Comparative evaluation of machine learning algorithms for greenhouse gas emission forecasting: a case study of Turkey (2012-2021).

Forecasting of Turkey's greenhouse gas emissions using linear and nonlinear rolling metabolic grey model based on optimization

Major US electric utility climate pledges have the potential to collectively reduce power sector emissions by one-third

Korean police agencies can discover useful trends and patterns using machine learning algorithms. However, which algorithm to select is difficult problem, because the accuracy and performance of machine learning algorithms may depend on dataset used for analysis and purpose of analysis itself. In this short paper, we describe a study on performance evaluation of machine learning algorithms for crime dataset. Unlike previous methods, we employ the real South Korean crime dataset that we collected from various sources, and carry out performance evaluation of well-known machine learning algorithms on this real crime dataset. As far as we know, this is the first attempt on performing such kind of research using South Korean crime dataset.

BackgroundPolicies to mitigate climate change by reducing greenhouse gas (GHG) emissions can yield public health benefits by also reducing emissions of hazardous co-pollutants, such as air toxics and particulate matter. Socioeconomically disadvantaged communities are typically disproportionately exposed to air pollutants, and therefore climate policy could also potentially reduce these environmental inequities. We sought to explore potential social disparities in GHG and co-pollutant emissions under an existing carbon trading program—the dominant approach to GHG regulation in the US and globally.Methods and findingsWe examined the relationship between multiple measures of neighborhood disadvantage and the location of GHG and co-pollutant emissions from facilities regulated under California’s cap-and-trade program—the world’s fourth largest operational carbon trading program. We examined temporal patterns in annual average emissions of GHGs, particulate matter (PM2.5), nitrogen oxides, sulfur oxides, volatile organic compounds, and air toxics before (January 1, 2011–December 31, 2012) and after (January 1, 2013–December 31, 2015) the initiation of carbon trading. We found that facilities regulated under California’s cap-and-trade program are disproportionately located in economically disadvantaged neighborhoods with higher proportions of residents of color, and that the quantities of co-pollutant emissions from these facilities were correlated with GHG emissions through time. Moreover, the majority (52%) of regulated facilities reported higher annual average local (in-state) GHG emissions since the initiation of trading. Neighborhoods that experienced increases in annual average GHG and co-pollutant emissions from regulated facilities nearby after trading began had higher proportions of people of color and poor, less educated, and linguistically isolated residents, compared to neighborhoods that experienced decreases in GHGs. These study results reflect preliminary emissions and social equity patterns of the first 3 years of California’s cap-and-trade program for which data are available. Due to data limitations, this analysis did not assess the emissions and equity implications of GHG reductions from transportation-related emission sources. Future emission patterns may shift, due to changes in industrial production decisions and policy initiatives that further incentivize local GHG and co-pollutant reductions in disadvantaged communities.ConclusionsTo our knowledge, this is the first study to examine social disparities in GHG and co-pollutant emissions under an existing carbon trading program. Our results indicate that, thus far, California’s cap-and-trade program has not yielded improvements in environmental equity with respect to health-damaging co-pollutant emissions. This could change, however, as the cap on GHG emissions is gradually lowered in the future. The incorporation of additional policy and regulatory elements that incentivize more local emission reductions in disadvantaged communities could enhance the local air quality and environmental equity benefits of California’s climate change mitigation efforts.

Quantifying the impact of energy consumption sources on GHG emissions in major economies: A machine learning approach

Addressing the challenge of rising carbon dioxide (CO2) and greenhouse gas (GHG) emissions is a critical priority in global efforts to combat climate change. The primary aim is to assess the relationship between energy intensity, private investments in energy, renewable energy consumption, export-related factors, and their influence on CO2 and GHG emissions in Turkey. The study employs a multi-level approach using correlation and regression analyses to explore the impact of the selected variables. A Bayesian correlation analysis was conducted to evaluate the strength of relationships between variables, and a regression model was used to test the significance of each factor. Data were gathered from official sources on energy intensity, renewable energy consumption, private investments in energy, and export-related variables in Turkey from 2007 to 2022. The study employed the JASP statistical software. The analysis showed that energy intensity and private energy investments are the most significant predictors of CO2 and GHG emissions. Energy intensity exhibited a strong negative correlation with CO2 emissions per capita (r = -0.717, BF₁₀ = 10.456) and GHG emissions (r = -0.802, BF₁₀ = 44.224), highlighting the critical role of energy efficiency in reducing emissions. Renewable energy consumption also played a role, though its influence was less pronounced than energy efficiency and investment. Based on the findings, it is recommended that policymakers prioritize energy efficiency improvements and create incentives for private investment in renewable energy technologies. Future studies should focus on sector-specific energy efficiency improvements and policy frameworks to enhance private sector engagement in clean energy initiatives.

Mitigating Curtailment and Carbon Emissions through Load Migration between Data Centers

Despite the availability of efficacious direct-acting antiviral (DAA) therapy, the number of people infected with hepatitis C virus (HCV) continues to rise, and HCV remains a leading cause of liver-related morbidity, liver transplantation, and mortality. We developed and validated machine learning (ML) algorithms to predict DAA treatment failure. Using the HCV-TARGET registry of adults who initiated all-oral DAA treatment, we developed elastic net (EN), random forest (RF), gradient boosting machine (GBM), and feedforward neural network (FNN) ML algorithms. Model performances were compared with multivariable logistic regression (MLR) by assessing C statistics and other prediction evaluation metrics. Among 6525 HCV-infected adults, 308 patients (4.7%) experienced DAA treatment failure. ML models performed similarly in predicting DAA treatment failure (C statistic [95% CI]: EN, 0.74 [0.69–0.79]; RF, 0.74 [0.69–0.80]; GBM, 0.72 [0.67–0.78]; FNN, 0.75 [0.70–0.80]), and all 4 outperformed MLR (C statistic [95% CI]: 0.51 [0.46–0.57]), and EN used the fewest predictors (n = 27). With Youden index, the EN had 58.4% sensitivity and 77.8% specificity, and nine patients were needed to evaluate to identify 1 DAA treatment failure. Over 60% treatment failure were classified in top three risk decile subgroups. EN-identified predictors included male sex, treatment < 8 weeks, treatment discontinuation due to adverse events, albumin level < 3.5 g/dL, total bilirubin level > 1.2 g/dL, advanced liver disease, and use of tobacco, alcohol, or vitamins. Addressing modifiable factors of DAA treatment failure may reduce the burden of retreatment. Machine learning algorithms have the potential to inform public health policies regarding curative treatment of HCV.

Aims/Purpose: We present a comprehensive evaluation of machine learning (ML) algorithms using Optical coherence tomography (OCT) and optical coherence tomography angiography (OCT‐A) for the diagnosis of Multiple sclerosis (MS).Methods: Prospective observational study using OCT volumes acquired with Cirrus HD‐OCT 5000 (Carl Zeiss, Meditec, Dublin, CA, EEUU) using the scanning protocols Optic Disc Cube 200 × 200 and Macular Cube 512 × 128, and OCT‐A macular volumes 175 × 350 with FOV 6 × 6 mm and OCT‐A of the Optic Nerve Head (ONH) with FOV 4.5 × 4.5 mm. The total data set consisted of 79 eyes of 40 patients with MS and 54 eyes of 27 control subjects. We evaluated classification algorithms such as Support Vector Machines, k‐Nearest Neighbour, Decision Trees, Random Forest, Extra Trees Classifier and Gaussian NB using data extracted from Ganglion Cell Analysis (8 parameters), Macular Thickness Analysis (7 parameters) and Optic Disc Analysis (5 parameters) from the OCT volumes and 5 parameters from the analysis of the macular OCT‐A volumes and 6 parameters from the analysis of the ONH. We studied all the ML algorithms considering three scenarios: OCT (20 parameters), OCT‐A (11 parameters), and OCT + OCT‐A (31 parameters). The models were trained using a 5‐fold cross‐validation strategy. We calculated the accuracy, F1‐score for each model and the area under the curve (AUC) was used to select the best ML algorithm. We applied SHAP (SHapley Additive exPlanations) to explain the output of the best ML model.Results: For the MS diagnosis, the best results were obtained for OCT with Gaussian NB (accuracy 0.865 ± 0.085, F1‐score 0.864 ± 0.086 and AUC 0.871 ± 0.081), for OCT‐A with Random Forest (accuracy 0.832 ± 0.032, F1‐score 0.832 ± 0.043 and AUC 0.831 ± 0.043) and for OCT + OCT‐A with Gaussian NB (accuracy 0.885 ± 0.028, F1‐score 0.886 ± 0.028 and AUC 0.891 ± 0.034). The input parameters with the highest contribution to the model's predictions were the minimum ganglion cell thickness and the temporal flow index in the ETDRS grid.Conclusions: This study highlights the efficacy of machine learning techniques in utilizing combined parameters from OCT and OCT‐A images of the macula and ONH tests to facilitate early diagnosis of MS. Moreover, it underscores the significance of ophthalmic examination in the comprehensive management of the disease.

This study determines whether greenhouse gas (GHG) emission statements are subject to assurance audits, and whether they affect the operating profitability of companies. Two secondary questions are explored. First, does the profitability of companies requesting assurance engagements vary by sector? Second, are regulations relating to the verification of GHG emissions in Turkey an effective control over the companies’ assurance demands? This study used data from fifty BIST Sustainability Index companies’ sustainability reports from 2011 to 2017. It was determined whether the sustainability reports included a company-disclosed assurance report on GHG emissions, and the results were summarized in Microsoft Excel. Profitability ratios were used to determine whether and to what extent assurance engagements were performed by verification bodies (ROA—return on assets ratio, ROE -Return on Equity). The data were subjected to frequency, crosstab, and parametric and nonparametric analyses using SPSS statistical software. The rate of BIST Sustainability Index companies requesting an assurance engagement was quite low (at 20%), and possession of an assurance engagement did not significantly affect company profitability. By sector, although there was a significant difference in the profitability of companies with an assurance engagement report, those in the financial sector were found to be more profitable than in other sectors. However, companies with an assurance engagement report in the transportation, telecommunication, and warehousing sectors exhibited the lowest profitability among the sectors. This study takes a research-based approach to investigate assurance engagements in GHG emission reports in Turkey. The study is original because there are few studies in Turkey.KeywordsCarbon emissionsAssurance engagementsSustainability ReportsOperating Profit

Energy-related activities are a major contributor of greenhouse gas (GHG) emissions. A growing body of knowledge clearly depicts the links between human activities and climate change. Over the last century the burning of fossil fuels such as coal and oil and other human activities has released carbon dioxide (CO2) emissions and other heat-trapping GHG emissions into the atmosphere and thus increased the concentration of atmospheric CO2 emissions. The main human activities that emit CO2 emissions are (1) the combustion of fossil fuels to generate electricity, accounting for about 37% of total U.S. CO2 emissions and 31% of total U.S. GHG emissions in 2013, (2) the combustion of fossil fuels such as gasoline and diesel to transport people and goods, accounting for about 31% of total U.S. CO2 emissions and 26% of total U.S. GHG emissions in 2013, and (3) industrial processes such as the production and consumption of minerals and chemicals, accounting for about 15% of total U.S. CO2 emissions and 12% of total ...

Taking Stock of Strategies on Climate Change and the Way Forward: A Strategic Climate Change Framework for Australia

Evaluation of machine learning algorithms for renin-angiotensin-aldosterone system inhibitors associated renal adverse event prediction

Livestock sheds are local and regional hotspots of greenhouse gases (GHG) emissions, but only very few studies analyse the intensities of GHG emissions from this source. The objective of this study was to quantify annual CH 4 , CO 2 and N 2 O emissions from inside and outside of sheepfolds and summer cattle sheds in a typical agro‐pastoral ecotone using static chamber technique. Both sheepfolds and cattle shed functioned as huge net sources of CH 4 and N 2 O at annual scale. Animal presence increased CH 4 , CO 2 and N 2 O effluxes for up to 1100 times compared to the animal sheds without animals. N 2 O emissions boosted for 160%–280% during and after rainfall and spring‐thaw events. The CH 4 and CO 2 fluxes increased exponentially with faeces temperature for the outside sheepfold and summer cattle shed. The annual GHG emissions from both sheepfolds and summer cattle shed were 56 t CO 2 equivalents ha −1 , of which N 2 O contributed to 94%. Sheepfold dominated the total GHG emissions from animal sheds and accounted for 83% of the annual GHG flux. Annual emission on a per animal basis was 15, 0.2 and 28 kg CO 2 eq year −1 sheep −1 and 26, 10 and 140 kg CO 2 eq year −1 cattle −1 for N 2 O, CH 4 and CO 2 , respectively. The annual N 2 O emissions from animal sheds were 70–250 times larger than nearby grassland soils, which were also net sink for atmospheric CH 4 . Concluding, animal sheds are very intensive local hotspots of GHG emissions, which should be considered at the local and regional scales.

ABSTRACTEnvironmental sustainability has emerged as a global imperative, necessitating a nuanced understanding of the determinants of greenhouse gas (GHG) emissions. Although previous studies have highlighted various environmental indicators the specific influence of technological innovation, renewable energy adoption, and forest cover on GHG emissions within G‐7 countries remains insufficiently explored. In this regard, this study examines the impact of these factors on GHG emissions across the G‐7 countries over the period 1995–2021. The empirical results reveal significant cross‐sectional dependence and confirm that technological innovation, renewable energy consumption, and forest conservation significantly contribute to reducing GHG emissions. In contrast, tourism development and economic growth are associated with increased environmental degradation. Moreover, causality analysis uncovers a unidirectional relationship from GHG emissions to technological innovations and from tourism to GHG emissions, alongside a bidirectional causal nexus between renewable energy and GHG emissions. Furthermore, advanced machine learning techniques, including Gradient Boosting (GB) and Extreme Gradient Boosting (XGBoost), identify these variables as key predictors of GHG emissions. These findings underscore the critical role of data‐driven, evidence‐based policymaking in advancing environmental sustainability. By strategically integrating technological advancements and renewable energy frameworks, G‐7 policymakers can more effectively mitigate climate change impacts. The study advocates for comprehensive and targeted policy interventions that address the intricate interplay between economic dynamics and environmental imperatives in the pursuit of a sustainable future.