• Communications for Statistical Applications and Methods
• /
• v.7 no.2
• /
• pp.427-433
• /
• 2000

In this paper, we review several two-stage randomized response techniques for gathering self-report data when persons are asked sensitive question. Also efficiencies and privacy protections based on the two-stage randomized response procedures are compared. Finally, we find optimal parameter conditions.

• Communications for Statistical Applications and Methods
• /
• v.10 no.1
• /
• pp.247-256
• /
• 2003

In this study, we systemize the theoretical validity for applying RRM to three-stage cluster sampling method and derive the estimate and it's variance of sensitive parameter. We derive the minimum variance form under the optimal values of the subsample sizes when the costs are fixed. Under the some given precision, we obtain the optimal values of the subsample sizes and derive the minimum cost form by using them. We apply the three-stage cluster RRM to field survey and suggest some necessary points for practical use.

• Communications for Statistical Applications and Methods
• /
• v.18 no.4
• /
• pp.423-431
• /
• 2011

For procuring more sensitive information and estimating stratum target population proportion as well as an overall one form a sensitive population composed of several strata we suggest a two-stage stratified unrelated question model that uses stratified random sampling instead of simple random sampling in the two-stage unrelated question model by Kim et al. (1992) and extend it to the three-stage stratified unrelated question model. We also deal with the proportional and optimal allocation problems in each suggested model, compare the relative efficiency of the suggested two models, and show that the three-stage stratified unrelated question model is more efficient than the two-stage one in view of the variance.

• Proceedings of the SAREK Conference
• /
• 2009.06a
• /
• pp.1484-1490
• /
• 2009

Since the cooling performance of a $CO_2$ cooling cycle is varied significantly with a variation of refrigerant charge amount and outdoor temperature, the reliability of $CO_2$ system is down. In this study, the performance characteristics of three kinds of $CO_2$ systems were measured and analyzed by varying refrigerant charge amount and outdoor temperature so as to study the characteristics of variation with cycle option. The applied system options are the single-stage compression(1C-1E) system, two-stage compression with 1-EEV(2C-1E) system, and two-stage compression with 2-EEV(2C-2E) system. The performances of two-stage compression with 2-EEV system were less sensitive than those of other systems and the system operated safely and steadily for wide charge amount. The performance of the two-stage compression with 1-EEV(2C-1E) system was the most sensitive to the charge amount, and that of the single-stage compression(1C-1E) system varied a lot with outdoor temperature.

• BMB Reports
• /
• v.55 no.11
• /
• pp.553-558
• /
• 2022

Hepatocellular carcinoma (HCC) is dangerous cancer that often evades early detection because it is asymptomatic and an effective detection method is lacking. For people with chronic liver inflammation who are at high risk of developing HCC, a sensitive detection method for HCC is needed. In a meta-analysis of The Cancer Genome Atlas pan-cancer methylation database, we identified a CpG island in the USP44 promoter that is methylated specifically in HCC. We developed methylation-sensitive high-resolution melting (MS-HRM) analysis to measure the methylation levels of the USP promoter in cell-free DNA isolated from patients. Our MS-HRM assay correctly identified 40% of patients with early-stage HCC, whereas the α-fetoprotein test, which is currently used to detect HCC, correctly identified only 25% of early-stage HCC patients. These results demonstrate that USP44 MS-HRM analysis is suitable for HCC surveillance.

• Journal of the Korean Society of Tobacco Science
• /
• v.7 no.1
• /
• pp.25-31
• /
• 1985

For shortening the tobacco breeding cycle, seedlings with 6, 8 and II leaves of 2 flue.cured tobacco varieties, day. neutral type and photoperiod-sensitive type, were grown in controlled-environment room (CER), programmed for temperature of $18^{\circ}C$ and 8.hour day period of 30 klux, for 20, 30 and 40 days. All of plants of day.neutral type variety treated in CER, regardless of seedling stage and duration, flowered earlier than the untreated plants. In the 6.leaf seedlings stage of photoperiod-sensitive type variety, plants treated for 20 and 30 days in CER did not accelerate the flowering. the tobacco plants, treated with low temperature for 20 days at 8.leaf seedling stage, flowered earlier in comparison with. the other treatment.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.28 no.3
• /
• pp.286-294
• /
• 1992

In this paper, the corrosion fatigue crack propagation behavior of structure rolled steel (SWS 41C) was investigated by changing the thickness, and this experiment was done by the three point bending corrosion fatigue tester. The main results obtained are as follows: 1) As the thickness of specimen becomes thicker, the corrosion sensitivity to initial stage crack becomes some sensitive, and that the fatigue life becomes more sensitive. 2) The crack growth rate to initial stage crack (da/dN) was retarded as the thickness of specimen becomes thicker. But after initial stage crack, as the thickness of specimen is more thicker, da/dN is more rapid. 3) As the corrosion fatigue crack length grows, the accelerative factor of thick specimen (t=12mm) is more higher than that of thin specimen (t=6mm). 4) As the corrosion fatigue crack length grows, the corroson potential of both thick specimen and thin specimen becomes more less noble potential, however thick specimen (t=12mm) tends to more less noble potential than that of thin specimen(t=6mm).

• Communications for Statistical Applications and Methods
• /
• v.29 no.2
• /
• pp.251-261
• /
• 2022

As AI has a wide range of influence on human social life, issues of transparency and ethics of AI are emerging. In particular, it is widely known that due to the existence of historical bias in data against ethics or regulatory frameworks for fairness, trained AI models based on such biased data could also impose bias or unfairness against a certain sensitive group (e.g., non-white, women). Demographic disparities due to AI, which refer to socially unacceptable bias that an AI model favors certain groups (e.g., white, men) over other groups (e.g., black, women), have been observed frequently in many applications of AI and many studies have been done recently to develop AI algorithms which remove or alleviate such demographic disparities in trained AI models. In this paper, we consider a problem of using the information in the sensitive variable for fair prediction when using the sensitive variable as a part of input variables is prohibitive by laws or regulations to avoid unfairness. As a way of reflecting the information in the sensitive variable to prediction, we consider a two-stage procedure. First, the sensitive variable is fully included in the learning phase to have a prediction model depending on the sensitive variable, and then an imputed sensitive variable is used in the prediction phase. The aim of this paper is to evaluate this procedure by analyzing several benchmark datasets. We illustrate that using an imputed sensitive variable is helpful to improve prediction accuracies without hampering the degree of fairness much.

• KIISE Transactions on Computing Practices
• /
• v.21 no.11
• /
• pp.681-688
• /
• 2015

The k nearest neighbor (k-NN) graph construction is an important operation with many web-related applications, including collaborative filtering, similarity search, and many others in data mining and machine learning. Despite its many elegant properties, the brute force k-NN graph construction method has a computational complexity of $O(n^2)$, which is prohibitive for large scale data sets. Thus, (Key, Value)-based distributed framework, MapReduce, is gaining increasingly widespread use in Locality Sensitive Hashing which is efficient for high-dimension and sparse data. Based on the two-stage strategy, we engage the locality sensitive hashing technique to divide users into small subsets, and then calculate similarity between pairs in the small subsets using a brute force method on MapReduce. Specifically, generating a candidate group stage is important since brute-force calculation is performed in the following step. However, existing methods do not prevent large candidate groups. In this paper, we proposed an efficient algorithm for approximate k-NN graph construction by regrouping candidate groups. Experimental results show that our approach is more effective than existing methods in terms of graph accuracy and scan rate.

• Journal of the Korean Society of Tobacco Science
• /
• v.11 no.2
• /
• pp.127-134
• /
• 1989

Several growth characteristics of two ornamental tobacco species, Nicotiana sanderae and N. affinis, were investigated in this study. Also effect of temperature and daylength on the flowering of the tobacco plants were evaluated to obtain basic information on breeding and cultivation. 1. The plants were great in high temperature-long day at the early stage and in low temperature-short day at the late stage of plant growth, for both Nicotana species. At the early growth stage the leaf length N. sanderae was great in high temperature-long day, and that of N. affinis was great in high temperature-short day period, while at the late stage of the plant growth the leaf lengths were more significantly effected by the temperature rather than daylength. Leaf width and leaf shape index were less sensitive to the conditions. 2. For both of the species, the total number of tobacco leaves not much influenced by the temperature and daylength. 3. There were no significant differences for budding and flowering period between the two species, both of which were sensitive to temperature and daylength with more influence by daylength than temperature. 4. Number of floral stalks, number of flower and flowering period were not much influenced by temperature and daylength; however, N. affinis had 2 more floral stalks, 31 more flowers, and 6 day longer flowering period than N. sanderae.