• Journal of Korean Society for Quality Management
• /
• v.21 no.1
• /
• pp.161-170
• /
• 1993

This study attempts to present the linear regression analysis that involves more than one regressor variable, because regression analysis is the most widely used statistical technique for describing, predicting and estimating the relationships between given data. The model of multiple linear regression may be solved directly by the two linear programming methods, i.e., to minimize the sum of the absolute deviation (MSD) and to minimize the maximum deviation(MMD). In addition, some results was compared to each techniques for accuracy and tested to the validity of statistical meaning.

• Journal of Digital Convergence
• /
• v.16 no.1
• /
• pp.231-242
• /
• 2018

The efficient management of the Length of Stay(LOS) is important in hospital. It is import to reduce medical cost for patients and increase profitability for hospitals. In order to efficiently manage LOS, it is necessary to develop an artificial intelligence-based prediction model that supports hospitals in benchmarking and reduction ways of LOS. In order to develop a predictive model of LOS for acute stroke patients, acute stroke patients were extracted from 2013 and 2014 discharge injury patient data. The data for analysis was classified as 60% for training and 40% for evaluation. In the model development, we used traditional regression technique such as multiple regression analysis method, artificial intelligence technique such as interactive decision tree, neural network technique, and ensemble technique which integrate all. Model evaluation used Root ASE (Absolute error) index. They were 23.7 by multiple regression, 23.7 by interactive decision tree, 22.7 by neural network and 22.7 by esemble technique. As a result of model evaluation, neural network technique which is artificial intelligence technique was found to be superior. Through this, the utility of artificial intelligence has been proved in the development of the prediction LOS model. In the future, it is necessary to continue research on how to utilize artificial intelligence techniques more effectively in the development of LOS prediction model.

• International Journal of Highway Engineering
• /
• v.19 no.4
• /
• pp.1-7
• /
• 2017

PURPOSES : This study aims to improve complex modeling of multivariable, nonlinear, and overdispersion data with an artificial neural network that has been a problem in the civil and transport sectors. METHODS: Deep learning, which is a technique employing artificial neural networks, was applied for developing a large bus fuel consumption model as a case study. Estimation characteristics and accuracy were compared with the results of conventional multiple regression modeling. RESULTS : The deep learning model remarkably improved estimation accuracy of regression modeling, from R-sq. 18.76% to 72.22%. In addition, it was very flexible in reflecting large variance and complex relationships between dependent and independent variables. CONCLUSIONS : Deep learning could be a new alternative that solves general problems inherent in conventional statistical methods and it is highly promising in planning and optimizing issues in the civil and transport sectors. Extended applications to other fields, such as pavement management, structure safety, operation of intelligent transport systems, and traffic noise estimation are highly recommended.

• Computers and Concrete
• /
• v.13 no.1
• /
• pp.17-30
• /
• 2014

This research presents the effect of various ground pozzolanic materials in blended cement concrete on the strength and chloride penetration resistance. An experimental investigation dealing with concrete incorporating ground fly ash (GFA), ground bottom ash (GBA) and ground rice husk ash (GRHA). The concretes were mixed by replacing each pozzolan to Ordinary Portland cement at levels of 0%, 10%, 20% and 40% by weight of binder. Three different water to cement ratios (0.35, 0.48 and 0.62) were used and type F superplasticizer was added to keep the required slump. Compressive strength and chloride permeability were determined at the ages of 28, 60, and 90 days. Furthermore, using this experimental database, linear and nonlinear multiple regression techniques were developed to construct a mathematical model of chloride permeability in concretes. Experimental results indicated that the incorporation of GFA, GBA and GRHA as a partial cement replacement significantly improved compressive strength and chloride penetration resistance. The chloride penetration of blended concrete continuously decreases with an increase in pozzolan content up to 40% of cement replacement and yields the highest reduction in the chloride permeability. Compressive strength of concretes incorporating with these pozzolans was obviously higher than those of the control concretes at all ages. In addition, the nonlinear technique gives a higher degree of accuracy than the linear regression based on statistical parameters and provides fairly reasonable absolute fraction of variance ($R^2$) of 0.974 and 0.960 for the charge passed and chloride penetration depth, respectively.

• Journal of Biosystems Engineering
• /
• v.38 no.1
• /
• pp.64-71
• /
• 2013

Purpose: The purpose of this study was to select standing posture parameters that have a significant difference according to the severity of spinal deformity, and to develop a novel Cobb angle prediction model for adolescent girls with idiopathic scoliosis. Methods: Five normal adolescents girls with no history of musculoskeletal disorders, 13 mild scoliosis patients (Cobb angle: $10^{\circ}-25^{\circ}$), and 14 severe scoliosis patients (Cobb angle: $25^{\circ}-50^{\circ}$) participated in this study. Six infrared cameras (VICON) were used to acquire data and 35 standing parameters of scoliosis patients were extracted from previous studies. Using the ANOVA and post-hoc test, parameters that had significant differences were extracted. In addition, these standing posture parameters were utilized to develop a Cobb-angle prediction model through multiple regression analysis. Results: Twenty two of the parameters showed differences between at least two of the three groups and these parameters were used to develop the multi-linear regression model. This model showed a good agreement ($R^2$ = 0.92) between the predicted and the measured Cobb angle. Also, a blind study was performed using 5 random datasets that had not been used in the model and the errors were approximately $3.2{\pm}1.8$. Conclusions: In this study, we demonstrated the possibility of clinically predicting the Cobb angle using a non-invasive technique. Also, monitoring changes in patients with a progressive disease, such as scoliosis, will make possible to have determine the appropriate treatment and rehabilitation strategies without the need for radiation exposure.

• Journal of Korean Institute of Industrial Engineers
• /
• v.28 no.3
• /
• pp.331-343
• /
• 2002

This paper proposes a fuzzy rule-based model as a means to build usability models between emotional satisfaction and design variables of consumer products. Based on a subtractive clustering algorithm, this model obtains partially overlapping rules from existing data and builds multiple local models each of which has a form of a linear regression equation. The best subset procedure and cross validation technique are used to select appropriate input variables. The proposed technique was applied to the modeling of luxuriousness, balance, and attractiveness of office chairs. For comparison, regression models were built on the same data in two different ways; one using only potentially important variables selected by the design experts, and the other using all the design variables available. The results showed that the fuzzy rule-based model had a great benefit in terms of the number of variables included in the model. They also turned out to be adequate for predicting the usability of a new product. Better yet, the information on the product classes and their satisfaction levels can be obtained by interpreting the rules. The models, when combined with the information from the regression models, are expected to help the designers gain valuable insights in designing a new product.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.21 no.5
• /
• pp.83-93
• /
• 2017

This study proposed the new weighting values and fault points of condition assessment items for reinforced earth retaining walls based on the combination the inspection data and hybrid weighting technique. Utilizing the inspection data of 161 reinforced earth retaining walls, multi regression analysis and entropy technique were applied to gain the weighting values of condition assessment items. In addition, the weighting values by AHP technique was analyzed based on the opinion of experts. By appling hybrid weighting technique to the calculated weighting values obtained by the individual technique, the new weighting values of condition assessment items were proposed, and the fault points and fault indices of reinforced earth retaining walls were proposed. Results showed that the rank of the weighting value of the condition evaluation items was fluctuated according to the multiple regression analysis, AHP technique, and entropy technique. There was no duplication of the rank of the weighting value while the current weighting value was overlapped. Specially, in the rsults of multi regression analysis, two condition assessment items were occupied 70% of the total weights. When the proposed weighting values were applied to existing reinforced earth retaining wall of 161, 16 reinforced earth retaining walls showed the increased risk rank and 31 represented the decreased risk rank.

• Journal of the Korean Data and Information Science Society
• /
• v.16 no.3
• /
• pp.665-682
• /
• 2005

The sample reuse bootstrap technique has been successful to attract both applied and theoretical statisticians since its origination. In recent years a good deal of attention has been focused on the applications of bootstrap methods in regression analysis. It is easier but more accurate computation methods heavily depend on high-speed computers and warrant tough mathematical justification for their validity. It is now evident that the presence of multiple unusual observations could make a great deal of damage to the inferential procedure. We suspect that bootstrap methods may not be free from this problem. We at first present few examples in favour of our suspicion and propose a new method diagnostic-before-bootstrap method for regression purpose. The usefulness of our newly proposed method is investigated through few well-known examples and a Monte Carlo simulation under a variety of error and leverage structures.

• The Transactions of the Korean Institute of Electrical Engineers C
• /
• v.52 no.3
• /
• pp.137-143
• /
• 2003

The degree of contamination on outdoor insulator is ons of the most importance factor to determine the pollution level of outdoor insulation, and the sea salt is known as the most dangerous pollutant. As shown through the preceding study, the generation of salt pollutant and the pollution degree of outdoor insulator have a close relation with meteorological conditions, such as wind velocity, wind direction, precipitation and so fourth. So, in this paper, we made an investigation on the prediction method, a statistical estimation technique for equivalent salt deposit density of outdoor insulator with multiple linear regression analysis. From the results of the analysis, we proved the superiority of the prediction method in which the variables had a very close(about 0.9) correlation coefficient. And the results could be applied to establish the Pollution Prediction System for power utilities, and the system could provide an invaluable information for the design and maintenance of outdoor insulation system.

• Journal of Korea Society of Industrial Information Systems
• /
• v.17 no.5
• /
• pp.89-96
• /
• 2012

The Falling Weight Deflectormeter has been widely used in evaluating the structural adequacy of pavement structures. The deflections measured from the FWD are capable of estimating the stiffness of pavement layers and measuring the pavement responses in the pavement structure. The objective of paper is to develop the pavement response model using a partial least square regression technique based on the FWD deflection data. The partial least square regression method enables to solve the multicollinearity problem occurred in multiple regression model. It is also found that the pavement response model can be developed using the raw data when a partial least square regression was used.