China
Africa
Explore chapters and articles related to this topic
Characterization of Seeds by a Fuzzy Clustering Algorithm
Published in Don Potter, Manton Matthews, Moonis Ali, Industrial and Engineering Applications of Artificial Intelligence and Expert Systems, 2020
Younes Chtioui, Dominique Bertrand, Dominique Barba
The measured features were correlated. It was necessary to reduce the redundancy by extracting (or selecting) the key features which were effective in discriminating between pattern classes. Principal component analysis (PCA), which is a well-known method in multivariate analysis, achieves a dimensionality reduction of a data set of correlated variables. This is achieved by creating new un-correlated variables, called « principal components », while maintaining as much as possible the diversity present in the original data set. These principal components are ranked so that the first few components describe the largest part of the variation of the original data set [Jol86]. PCA was applied here on the learning data as active observations, and on the test data as supplementary ones.
Applied Statistics
Published in Vinayak Bairagi, Mousami V. Munot, Research Methodology, 2019
Varsha K. Harpale, Vinayak K. Bairagi
Most of the systems are defined with multiple independent or response variables and analysis of such system using multiple variables is a multivariate Data Analysis. This statistical technique is used to perform operable studies across multiple dimensions considering the effects of all variables on the responses of interest. The uses of MVA comprises: Inverse and Capability-based designAnalysis of Alternatives (AoA)Analysis of concepts with recent trendsIdentification of critical design-drivers and correlations across hierarchical levels.Multivariate analysis is suffered by high dimensionality and thus observed to be computationally costly. As most of the systems are hierarchically designed this complexity increases with its level of design. Thus in MVA parametric optimization is one of the preferred concepts of dimensionality reduction and in turn, reduces the complexity of the analysis.
A study of a low-carbon and intelligent city index system
Published in Jimmy C.M. Kao, Wen-Pei Sung, Civil, Architecture and Environmental Engineering, 2017
In order to avoid the subjectivity of selecting and determining index system, correlation analysis and factor analysis rating method are used to construct the index system in this paper. The factor analysis rating method decides the index system by factor analysis, factor analysis is a multivariate analysis that simplifies a lot of complicated indicators to the few meaningful aspects/component using the correlation matrix. Therefore, factor analysis is not only used to simplify the numbers of indicators and select the meaningful indicators in this paper, but also determine the relative weight of each aspect and indicators. In the empirical analysis of factor analysis, principal-axis factoring analysis (PFA) is used as the way of factor extraction, and varimax is used as the way of rotation.
EASIER System. Evaluating a Spanish Lexical Simplification Proposal with People with Cognitive Impairments
Published in International Journal of Human–Computer Interaction, 2022
Rodrigo Alarcon, Lourdes Moreno, Paloma Martínez, José A. Macías
We have applied multivariate analysis to find further insights, principally based on human-computer interaction (HCI) concerns, and thus reinforce the above hypotheses with user-based evidence (Alboukadel & Fabian, 2019). Multivariate analysis helps identify structures and relations between different variables comprising the data and might include applications like dimensionality reduction, principal component analysis (PCA), clustering, or multidimensional scaling, to name just a few. This kind of statistics has already been applied to HCI studies (Castells & Macías, 2002; Macías, 2021; Macías & Castells, 2001; Macías & Culén, 2021; Macías et al., 2009; Sánchez & Macías, 2019), obtaining successful results to generate new knowledge, improve decision-making and further corroborate research hypotheses.
Key Success Factors and Benefits of Kaizen Implementation
Published in Engineering Management Journal, 2020
Vesna Janjić, Mirjana Todorović, Dejan Jovanović
Questionnaire data was analyzed using SPSS (Version 20.0). Measurement of reliability and internal consistency of variables was performed using the Cronbach’s Alpha coefficient. Descriptive statistics were used to measure the central tendency and variability. In this article, factor analysis was used, similar to other studies (Garcia et al., 2013; Hailu et al., 2017). Factor analysis is often used in this type of research where there are many variables and a need for their compression. Factor analysis is a multivariate analysis method that is used if the number of variables is large, as is the case in the conducted empirical research, and it can help select a representative subset of variables or create new variables as substitutes for original variables, retaining their original character (Hair, Anderson, Tatham, & Black, 1998). The formation of a factor model is based on the assumption that the variables can be grouped according to their correlations in different groups. In this article, an exploratory approach to factor analysis was used, specifically Principal Components Analysis (PCA) method was applied, while the VARIMAX criterion was chosen as a rotation technique, since it provides a clearer separation of the factors and proved to be a very successful analytical orthogonal factor rotation approach (Hair et al., 1998). It is often said that factor analysis can be only as good as the data allows. Even though there are many limitations, factor analysis is used widely due to its wide scope and positive features.
The assessment of soil contamination by heavy metals using geostatistical sequential Gaussian simulation method
Published in Human and Ecological Risk Assessment: An International Journal, 2018
Adem Ersoy, Tayfun Yusuf Yünsel
Several methods or techniques have been suggested to investigate for quantification of soil contamination by heavy metals. These techniques can be given as Geographic Information Systems (GIS), geostatistical studies, multivariate statistical analysis and other methods. GIS was initially developed as a tool for the data retrieval and displaying geographic information, and later enhanced for spatial analysis (Fotheringham and Rogerson 1994). GIS with various spatial interpolation methods including inverse distance and Kriging is used in several regional scales for soil quality survey studies (Lu et al.2012; Kelepertzis 2014; Huang et al.2015; Moore et al.2016). Geostatistical methods are used to estimate unknown soil properties between known sampling locations. Kriging is commonly applied. Multivariate analysis methods include analyses of principal component, cluster, Pearson correlation, factor, and multiple linear regression. Last three are less commonly used methods. Heavy metal contamination in soils has been well documented by a combination of GIS techniques and multivariate statistical analysis or Kriging (Yesilonis et al.2008; Davis et al.2009; Shan et al.2013; Gabarrón et al.2017; Hou et al.2017). Other methods are geo-accumulation index, enrichment factor, contamination factor and integrated pollution index to evaluate the risk of metal contamination. There are many studies of contamination indices in the literature (e.g. Li et al. 2014; Teh et al. 2016). Geo-accumulation index has been most commonly used for evaluation of metals background concentrations (e.g. (Solgi and Parmah 2015).