Graphical group ridge

Aldahmani, Saeed; Zoubeidi, Taoufik

doi:10.1080/00949655.2020.1803320

Cited by 6 publications

(3 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This method could classify ridge regression predictors in disjoint groups of conditionally correlated variables and derived different penalties (i.e., shrinkage parameters) for predictors groups. [ 56 ] It had the advantage of combining the ridge regression method with graphical model for ill‐conditioned data or high‐dimensional data.…”

Section: Methodsmentioning

confidence: 99%

Temperature and Rainfall Patterns Constrain the Multidimensional Rewilding of Global Forests

Zhou

Eldridge

et al. 2022

Advanced Science

View full text Add to dashboard Cite

The long-term contribution of global forest restoration to support multiple dimensions of biodiversity and ecosystem function remains largely illusive across contrasting climates and forest types. This hampers the capacity to predict the future of forest rewilding under changing global climates. Here, 120 studies are synthesized across five continents, and it is found that forest restoration promotes multiple dimensions of biodiversity and ecosystem function such as soil fertility, plant biomass, microbial habitat, and carbon sequestration across contrasting climates and forest types. Based on global relationship between stand age and soil organic carbon stock, planting 350 million hectares of forest under the UN Bonn Challenge can sequester >30 Gt soil C in the surface 20 cm over the next century. However, these findings also indicate that predicted increases in temperature and reductions in precipitation can constrain the positive effects of forest rewilding on biodiversity and ecosystem function. Further, important tradeoffs are found in very old forests, with considerable disconnection between biodiversity and ecosystem function. Together, these findings provide evidence of the importance of the multidimensional rewilding of forests, suggesting that on-going climatic changes may dampen the expectations of the positive effects of forest restoration on biodiversity and ecosystem function.

show abstract

Section: Methodsmentioning

confidence: 99%

Temperature and Rainfall Patterns Constrain the Multidimensional Rewilding of Global Forests

Zhou

Eldridge

et al. 2022

Advanced Science

View full text Add to dashboard Cite

show abstract

“…Distributions are coloured to represent the designation of ‘left’, ‘right’, or ‘neither’ as described by the LI. Distributions are presented as geom_density functions form the ggplot2 and ggridge packages in R [43,44]. These are kernel density estimates and can be considered as smoothed versions of histograms and are designed to provide a non-binned density distribution of continuous data and have scales relevant to the number of dives for each bird.…”

Section: Resultsmentioning

confidence: 99%

Handedness and individual roll-angle specialism when plunge diving in the northern gannet

Bennison,

Clark,

Votier

et al. 2023

Biol. Lett.

View full text Add to dashboard Cite

Many vertebrates show lateralized behaviour, or handedness, where an individual preferentially uses one side of the body more than the other. This is generally thought to be caused by brain lateralization and allows functional specializations such as sight, locomotion, and decision-making among other things. We deployed accelerometers on 51 northern gannets, Morus bassanus , to test for behavioural lateralization during plunge dives. When plunge diving, gannets ‘roll’ to one side, and standard indices indicated that 51% of individuals were left-sided, 43% right-sided, and 6% ‘non-lateralized’. Lateralization indices provide no measure of error and do not account for environmental covariance, so we conducted two repeatability analyses on individuals' dive roll direction and angle. Dive side lateralization was highly repeatable among individuals over time at the population level ( R = 0.878, p < 0.001). Furthermore, roll angle was also highly repeatable in individuals ( R = 0.751, p < 0.001) even after controlling for lateralized state. Gannets show individual specializations in two different parts of the plunge diving process when attempting to catch prey. This is the first demonstration of lateralization during prey capture in a foraging seabird. It is also one of the few demonstrations of behavioural lateralization in a mixed model approach, providing a structure for further exploring behavioural lateralization.

show abstract

“…These datasets are usually known as (n < p) or highdimensional data. Such high-dimensional gene expression datasets are usually not effective for the purpose of classification because of the issue of dimensionality [1], [2]. Moreover, most of the genes in such type of datasets are noisy and have little contribution in the true classification of observations (phenotypes).…”

Section: Introductionmentioning

confidence: 99%

Gene Selection in Binary Classification Problems Within Functional Genomics Experiments via Robust Fisher Score

Hamraz

Khan

et al. 2022

IEEE Access

View full text Add to dashboard Cite

This study proposes a supervised feature selection technique for classification in high dimensional binary class problems by adding robustness in the conventional Fisher Score. The proposed method utilizes the more robust measure of location, i.e. the median and measure of dispersion known as Rousseeuw and Croux statistic (Q n ). Initially, a minimum subset of genes is identified by the greedy search approach, which is then combined with the top ranked genes obtained via the proposed Robust Fisher Score (RFish). To remove redundancy in the selected genes, Least Absolute Shrinkage and Selection Operator (LASSO) is then applied. The proposed method is validated on five publicly available datasets and is further assessed in a detailed simulation study. The results of the proposed method are compared with six well known feature selection methods based on prediction performance via Random Forest (RF), Support Vector Machine (SVM) and k Nearest Neighbour (k-NN) classifiers. The findings are presented in boxplots and barplots, which show that the proposed method (RFish) outperforms all the other methods in the majority of cases.INDEX TERMS Classification, feature selection, high dimensional gene expression datasets, Fisher Score, Rousseeuw and Croux statistic.

show abstract

Graphical group ridge

Cited by 6 publications

References 21 publications

Temperature and Rainfall Patterns Constrain the Multidimensional Rewilding of Global Forests

Temperature and Rainfall Patterns Constrain the Multidimensional Rewilding of Global Forests

Handedness and individual roll-angle specialism when plunge diving in the northern gannet

Gene Selection in Binary Classification Problems Within Functional Genomics Experiments via Robust Fisher Score

Contact Info

Product

Resources

About