• Journal of the Korean Data Analysis Society
• /
• 제20권6호
• /
• pp.2733-2746
• /
• 2018

The two inverse regression estimation methods, SIR and SAVE to estimate the central space are computationally easy and are widely used. However, SIR and SAVE may have poor performance in finite samples and need strong assumptions (linearity and/or constant covariance conditions) on predictors. The two non-parametric estimation methods, MAVE and dMAVE have much better performance for finite samples than SIR and SAVE. MAVE and dMAVE need no strong requirements on predictors or on the response variable. MAVE is focused on estimating the central mean subspace, but dMAVE is to estimate the central space. This paper explores and compares four methods to explain the dimension reduction. Each algorithm of these four methods is reviewed. Empirical study for simulated data shows that MAVE and dMAVE has relatively better performance than SIR and SAVE, regardless of not only different models but also different distributional assumptions of predictors. However, real data example with the binary response demonstrates that SAVE is better than other methods.

• International journal of advanced smart convergence
• /
• 제13권2호
• /
• pp.80-87
• /
• 2024

This paper focuses on improving accuracy in constrained computing settings by employing the ReLU (Rectified Linear Unit) activation function. The research conducted involves modifying parameters of the ReLU function and comparing performance in terms of accuracy and computational time. This paper specifically focuses on optimizing ReLU in the context of a Multilayer Perceptron (MLP) by determining the ideal values for features such as the dimensions of the linear layers and the learning rate (Ir). In order to optimize performance, the paper experiments with adjusting parameters like the size dimensions of linear layers and Ir values to induce the best performance outcomes. The experimental results show that using ReLU alone yielded the highest accuracy of 96.7% when the dimension sizes were 30 - 10 and the Ir value was 1. When combining ReLU with the Adam optimizer, the optimal model configuration had dimension sizes of 60 - 40 - 10, and an Ir value of 0.001, which resulted in the highest accuracy of 97.07%.

• 대한지역사회영양학회지
• /
• 제5권4호
• /
• pp.662-671
• /
• 2000

The purpose of this study was to analyze university students' perception of service quality attributes in residence hall foodservice. Questionnaires were hand delivered to 1,210 university students residing in a residence hall. A total of 1,011 was usable, resulting in an 83.6% response rate. The survey was conducted between October, 1998 and May, 1999. A statistical data analysis was completed using the SAS/Win 6.12 for Descriptive Analysis, $X^2$-test, t-test, ANOVA, Spearman Correlation, and Common factor Analysis, and Importance- Performance Analysis(IPA) was completed. Significant negative correlations between importance and performance were found with 'abundance of foods'(p < 0.001), 'discontent handling'(p < 0.001), and 'availability of new menus'(p < 0.01). factors including food, menu, sanitation, atmosphere, facilities, employee's attitude, and convenience were rearranged and a new dimension was created with the service. Employee's attitude and service factors received the highest scores in customer perception of importance and performance. Food, menu, and convenience factors were included in Quadrant A. female students had significantly higher importance mean scores than males, while males had significantly higher Performance mean scores. (Korean J Community Nutrition 5(4) : 662~671,2000)

• 전자공학회논문지
• /
• 제50권8호
• /
• pp.61-66
• /
• 2013

본 논문에서는 채널부호화 다중 안테나 시스템에서 공간다중화 전송된 신호들을 효과적으로 복조하기 위한 저연산 연판정 복조 다중 안테나 반복검출 기법을 제시한다. 반복 검출기법의 경우 우수한 성능에도 불구하고 연산량의 복잡성으로 수신단에 높은 복잡도를 요청하게 된다. 이러한 복잡도 감소를 위해 차원감소 소프트 검출 기법 (DRSD)과 모든 순서 순차적 간섭 제거(AOSIC) 기법을 사용한다. 이 기법의 경우 기존 기법들에 비해 반복검출 기법의 연산량의 복잡성을 줄일 수 있으며 향상된 성능을 얻을 수 있다.

• Membrane and Water Treatment
• /
• 제2권2호
• /
• pp.105-119
• /
• 2011

This study presents the tests of an Immersed Membrane BioReactor (IMBR) equipped with a draft tube and focuses on the influence of hydrodynamic conditions on membrane fouling in a pilot-scale using a hollow fiber membrane module of ZW-10 under ambient conditions. In this system, the cross-flow velocities across the membrane surface were induced by a cylindrical draft-tube. The relationship between cross-flow velocity and aeration strength and the influence of the cross-flow on fouling rate (under various hydrodynamic conditions) were investigated using Multi-Dimension Scaling (MDS) analysis. MDS technique is especially suitable for samples with many variables and has relatively few observations, as the data about Membrane Bio-Reactor (MBR) often is. Observations and variables are analyzed simultaneously. According to the results, a specialized form of MDS, CoPlot enables presentation of the results in a two dimensional space and when plotting variables ratio (output/input) rather than original data the efficient units can be visualized clearly. The results indicate that: (i) aeration plays an important role in IMBR performance; (ii) implementing the MDS approach with reference to the variables ratio is consequently useful to characterize performance changes for data classification.

• Journal of Information Technology Applications and Management
• /
• 제12권3호
• /
• pp.81-96
• /
• 2005

ASP (Application service provider) as a new type of information systems service has been popular recently and has taken a much attention of information systems professionals and researchers. One stream of the research on ASP performed by IS community dealt with utilization and benefits of ASP systems use. The other researchers argue the importance of ASP vendors' capability to provide users with appropriate services. Kern, Lacity and Willcock (2002) summarized the components of core competency of service providers based on their extensive case studies on application service providers. Also, Levina and Ross (2003) developed a value delivery process of application service Providers in their case study This study is developing Prepositions and an empirical model based on the theory of core competency from strategic management study. The dimension of core competency is composed of management competency, technical service competency and customer service competency. Each dimension is made up of competency items derived from previous studies and adapted by professionals in ASP industry. The validity of the dimensions and items in each dimension has been checked with factor analytic method for empirical test. The core competency dimensions are correlated with performance factors of application service providers including growth of revenue, asset and customer base.

• 한국멀티미디어학회논문지
• /
• 제21권8호
• /
• pp.952-959
• /
• 2018

Clustering is one of the most fundamental algorithms in machine learning. The performance of clustering is affected by the distribution of data, and when there are more data or more dimensions, the performance is degraded. For this reason, we use a stacked auto encoder, one of the deep learning algorithms, to reduce the dimension of data which generate a feature vector that best represents the input data. We use k-means, which is a famous algorithm, as a clustering. Sine the feature vector which reduced dimensions are also multi dimensional, we use the Euclidean distance as well as the cosine similarity to increase the performance which calculating the similarity between the center of the cluster and the data as a vector. A deep clustering networks combining a stacked auto encoder and k-means re-trains the networks when the k-means result changes. When re-training the networks, the loss function of the stacked auto encoder and the loss function of the k-means are combined to improve the performance and the stability of the network. Experiments of benchmark image ad document dataset empirically validated the power of the proposed algorithm.

• 한국산학기술학회논문지
• /
• 제6권3호
• /
• pp.249-254
• /
• 2005

본 논문은 얼굴인식에서 사용되고 있는 PCA/LDA 방식의 유사도 측정 방식에 따른 인식 성능을 비교 분석하였다. 총 14가지의 거리 척도를 ORL 얼굴 데이터베이스에 적용하였으며, PCA와 PCA/LDA로 나누어 성능 비교를 하였다. PCA의 경우에는 맨하튼 거리, Weighted SSE 거리의 인식률이 좋지만, PCA/LDA인 경우에는 Angle-based 거리, Modified SSE거리에 대한 인식률이 좋음이 확인되었다. 또한 PCA보다 PCA/LDA의 경우 유사도 비교 차원의 수를 줄이면서 높은 인식률을 유지할 수 있어, PCA/LDA와 Angle-based 거리 척도를 적용하여 얼굴인식을 할 경우 계산의 경제성과 인식률에서 높은 경쟁력을 갖출 수 있다.

• The Journal of Asian Finance, Economics and Business
• /
• 제7권2호
• /
• pp.249-260
• /
• 2020

This study aims to explore the role of Khan's engagement dimension previous study in mediating the relationship between work-life balance and in-role and extra-role performance of salesforce. To measure the data, a quantitative approach conducted through a survey of the sales force in the financial services industry sector in Makassar City, Indonesia. We determine the sample by using purposive sampling technique, and a total of 188 salesforce respondents. SEM-PLS with the second-order model was applied to test the measurement model and the mediator role of the engagement. The results show that the engagement dimension fully mediates the effect of work-life balance on both in-role and extra-role performance direct or indirect effect. the findings in this study also address the great concern of both individuals and organizations that the antecedents of the work-life balance variable are fundamental reasons for improving performance. The results of this study indicate that the work-life balance program positively influences engagement. Overall, we can conclude that commitment not only serves as a means for employees to fulfill formal job requirements. This can also encourage them to take action that goes beyond the precise requirements of their work by presenting them physically, emotionally, and cognitively in the work environment.

• 임상간호연구
• /
• 제14권1호
• /
• pp.71-84
• /
• 2008

Purpose: The purpose of this study was to develop a performance tool to evaluate operating room nurses. Method: In the first stage, 59 preliminary items were selected after two focus group meetings were carried out using as background a reference review. At this stage 59 items were developed. Next, the selected preliminary items were modified by an expert group of nurses with over 10 years experience, nurses holding doctoral degrees and professors in nursing. In the third and fourth stage, these items were analyzed by the expert group to determine validity and tested by hospital nurses with over 10 years experience to determine reliability Results: For the final tool, the 59 items, 47 items for the performance evaluation dimension and 12 items for the competence evaluation dimension were found to be valid and reliable. Cronbach's $\alpha$ was .9622, a promising alpha value for the total instrument. Conclusion: This tool is an efficient performance evaluation instrument providing meaningful feedback for operating room nurses.