate how much flood vulnerability in urban area changed for the past 32 years by using the panel model. At the same time, this study strives to determine the primary factors and to construct an effective counter-plan by means of empirical research. After selecting research hypotheses based on considerations of issues concerning causes for urban flooding, their relevance is put to the test by conducting empirical research in individual case locations. This research verifies the four research hypotheses by using one-way error component regression model. In conclusion, this research has shown that urban land use and local characteristics act as significant flood determinants, with forests acting to reduce flood dangers. Moreover, constructing embankments can no longer represent a reliable flood control policy. The changes in future flood control policies need to incorporate local characteristics and to minimize natural destruction, so that humans and nature can coexist through environmentally friendly flood management policies.

• The monthly packaging world
• /
• s.277
• /
• pp.72-75
• /
• 2016

• The monthly packaging world
• /
• s.278
• /
• pp.83-87
• /
• 2016

• Journal of Korean Library and Information Science Society
• /
• v.30 no.3
• /
• pp.427-456
• /
• 1999

This paper deals with the problems of information retrieval through full-test database which arise from both the deficiency of searching strategies or methods by information searcher and the difficulties of query representation, generation, extension, etc. In oder to solve these problems, we should use automatic retrieval instead of manual retrieval in the past. One of the ways to make the gap narrow between the terms by the writers and query by the searchers is that the query should be searched with the terms which the writers use. Thus, the preconditions which should be taken one accorded way to solve the problems are that all areas of information retrieval such as should taken one accorded way to solve the problems are that all areas of information retrieval such as contents analysis, information structure, query formation, query evaluation, etc. should be solved as a coherence way. We need to deal all the ares of automatic information retrieval for the efficiency of retrieval thought this paper is trying to solve the design of thesaurus browser. Thus, this paper shows the theoretical analyses about the form of information retrieval, automatic indexing, clustering technique, establishing and expressing thesaurus, and information retrieval technique. As the result of analyzing them, this paper shows us theoretical model, that is to say, the thesaurus browser by clustering algorithm. The result in the paper will be a theoretical basis on new retrieval algorithm.