Variable selection and the interpretation of principal subspaces

Cadima, Jorge; Jolliffe, Ian T.

doi:10.1198/108571101300325256

Cited by 82 publications

(50 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Software has been written that implements the relevant methodology. 28 Future work may also find that the PCA method may reduce errors in classifying tasks as inside or outside control limits for acceptable driving performance. Type I errors (false positives) mean that a task is classified as outside control limits when it is really inside control limits, leading to unnecessary redesign or lockout of the task while driving.…”

Section: Discussionmentioning

confidence: 99%

The Dimensions of Driver Performance during Secondary Manual Tasks

Young

Angell

2005

Proceedings of the 2nd International Driving Symposium on Human Factors in Driver Assessment, Training and Vehicle Design: Driv

View full text Add to dashboard Cite

Summary: This analysis identified the underlying dimensions of driver performance using data obtained from drivers engaged in secondary manual tasks. Randomly chosen subjects balanced for age and gender used one of five advanced navigation and communication systems while driving on a closed roadway. Fifteen driver performance variables were averaged and standardized across subjects for 79 tasks. There were high correlations between all variables. Principal Component Analysis (PCA) found that the vector of loadings defining the first principal component (PC1) was positive for all 15 variables, accounting for 61 percent of the total variation across all tasks. It is interpreted as overall driver demand. PC2 loaded with one sign on event detection and response variables, but opposing sign on visual-manual workload variables. It identified tasks making drivers more inattentive to outside events than expected, given a task's visual-manual workload, and accounted for 17 percent of total variation. It is interpreted as low-workloadbut-high-inattentiveness. PC3 had loadings of opposing sign for peripheral vs. central event variables (5 percent of total variation). It is interpreted as peripheral insensitivity. The first three components together accounted for 83 percent of total variation, which is deemed substantial. Thus most of the information available through the 15 original variables can be summarized by only three PC variables. Because the vectors of loadings defining the components are orthogonal to each other as defined by PCA, no single variable by itself can capture all the important variations in driver performance during secondary manual tasks. A multivariate design and analysis is required.

show abstract

Section: Discussionmentioning

confidence: 99%

The Dimensions of Driver Performance during Secondary Manual Tasks

Young

Angell

2005

Proceedings of the 2nd International Driving Symposium on Human Factors in Driver Assessment, Training and Vehicle Design: Driv

View full text Add to dashboard Cite

show abstract

“…The percentages of total variation accounted for by each of the three principal components were 72.76, 7.71 and 6.44% respectively (Table 1). The proportion of total variation more than 75% is acceptable (Cadima & Jolliffe 2001;Jolliffe 2002) for characterization and evaluation of accessions in this genebank collection.…”

Section: Resultsmentioning

confidence: 99%

Genetic diversity of Albanian pea (Pisum sativum L.) landraces assessed by morphological traits and molecular markers

Gixhari¹,

Pavelková²,

Ismaili³

et al. 2014

Czech J. Genet. Plant Breed.

View full text Add to dashboard Cite

Gixhari B., Pavelková M., Ismaili H., Vrapi H., Jaupi A., : Genetic diversity of Albanian pea (Pisum sativum L.) landraces assessed by morphological traits and molecular markers. Czech J. Genet. Plant Breed., 50: 177-184.In order to investigate the genetic diversity present in the pea germplasm stored in the Albanian genebank, we analyzed 28 local pea genotypes of Albanian origins for 23 quantitative morphological traits, as well as 14 retrotransposonbased insertion polymorphism (RBIP) molecular markers. The study of morphological characters carried out during three growing seasons (2010, 2011 and 2012) had the objective of characterization of traits useful in breeding programs. RBIP marker analysis revealed the genetic similarity in range from 0.06 to 0.45. ANOVA, principal component analysis (PCA) and cluster analysis was used to visualize the association among different traits. Most of the quantitative morphological traits showed significant differences. PCA and cluster analysis (Ward's method) carried out for morphological traits divided the local pea genotypes into three clusters. Finally, the study identified the agronomicaly important traits which will facilitate the maintenance and agronomic evaluation of the collections.

show abstract

“…At last, to clarify contrast between two gatherings, and to apply right model to another program, i.e., excluded in dataset, it was assessed on the possibility to characterized modules into gatherings utilizing program measurements. To choose best components, i.e., measurements, to use in classifier, feature choice computers was applied, in particular independent features methodology [62]. This technique performs a statistical test for each individual component, showing that distinction is unrealistic to be random variation; if contrast is sig times lower than standard mistake, then highlight is not regarded valuable for ordering.…”

Section: Resultsmentioning

confidence: 99%

“…To get an exact model, Principal Component Analysis (PCA) strategy was adopted, which changes dependant variables in a small number of uncorrelated variables [62]; yet, this methodology did not enhance model. An exponential and a logarithmic model was assessed, however they were not ready to give better accuracy.…”

Section: Resultsmentioning

confidence: 99%

Program Aging and Service Crash

Ahamad¹

2016

IJCATR

View full text Add to dashboard Cite

Abstract:Program aging is a degradation of performance or functionality caused by resource depletion. The aging affects the cloud services which provide access to big data bank and computing resources. This suffers large budget and delays of defect removal, which requires other related solutions including renewal in the form of controlled restarts. Collection of various runtime metrics are more significant source for further study of detection and analysis of aging issues. This study highlights the method for detecting aging immediately after their introduction by runtime comparisons of different development scenarios. The study focuses on aging of program and service crash as a consequence.

show abstract

Variable selection and the interpretation of principal subspaces

Cited by 82 publications

References 18 publications

The Dimensions of Driver Performance during Secondary Manual Tasks

The Dimensions of Driver Performance during Secondary Manual Tasks

Genetic diversity of Albanian pea (Pisum sativum L.) landraces assessed by morphological traits and molecular markers

Program Aging and Service Crash

Contact Info

Product

Resources

About