China 
Africa 
Explore chapters and articles related to this topic
Microbiology of Metalworking Fluids
Published in Jerry P. Byers, Metalworking Fluids, Third Edition, 2018
The actual causes of MWF-related dermatitis are explained in Chapter 16. In this discussion of allergenic disease, it is relevant to address the misconception that dermatitis is caused by exposure to MWF microbes. The relationship between microbial contamination in MWFs and dermatitis is indirect. There is no evidence that dermatitis is caused by MWF microbes. However, there is strong evidence that when MWFs are treated to control microbial contamination, the combination of increased pH and biocide concentration can trigger allergic contact dermatitis in susceptible individuals. Note that the allergic reaction is not caused by the microbes. Rather, it is caused by the chemicals used to reduce MWF bioburden. This is a classic example of the statisticians’ adage “Correlation does not imply causation.”
Pitfalls of machine learning for tail events in high risk environments
Published in Stein Haugen, Anne Barros, Coen van Gulijk, Trond Kongsvik, Jan Erik Vinnem, Safety and Reliability – Safe Societies in a Changing World, 2018
C. Agrell, S. Eldevik, A. Hafver, F.B. Pedersen, E. Stensrud, A. Huseby
“Correlation does not imply causation” is a well used phrase within statistics, and describes what still remains as one of the major pitfalls in the analysis of data. The importance of this distinction depends on the degree to which one intends to intervene, and the consequence of erroneous intervention. See e.g (Pearl 2010). For many ML applications this may not be significant. But for the use cases considered in this paper it plays an important role, and we will argue that causality constraints from phenomenological knowledge should be incorporated in ML models–to strengthen model performance for tail events, to increase model transparency, and to make it easier to falsify models that do not comply with observations.
Statistics
Published in Benjamin D. Shaw, Uncertainty Analysis of Experimental Data with R, 2017
In R, the covariance and correlation of two data sets are calculated using the functions cov() and cor(), respectively. Here is an example where we plot a data set y as a function of x (Figure 3.1): The y data are noisy and we calculate the correlation coefficient as being ρxy = 0.96. This indicates that the data are well correlated, which is also evident in the plot. Always remember, though, that correlation does not imply causation. There could be another variable that is causing x and y to vary.
The spatial association of social vulnerability with COVID-19 prevalence in the contiguous United States
Published in International Journal of Environmental Health Research, 2022
Chuyuan Wang, Ziqi Li, Mason Clay Mathews, Sarbeswar Praharaj, Brajesh Karna, Patricia Solís
There are four limitations to this study. First, correlation does not imply causation. Our results only indicate the statistical relationship between SVI and COVID-19 prevalence in the contiguous US, but do not imply the potential cause–effect relationship between these two variables. Second, the results of a locally weighted correlation analysis depend on a pre-defined bandwidth or the number of neighboring features. The selection of a bandwidth is arbitrary and based on researchers’ experiences. A change of the bandwidth can result in changes in results. Third, we did not perform significance tests for local Spearman’s ρ because, to the best of our knowledge, there is no software that can compute local p-values for locally weighted correlation analysis. Fourth, the quality of COVID-19 data influences the accuracy of the results. There are some uncertainties within the national level COVID-19 data set because counties report COVID-19 cases and deaths using different methods or definitions. Future studies can use a wider selection of social, demographic, economic, and health indicators to perform a comprehensive analysis between vulnerability factors and COVID-19 prevalence for the entire United States or other countries around the world.
Demographic and socio-economic factors including sustainability related indexes in waste generation and recovery
Published in Energy Sources, Part A: Recovery, Utilization, and Environmental Effects, 2021
Yee Van Fan, Jiří Jaromír Klemeš, Chew Tin Lee, Raymond R. Tan
Correlation does not imply causation. Multiple linear regression, a statistical learning algorithm based on supervised learning (Goodfellow et al. 2016), is applied in this study to identify the best fitting model for EU-27 municipal waste generation prediction, as shown in Eq(2). The definition of municipal waste is produced by households, including similar wastes from sources such as commerce, offices, and public institutions collected by or on behalf of municipal authorities (Eurostat 2021c). The method of the ordinary least squares (OLS) model is followed. The model is built in Jupyter notebook, Version 6.3.0 (Jupyter 2021), with Scikit-learn, Version 0.24.1 (Scikit-learn 2021) and Numpy, Version 1.19.2 (NumPy 2021) as the main machine learning library, using Python programming language (Python Software Foundation 2021). The dataset is split into the ratio of 80% training data, 20% testing (27 observations) for cross-validation (Holdout Method).
Modeling interaction as a complex system
Published in Human–Computer Interaction, 2021
Niels van Berkel, Simon Dennis, Michael Zyphur, Jinjing Li, Andrew Heathcote, Vassilis Kostakos
The primary purpose of experiments and observational studies is to help researchers estimate and infer causal effects – although more exploratory approaches can also be used these are less common and are not our focus here. Consistent with the well-known dictum that ‘correlation does not imply causation,’ the problem is that many observed associations among variables cannot simply be understood as causal. In our example, researchers may find a relationship between mobile notifications and smartphone usage, but then not know if both variables have common causes, one causes the other, or bidirectional causality exists.