• Communications for Statistical Applications and Methods
• /
• v.28 no.3
• /
• pp.267-279
• /
• 2021

A regression with multi-dimensional responses is quite common nowadays in the so-called big data era. In such regression, to relieve the curse of dimension due to high-dimension of responses, the dimension reduction of predictors is essential in analysis. Sufficient dimension reduction provides effective tools for the reduction, but there are few sufficient dimension reduction methodologies for multivariate regression. To fill this gap, we newly propose two fused slice-based inverse regression methods. The proposed approaches are robust to the numbers of clusters or slices and improve the estimation results over existing methods by fusing many kernel matrices. Numerical studies are presented and are compared with existing methods. Real data analysis confirms practical usefulness of the proposed methods.

• Communications of the Korean Mathematical Society
• /
• v.19 no.3
• /
• pp.469-475
• /
• 2004

We generalize S. Ikeda's results for perturbed cantor sets showing how we get the dimensions for deformed self-similar sets.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2005.03a
• /
• pp.854-861
• /
• 2005

Studied accuracy and practical use possibility of joint measurement that using 3D laser scanner to rock slope. Measured joint of Rock slope and comparison applied 3 dimension laser scanner and clinometer. 3D laser scanning system preserves on computer calculating to 3 dimension coordinate scaning laser to object. and according to laser measurement method of interior, produce correct vector value from charge-coupled device(CCD) or laser reciver and telegram register and time measuring equipment. Create of object x, y, z point coordinates to 3 dimension space of computer. Such 3 dimension point datum (Point Clouds) forms relocate position informations that exist to practical space to computer space. Practical numerical values related between each other. Compared joint distribution and direction that measured by laser scanner and clinometer. By the result, Distribution of joint projected almost equally. Could get more joint datas by measurement of 3 dimension scanner than measured by clinometer. Therefore, There is effect that objectification of rock slope investigation data, shortening of investigation periods, investigation reduction of cost. could know that it is very effective method in joint measuring.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2009.03a
• /
• pp.123-131
• /
• 2009

The study decided the improvement depth of soft ground of deep mixing method through 2 and 3 dimension finite element method and following results were acquired. 1. 2 dimension analysis shows settlement 10% more estimated than 3 dimension analysis. 2. When the rate of replacement is under 5%, the settlement sharply increased. 3. The most economical design for the levee was decided 3.0m for width direction, 6.0m for length direction and 8.0m for improvement depth. 4. When the soft ground is developed through deep mixing method, the decision of improvement should be decided through 3 dimension analysis than 2 dimension analysis.

• IE interfaces
• /
• v.16 no.3
• /
• pp.344-351
• /
• 2003

Critical dimension is one of the most important characteristics of up-to-date integrated circuit devices. Hence, critical dimension control in a semiconductor wafer fabrication process is inevitable in order to achieve optimum device yield as well as electrically specified functions. Currently, in complex semiconductor wafer fabrication processes, statistical methodologies such as Shewhart-type control charts become crucial tools for practitioners. Meanwhile, given a critical dimension sampling plan, the analysis of variance technique can be more effective to investigating critical dimension variation, especially for on-chip and on-wafer variation. In this paper, relating to a typical sampling plan, linear statistical models are presented for the analysis of critical dimension variation. A case study is illustrated regarding a semiconductor wafer fabrication process.

• Journal of the Korean Home Economics Association
• /
• v.26 no.3
• /
• pp.1-12
• /
• 1988

The objectives of the study were two folds. The first objective was to determine the dimensions of the evaluative criteria of various clothing items (underwear, pajamas, jeans, blouse, two-piece, coat). The second objective was to compare the importance of the dimensions according to the clothing items and the socioeconomic status of the subjects. The questionnaires were administered to college female students living in Seoul. Principal component factor analysis with varimax rotation and ANOVA were used for the analysis. The results were as follows; 1) The evaluative criteria dimensions were found to be different according to clothing items. (1) In underwear, pajamas, jeans, evaluative criteria were classified into Aesthetic dimension, economic dimension and Functional dimension. (2) In blouse, two-piece, coat, evaluative criteria were classified into Aesthetic dimension and practical dimension. 2) there were partially significant differences in placing importance on each evaluative criteria dimension between socio-economic groups. (1) In jeans, there was a significant difference in placing importance on Aesthetic dimension between socioeconomic status groups. (2) In blouse and two-piece there was a significant difference in placing importance on Practical dimension between socioeconomic status groups.

• Journal of Agricultural Extension & Community Development
• /
• v.9 no.2
• /
• pp.215-231
• /
• 2002

The objectives of this study were (1) to measure the level of satisfaction on rural for local public service. (2) and analyse variables related to the public service. Reviewing the literature and related studies for theoretical framework, Six criteria characterize public service : (1) responsiveness. (2) convenience, (3) speed, (4) correctness, (5) pleasantness, (6) equality, which are reflected in the instrument of this study. By the results of analysis, two dimensions (convenience, responsiveness) and total satisfaction are related to their residential lengths of the same rural area. For the total public service satisfaction gender, age, educational level, income level, occupation, and residential years explained in 13.3% For the convenience dimension their variables accounted for 22.4%. For responsiveness dimension their variables interpreted in 23.2%. For speed and correctness dimension their variables explained in 21.6%. For pleasantness dimension their variables accounted for local public services in 15.3%. For equality dimension their variables explained in 16.2%.

• Journal of Korea Technology Innovation Society
• /
• v.3 no.2
• /
• pp.1-13
• /
• 2000

This paper maps the overall features of technology valuation through a conceptional framework. The framework is composed of 4 dimensions such as basic, compositional, technical and behavioral dimension. At basic dimension, what is value and what is technology are discussed. The valuation of technology or the valuation of other assets are compared at the compositional dimension. The techniques of the valuation of technology and its difference with the valuation methods of other assets are examined at the technical dimension. The effectiveness, possibility and error of valuation are discussed at the behavioral dimension.

• Speech Sciences
• /
• v.3
• /
• pp.148-155
• /
• 1998

As an application of dimensional analysis in the theory of chaos and fractals, we studied and estimated the information dimension for various phonemes. By constructing phase-space vectors from the time-series speech signals, we calculated the natural measure and the Shannon's information from the trajectories. The information dimension was finally obtained as the slope of the plot of the information versus space division order. The information dimension showed that it is so sensitive to the waveform and time delay. By averaging over frames for various phonemes, we found the information dimension ranges from 1.2 to 1.4.

• Journal of Korea Water Resources Association
• /
• v.31 no.5
• /
• pp.587-597
• /
• 1998

Researchers have suggested that the fractal dimension of the stream length is uniform in all the streams of the basin and the estimates of the fractal dimension are in between 1.09 and 1.13 which may be considerably large values. In this study, the fractal dimension for the Ieemokjung subbasin streams in the Pyungchang River basin which is one of the IHP representative basins in Korea are estimated for each stream order using three scale maps of a 1/50,000, 1/25,000, and 1/5,000. As a result, the fractal dimension of the stream length is different by stream order and the fractal dimension of all streams shows a lower value in comparison to that of the previous studies. As a result of the fractal dimension estimation for the Ieemokjung subbasin streams, we found that the fractal dimension of the stream length shows different estimates in stream orders. The fractal dimension of 1st and 2nd order stream is 1.033, and the fractal dimension of 3rd and 4th order stream is 1.014. This result is different from the previous studies that the fractal dimension of the stream length is uniform in all streams of the basin. The fractal dimension for a whole stream length is about 1.027. Therefore, the previous estimates of 1.09 and 1.13 suggested as the fractal dimension of the stream length may be overestimated in comparison with estimated value in this study.