Search

IntroductionPrevious studies have sought to identify risk factors for malnutrition in populations of schoolchildren, depending on traditional logistic regression methods. However, holistic machine learning (ML) approaches are emerging that may provide a more comprehensive analysis of risk factors.MethodsThis study employed feature selection and association rule learning ML methods in conjunction with logistic regression on epidemiological survey data from 1,036 Ethiopian school children. Our first analysis used the entire dataset and then we reran this analysis on age, residence, and sex population subsets.ResultsBoth logistic regression and ML methods identified older childhood age as a significant risk factor, while females and vaccinated individuals showed reduced odds of stunting. Our machine learning analyses provided additional insights into the data, as feature selection identified that age, school latrine cleanliness, large family size, and nail trimming habits were significant risk factors for stunting, underweight, and thinness. Association rule learning revealed an association between co-occurring hygiene and socio-economical variables with malnutrition that was otherwise missed using traditional statistical methods.DiscussionOur analysis supports the benefit of integrating feature selection methods, association rules learning techniques, and logistic regression to identify comprehensive risk factors associated with malnutrition in young children.

The Karner blue butterfly (Lycaeides melissa samuelis, or Kbb), a federally endangered species under the U.S. Endangered Species Act in decline due to habitat loss, can be further threatened by climate change. Evaluating how climate shapes the population trend of the Kbb can help in the development of adaptive management plans. Current demographic models for the Kbb incorporate in either a density-dependent or density-independent manner. We instead created mixed density-dependent and -independent (hereafter "endo-exogenous") models for Kbbs based on long-term count data of five isolated populations in the upper Midwest, United States during two flight periods (May to June and July to August) to understand how the growth rates were related to previous population densities and abiotic environmental conditions, including various macro- and micro-climatic variables. Our endo-exogenous extinction risk models showed that both density-dependent and -independent components were vital drivers of the historical population trends. However, climate change impacts were not always detrimental to Kbbs. Despite the decrease of population growth rate with higher overwinter temperatures and spring precipitations in the first generation, the growth rate increased with higher summer temperatures and precipitations in the second generation. We concluded that finer spatiotemporally scaled models could be more rewarding in guiding the decision-making process of Kbb restoration under climate change.

Watershed management is the art and practice of understanding stakeholder values for ecosystem services within a watershed and instituting management practices that consider trade-offs to sustain these goods and services. Effective watershed management practices are hydrologically defined, ecosystem-based, inclusive, and integrate biophysical as well as socioeconomic decisions. The uncertainties and unpredictability of climate change create an ambiguous backdrop to the increasingly social problem of water resource management. Inequities in watershed decision-making processes often lead to the reinforcement of power and resource imbalances. Future watershed managers must be able to engage across socioeconomic and cultural boundaries to support decisions that advance water as a human right in an uncertain future. We offer a design for a graduate level, 15-week university course that uses publicly available resources to help emerging watershed leaders prepare for an uncertain future. The design is interactive and constructivist, engaging the refereed literature and leading to an increased understanding of ecosystem-based watershed management under climate scenarios, with special attention to vulnerable populations.