• 제목/요약/키워드: MRA(Multiple Regression Analysis)

검색결과 20건 처리시간 0.019초

Evaluation of Barley Bran Sauce Aroma by Multiple Regression Analysis

  • Choi, Ung-Kyu
    • Food Science and Biotechnology
    • /
    • 제14권5호
    • /
    • pp.656-660
    • /
    • 2005
  • The relationship between the gas chromatographic (GC) patterns of sauce made of barley bran and ranked order in sensory analysis was investigated by multiple regression analysis (MRA). Most of the 42 barley bran sauce samples comprised about 34 peaks, in which the content of 9, 12-octadecanoic acid methyl ester was the highest, followed by those of 2-furanmethanol and 2-furancarboxaldehyde. It is difficult to estimate the aroma quality of barley bran sauce samples on the basis of only one peak. The 34 aroma compounds of the 42 samples were analyzed by an MRA model featuring six transformations. The most precise fit was calculated from the absolute value transformed with the root square of each peak, and the multiple determination coefficient showed that 91.6% of the variation in the sensory score could be explained on the basis of GC data.

Shear strength of steel beams with trapezoidal corrugated webs using regression analysis

  • Barakat, Samer;Mansouri, Ahmad Al;Altoubat, Salah
    • Steel and Composite Structures
    • /
    • 제18권3호
    • /
    • pp.757-773
    • /
    • 2015
  • This work attempts to implement multiple regression analysis (MRA) for modeling and predicting the shear buckling strength of a steel beam with corrugated web. It was recognized from theoretical and experimental results that the shear buckling strength of a steel beam with corrugated web is complicated and affected by several parameters. A model that predicts the shear strength of a steel beam with corrugated web with reasonable accuracy was sought. To that end, a total of 93 experimental data points were collected from different sources. Then mathematical models for the key response parameter (shear buckling strength of a steel beam with corrugated web) were established via MRA in terms of different input geometric, loading and materials parameters. Results indicate that, with a minimal processing of data, MRA could accurately predict the shear buckling strength of a steel beam with corrugated web within a 95% confidence interval, having an $R^2$ value of 0.93 and passing the F- and t-tests.

GA-optimized Support Vector Regression for an Improved Emotional State Estimation Model

  • Ahn, Hyunchul;Kim, Seongjin;Kim, Jae Kyeong
    • KSII Transactions on Internet and Information Systems (TIIS)
    • /
    • 제8권6호
    • /
    • pp.2056-2069
    • /
    • 2014
  • In order to implement interactive and personalized Web services properly, it is necessary to understand the tangible and intangible responses of the users and to recognize their emotional states. Recently, some studies have attempted to build emotional state estimation models based on facial expressions. Most of these studies have applied multiple regression analysis (MRA), artificial neural network (ANN), and support vector regression (SVR) as the prediction algorithm, but the prediction accuracies have been relatively low. In order to improve the prediction performance of the emotion prediction model, we propose a novel SVR model that is optimized using a genetic algorithm (GA). Our proposed algorithm-GASVR-is designed to optimize the kernel parameters and the feature subsets of SVRs in order to predict the levels of two aspects-valence and arousal-of the emotions of the users. In order to validate the usefulness of GASVR, we collected a real-world data set of facial responses and emotional states via a survey. We applied GASVR and other algorithms including MRA, ANN, and conventional SVR to the data set. Finally, we found that GASVR outperformed all of the comparative algorithms in the prediction of the valence and arousal levels.

머신러닝 기법과 계측 모니터링 데이터를 이용한 광안대교 신축거동 모델링 (Modeling on Expansion Behavior of Gwangan Bridge using Machine Learning Techniques and Structural Monitoring Data)

  • 박지현;신성우;김수용
    • 한국안전학회지
    • /
    • 제33권6호
    • /
    • pp.42-49
    • /
    • 2018
  • In this study, we have developed a prediction model for expansion and contraction behaviors of expansion joint in Gwangan Bridge using machine learning techniques and bridge monitoring data. In the development of the prediction model, two famous machine learning techniques, multiple regression analysis (MRA) and artificial neural network (ANN), were employed. Structural monitoring data obtained from bridge monitoring system of Gwangan Bridge were used to train and validate the developed models. From the results, it was found that the expansion and contraction behaviors predicted by the developed models are matched well with actual expansion and contraction behaviors of Gwangan Bridge. Therefore, it can be concluded that both MRA and ANN models can be used to predict the expansion and contraction behaviors of Gwangan Bridge without actual measurements of those behaviors.

Factors Affecting Employee Loyalty in Railway Rolling Stock Maintenance Companies in Thailand

  • LIEOPHAIROT, Ratchaphong;ROJNIRUTTIKUL, Nuttawut
    • The Journal of Asian Finance, Economics and Business
    • /
    • 제9권10호
    • /
    • pp.115-127
    • /
    • 2022
  • The study's goal was to investigate the levels of employee loyalty (EL) in two Thai railway rolling stock maintenance (RRSM) companies. Simple random sampling was used to obtain a final sample of 118 individuals from October 2021 through December 2021. The research instrument was a questionnaire with an expert IOC value between 0.67 to 1.00 and a questionnaire reliability Alpha (𝛼) average value of 0.82. Descriptive statistics included the mean and standard deviation (SD). SPSS for Windows Version 21 and Multiple Regression Analysis (MRA) was used for the analysis. Results showed that the 118 employee's overall perceptions of their RRSM employers' motivating factors, human resource management, satisfaction, and loyalty were high. HRM's performance evaluation had the most significant overall influence on EL. Moreover, from the analysis of the five EL questionnaire items, the most influential item was the employee's income as a contributing factor to their EL. This was followed by the suitability of their work. Also, it seems the employees had a high level of loyalty to their firms even if a better offer of more money was made. They also indicated a high level of pride in their respective firms.

2017 International Ice Hockey Federation World Championship의 승리 결정요인 분석 (Factors Contributing to Winning in Ice Hockey: Analysis of 2017 Ice Hockey World Championship)

  • 이주성;김혜영;김채은;프라밧;문제헌
    • 한국체육학회지인문사회과학편
    • /
    • 제57권4호
    • /
    • pp.387-394
    • /
    • 2018
  • 본 연구의 목적은 국제아이스하키연맹(IIHF)이 개최하는 대회에서 승리 팀을 결정하는 주요 변인들을 파악하여 전략 및 전술 수립에 필요한 정보를 제공하는 것이다. 2017 IIHF World Championship 1부 리그 14개 팀의 예선 및 본선을 포함한 64개 경기의 기록지를 분석대상으로 하였다. 분석변인은 save, shot on goal, penalty in minute, time on power play, power play goal, face off win의 비율, 승패 간 로지스틱 회귀분석, 중다회귀분석, 주성분분석을 수행하였다. 로지스틱 회귀분석 결과 승리와 관련이 있는 변인은 shot on goal(p<.001)와 face off win(p<.001)이고 penalty in minute(p<.01)과 time on power play(p<.01)는 부정적인 영향을 미친다. 중다회귀분석에 의하여 산출한 승패 비율과 각 변인과의 상관분석에서는 save(p<.01), face off win(p<.001)가 정적인 상관관계이고 penalty in minute(p<.001)이 부정적인 상관관계이다. 주성분분석 결과에서는 승리한 팀의 경우 페널티 요인, 공격 요인, 수비 요인으로 구성되는 반면 패배한 팀에서는 페널티 요인을 제외하고 공격과 수비 요인이 혼합되어 구성되었다. 따라서 최상위 팀이 참가하는 아이스하키 경기에서 승리하기 위해서는 페널티를 받지 않는 내에서 거친 플레이가 이루어져야 하고 face off win 비율을 높일 수 있는 방안이 마련되어야 할 것이다.

2017년 한국 가막만 북서내만해역 빈산소수괴 발생의 특성 (Characteristics of Hypoxic Water Mass Occurrence in the Northwestern Gamak Bay, Korea, 2017)

  • 정희호;최상덕;조현서
    • 해양환경안전학회지
    • /
    • 제27권6호
    • /
    • pp.708-720
    • /
    • 2021
  • 가막만 북서내만해역에서 매년 여름 발생하는 빈산소수괴는 해양환경에 악영향을 미쳐왔다. 따라서, 본 연구는 빈산소수괴 발생시기의 집중적인 현장조사 결과를 바탕으로 다중회귀분석(MRA)을 이용하여 빈산소수괴의 종합적인 발생 메커니즘을 밝혀내고, 그 주요인에 따른 빈산소수괴 관리방안의 방향성을 모색하였다. 그 결과, 2017년 첫 빈산소수괴는 6월 26일에 발생하였으며, 기상 조건에 의한 수온약층의 형성과 퇴적된 유기물의 영향으로 형성되었다. 이어 7월 12일에는 강우량의 증가에 의한 염분약층의 형성으로, 조사 시기 중 빈산소수괴가 수직 및 수평적으로 가장 크게 확장되었다. 그리고, 8월 8일에는 소량의 강우로 빈산소수괴가 크게 약화되었으며, 이때 주 요인은 Chlorophyll-a 농도 증가(식물플랑크톤 증식)과 퇴적된 유기물이었다. 그리고, 약 1주일 후인 8월 16일에는 많은 강우량에 기인한 매우 안정된 염분약층과 Chlorophyll-a 농도 증가(식물플랑크톤 증식)에 의해 크게 확장된 빈산소수괴가 재발생하였다. 이후 9월 13일의 빈산소수괴 소멸시기에서는 빈산소수괴가 해저 면을 따라 얕게 확장되었으며, 퇴적된 유기물에 의해 주로 영향을 받은 것으로 나타났다. 이는 빈산소수괴 관리를 위해서는 퇴적된 유기물의 개선뿐만 아니라 성층의 완화 기술이 필요함을 암시하였다.

감정예측모형의 성과개선을 위한 Support Vector Regression 응용 (Application of Support Vector Regression for Improving the Performance of the Emotion Prediction Model)

  • 김성진;유은정;정민규;김재경;안현철
    • 지능정보연구
    • /
    • 제18권3호
    • /
    • pp.185-202
    • /
    • 2012
  • 오늘날 정보사회에서는 정보에 대한 가치를 인식하고, 이를 위한 정보의 활용과 수집이 중요해지고 있다. 얼굴 표정은 그림 하나가 수천개의 단어를 표현할 수 있듯이 수천 개의 정보를 지니고 있다. 이에 주목하여 최근 얼굴 표정을 통해 사람의 감정을 판단하여 지능형 서비스를 제공하기 위한 시도가 MIT Media Lab을 필두로 활발하게 이루어지고 있다. 전통적으로 기존 연구에서는 인공신경망, 중회귀분석 등의 기법을 통해 사람의 감정을 판단하는 연구가 이루어져 왔다. 하지만 중회귀모형은 예측 정확도가 떨어지고, 인공신경망은 성능은 뛰어나지만 기법 자체가 지닌 과적합화 문제로 인해 한계를 지닌다. 본 연구는 사람들의 자극에 대한 반응으로서 나타나는 얼굴 표정을 통해 감정을 추론해내는 지능형 모형을 개발하는 것을 목표로 한다. 기존 얼굴 표정을 통한 지능형 감정판단모형을 개선하기 위하여, Support Vector Regression(이하 SVR) 기법을 적용하는 새로운 모형을 제시한다. SVR은 기존 Support Vector Machine이 가진 뛰어난 예측 능력을 바탕으로, 회귀문제 영역을 해결하기 위해 확장된 것이다. 본 연구의 제안 모형의 목적은 사람의 얼굴 표정으로부터 쾌/불쾌 수준 그리고 몰입도를 판단할 수 있도록 설계되는 것이다. 모형 구축을 위해 사람들에게 적절한 자극영상을 제공했을 때 나타나는 얼굴 반응들을 수집했고, 이를 기반으로 얼굴 특징점을 도출 및 보정하였다. 이후 전처리 과정을 통해 통계적 유의변수를 추출 후 학습용과 검증용 데이터로 구분하여 SVR 모형을 통해 학습시키고, 평가되도록 하였다. 다수의 일반인들을 대상으로 수집된 실제 데이터셋을 기반으로 제안모형을 적용해 본 결과, 매우 우수한 예측 정확도를 보임을 확인할 수 있었다. 아울러, 중회귀분석이나 인공신경망 기법과 비교했을 때에도 본 연구에서 제안한 SVR 모형이 쾌/불쾌 수준 및 몰입도 모두에서 더 우수한 예측성과를 보임을 확인할 수 있었다. 이는 얼굴 표정에 기반한 감정판단모형으로서 SVR이 상당히 효과적인 수단이 될 수 있다는 점을 알 수 있었다.

Factors Influencing the Perception of the Selling Price of Luxury Apartments

  • NGUYEN, Huu Cuong;DO, Duc Tai
    • The Journal of Asian Finance, Economics and Business
    • /
    • 제7권5호
    • /
    • pp.185-194
    • /
    • 2020
  • The study aims to identify and measure factors affecting the perception of the selling price of luxury apartments in Hanoi. We conducted a questionnaire consisting of 29 observation variables with a 5-point Likert scale. Independent variables were measured from 1 "without effect" to 5 "strongly". Based on the desk review and results of interviews, a total of 500 questionnaires were sent to research participants for collection; 458 of them met standard and were subject to be analyzed. This study employs Cronbach's Alpha test, and regression model. The results of Exploratory Factor Analysis (EFA) and Multiple Regression Analysis (MRA) identify five main determinants influencing the perception of the selling price of luxury apartments in Hanoi, including Physical characteristics of a luxury apartment (PC); Location and position of an apartment (LP); Surrounding Area (SA); Quality of service provided by managers; (QS) and Demographics factor (DF). Based on the findings, some recommendations have been proposed to help the firm leaders design appropriate personnel policies for creating better price satisfaction for customers in the future. On this basis, the authors propose a number of recommendations to improve the quality of luxury apartments, thereby contributing to the development of the market for luxury apartments in Hanoi.

BEMS 데이터의 통계적 분석에 기반한 공조기 최적 예냉운전 모델 개발 (Developing Optimal Pre-Cooling Model Based on Statistical Analysis of BEMS Data in Air Handling Unit)

  • 최선규;곽노열;구상헌
    • 설비공학논문집
    • /
    • 제26권10호
    • /
    • pp.467-473
    • /
    • 2014
  • Since the operating conditions of HVAC systems are different from those for which they are designed, on-going commissioning is required to optimize the energy consumed and the environment in the building. This study presents a methodology to analyze operational data and its applications. A predicted operation model is to be produced through a statistical data analysis using multiple regressions in SPSS. In this model, the dependent variable is the pre-cooling time, and the independent variables include the power output of the supply air inverter during pre-cooling, the supply air set temperature during pre-cooling, the indoor temperature-indoor set temperature just before pre-cooling, supply heat capacity, and the lowest outdoor air temperature during non-cooling/non-heating hours. The correlation coefficient R2 of the multiple regression model between the pre-cooling hour and the internal/external factors is of 0.612, and this could be used to provide information related to energy conservation and operating guidance.