4653 research outputs found
Sort by
Fire detection and risk assessment via Support Vector Regression with Flattening-Samples Based Augmented Regularization
No full textWe propose a Hybrid Support Vector Regression (SVR) with Flattening-Samples Based Augmented Regularization (Hybrid FSR-SVR) architecture for multi-sensor fire detection and forest fire risk assessment. The Hybrid FSR-SVR is a lightweight architecture built upon the novel Flattening-Samples Based Augmented Regularization (FSR) approach and temporal trends of environmental variables. The FSR approach augments l2 norm based smoothing term into an l1-l2 combination, facilitating the integration of l1 regularization into the SVR method, thereby enhancing generalization with minimal computational load. We evaluate the performance of Hybrid FSR-SVR using two distinct datasets covering indoor and forest fires, benchmarking against 15 machine learning models, including state-of-the-art techniques, such as Recurrent Trend Predictive Neural Network (rTPNN), Long-Short Term Memory (LSTM), Multi-Layer Perceptron (MLP), Gated Recurrent Unit (GRU), and Gradient Boosting. Our findings demonstrate that Hybrid FSR-SVR effectively assesses the risk of forest fire, enabling early preventive measures. Notably, it achieves a remarkable accuracy of 0.95 for forest fire detection and ranks third with 0.88 accuracy for indoor fire detection. Importantly, it exhibits computation times significantly lower - by 1 to 2 orders of magnitude - than the majority of compared techniques. The superior generalization ability of Hybrid FSR-SVR, facilitated by flattening-samples based augmented regularization, allows for high detection performance even with smaller training sets.Computer Science, Artificial Intelligence || Computer Science, Interdisciplinary Application
Construction of a Waddington-like landscape model that can guide clinical exploration of p53-dynamics-activating parameters in the face of divergent p53 dynamics
No full textThe primary strategy of radiotherapies is to manipulate cell fate decision, a process mainly regulated by a spectrum of p53 dynamics. Based on their biological relevance, we analytically categorize the range of p53 levels into seven distinct level -forms, leading to the identification of eleven non -chaotic phenomena of p53 level -form dynamics. The superimposing of cell fate attractors on the co -dimension two bifurcation diagram of eleven p53 level -form dynamics, under quasi -steady state assumption, provides a mechanistic tool that can be posted as a Waddingtonlike landscape model for cell fate regulation by p53 dynamics. In the proposed model, the (location of) cell is represented as a control point in the bifurcation diagram representing a flattened landscape composed of 11 distinct behavioral regions, and the effort that moves the cell on the landscape is exerted by accumulating death factors upon DNA damage. Further analysis reveals that intrinsically -resistant cancer attractors inevitably exist on the landscape, and cells might have evolved to use a safe operational area whose cusp bifurcation shape is contributing to robustness via hysteresis. We further reveal specific mechanisms through which tumors acquire resistance under therapy. The proposed landscape model can be put to productive use via a reverse control methodology. The reconstruction of cancer -specific landscape can inform the design of personalized p53 -dynamics -based drug combination strategies suffering from the combinatorial explosion of target parameters and the divergent p53 dynamics. We conclude that the reverse control of the proposed landscape model has the potential to bridge the gap between theoretical and clinical studies of p53 dynamics for p53 -dynamics -based cancer therapies.Mathematics, Applied || Mathematics, Interdisciplinary Applications || Mechanics || Physics, Fluids & Plasmas || Physics, Mathematica
Multi-Sensor E-Nose Based on Online Transfer Learning Trend Predictive Neural Network
No full textElectronic Nose (E-Nose) systems, widely applied across diverse fields, have revolutionized quality control, disease diagnostics, and environmental management through their odor detection and analysis capabilities. The decision and analysis of E-Nose systems often enabled by Machine Learning (ML) models that are trained offline using existing datasets. However, despite their potential, offline training efforts often prove intensive and may still fall short in achieving high generalization ability and specialization for considered application. To address these challenges, this paper introduces the e-rTPNN decision system, which leverages the Recurrent Trend Predictive Neural Network (rTPNN) combined with online transfer learning. The recurrent architecture of the e-rTPNN system effectively captures temporal dependencies and hidden sequential patterns within E-Nose sensor data, enabling accurate estimation of trends and levels. Notably, the system demonstrates the ability to adapt quickly to new data during online operation, requiring only a small offline dataset for initial learning. We evaluate the performance of the e-rTPNN decision system in two domains: beverage quality assessment and medical diagnosis, using publicly available wine quality and Chronic Obstructive Pulmonary Disease (COPD) datasets, respectively. Our evaluation indicates that the proposed e-rTPNN achieves decision accuracy exceeding 97 % while maintaining low execution times. Furthermore, comparative analysis against established Machine Learning (ML) models reveals that the e-rTPNN decision system consistently outperforms these models by a significant margin in terms of accuracy.Computer Science, Information Systems || Engineering, Electrical & Electronic || Telecommunication
Assessing the Spatial and Temporal Characteristics of Meteorological Drought in Afghanistan
No full textAfghanistan is suffering from periodic events of drought, which has exacerbated in recent years due to extreme climate events in the region. Having an arid to semi-arid climate, the country faces significant challenges of water resources management, especially for irrigation as reliance on agriculture is cumbersome. This study is undertaken to characterize historical meteorological drought in Afghanistan to provide an insight on where and when meteorological drought events happened in different River Basins (RBs). The study mainly employs the gamma-Standardized Precipitation Index (gamma-SPI) to analyze historical meteorological droughts across Afghanistan from 1979 to 2019. Monthly precipitation data is obtained from the Ministry of Energy and Water (MEW) of Afghanistan, which is a combination of observed data from ground stations and gap-filled data by the MEW for the study period. Gridded gamma-SPI values are interpolated and mapped to visualize patterns of spatial drought across the entire country. The results indicate that countrywide extreme drought events occurred in 1999, 2000, 2001, 2010, 2016, 2017, and 2019, particularly affecting southern, western, and southwestern regions. Decreasing rainfall occurred in all five RBs, with the most considerable decline observed in the 1999-2008 period. The study reveals the increasing frequency and severity of meteorological droughts in Afghanistan. It also emphasizes on the vulnerability of agriculture and water sectors due to the drought events. The findings of the study suggest the need for better drought monitoring, preparedness, awareness, and adaptation of strategies to ensure water security and agricultural sustainability in the face of climate change.Geochemistry & Geophysic
Asymmetric nexus of coal consumption with environmental quality and economic growth: Evidence from BRICS, E7, and Fragile Five countries by novel quantile approaches
No full textThe study analyzes the asymmetric nexus of coal consumption with environmental quality and economic growth. In this context, the study focuses on eight leading emerging countries that take place in BRICS, E7, and Fragile Five groups. Also, the study uses yearly data from 1989 to 2021 and performs novel quantile methods, such as Granger Causality-in-Quantiles and Quantile-on-Quantile Regression (QQR). Also, quantile regression is used for robustness check. The results present that (i) there are causalities from coal consumption to both environmental quality and economic growth at 10% significance, whereas quantile and country-based results differ || (ii) effects of coal consumption on environmental quality are much stronger in lower quantiles for Brazil, Indonesia, India, South Africa, and, Turkey, but in higher quantiles for China, Mexico, and Russia || (iii) effects of coal consumption on economic growth are much stronger in lower quantiles for Brazil, Indonesia, India, Russia, South Africa, and Turkey || in higher quantiles for China || lower and middle quantiles for Russia || and all quantiles for Mexico || and (iv) the robustness of the QQR results are validated. Hence, empirical outcomes underline the highly crucial effects of coal consumption on environmental quality and economic growth in the countries. The results imply that policymakers should focus on efforts to decrease coal consumption gradually by applying a macro transition plan to increase environmental quality without causing economic decline by considering changing effects of coal consumption at quantiles and countries.Environmental Studie
Shock resistors or transmitters? Contagion across industries and countries during the COVID-19 pandemic and the global financial crisis
No full textWe examine how global shocks from various sources propagate across industries and countries. Financial contagion is measured using residual-based and volatility-adjusted correlation. Specific industries and countries were resilient during both global crises, while others played a significant role in transmitting shocks.Economic
A numerical approach to exergy-based sustainability and environmental assessments of solar energy-powered district cooling systems using actual operational data
No full textThe demand for cooling in buildings has been increasing at a higher rate than heating, and more energy is expected to meet this demand. Solar energy can be vital in fulfilling this energy requirement based on its unique renewable energy features. The solar thermal powered absorption cooling (STAC) and solar electrical assisted vapor compression cooling (SEVC) systems are assessed in this study by conducting the conventional and advanced exergy analyses and environmental assessment. Determining the unavoidable part of exergy destruction, as in this study, provides a unique convenience in design problems where the thermodynamic performances of distinct systems are compared. Under current technological conditions, removing the thermodynamically optimized parameters of the designed systems from the minima-maxima dichotomy and rationally evaluating the avoidable part of exergy destruction will protect the researcher from the arbitrariness of the design. The obtained results based on conventional exergy analysis in a component manner showed that priority should be given to solar technologies due to their lowest exergy efficiencies (0.16 for a photovoltaic (PV) and 0.19 for a collector) and sustainability indices (1.20 for the PV and 1.24 for the collector). Advanced exergy analysis results revealed that the exergy destruction significantly originated from the unavoidable part of the total exergy destruction of the components for the solar technologies (93.02 % for the collector and 96.41 % for the PV), cooling (92.12 % for the absorption and 98.42 % for the vapor compression), and overall system (99.92 % for the SEVC and 99.99 % for the STAC). The initial estimated carbon dioxide emissions from the STAC were 0.28 kg CO2-eq, attributed to pump power consumption. However, these emissions varied dynamically for the SEVC, ranging from 0 (when the solar PV field meets the total power) to 5.58 kg CO2-eq (when radiation is not available), depending on the power-consuming components (compressor and pumps).Engineering, Environmental || Engineering, Chemica
Decoding the Nexus: A Bibliometric Review on Sustainability, Circular Economy, and Consumer Food Waste
No full textOne of the issues that has gained importance within the scope of the United Nations Sustainable Development Goals (SDGs) is the issue of food waste. These goals, which represent very important and urgent problems to be solved at the global level, are extremely critical in terms of sustainability. Food waste, by its nature at the intersection of economic, social, and environmental sustainability goals, has become a global problem linked to key challenges in the global environment in terms of food security, climate change, malnutrition, and economic sustainability. The fact that consumers are one of the most important factors affecting food waste in the transition to a circular economy increases the importance of this study once again. Due to the lack of systematic, chronological studies showing how food waste develops over time, this study will examine the development and evolution of food waste research using a bibliometric analysis. In this way, it aims to gain a comprehensive insight into the field's current state and shed light on this highly important area of study. In addition to informing policymakers, practitioners, and consumers with the results of this research, it is also aimed to support all relevant individuals, institutions, and organizations in the efforts to combat food waste. One of the main objectives of this study is to contribute to the achievement of the United Nations Sustainable Development Goals (SDGs). For this reason, it can be stated that the research has objectives in line with SDG 12: Responsible Consumption and Production and SDG 13: Climate Action.Green & Sustainable Science & Technology || Environmental Sciences || Environmental Studie