Efficient Utilization Research Articles

Hybrid Algorithm Selection and Hyperparameter Tuning on Distribute Machine Learning Resources: Hierarchical Agent-based Approach

Algorithm selection and hyperparameter tuning are critical steps in both academic and applied machine learning (ML). These steps are becoming increasingly delicate due to the extensive rise in the number, diversity, and distributed nature of ML resources. Multi-agent systems, when applied to the design of ML platforms, bring about several distinctive characteristics, such as scalability, flexibility, and robustness, just to name a few. This article proposes a fully automatic and collaborative agent-based mechanism for selecting distributed ML algorithms and simultaneously tuning their hyperparameters. Our method builds upon an existing agent-based hierarchical ML platform and augments its query structure to support the aforementioned functionalities without being limited to specific learning, selection, and tuning mechanisms. We have conducted theoretical assessments, formal verification, and analytical study to demonstrate the correctness, resource utilization, and computational efficiency of our technique. According to the results, our solution is algorithmically correct and exhibits linear time and space complexity in relation to the size of available resources. To further verify its correctness and demonstrate its effectiveness and flexibility across a range of algorithmic options and datasets, the article also presents a series of empirical results on a system composed of 24 algorithms and 9 datasets. The findings not only highlight the efficiency and scalability of the proposed approach, but also show its flexibility and openness to responding to the dynamic and distributed ML ecosystem.

How to Effectively Cool Blade Batteries in Extreme High-Temperature Environments?

The market share of blade batteries is rising rapidly due to their high energy density, efficient space utilization, and low cost. Nevertheless, effective cooling solutions for blade batteries are crucial to ensure the safe operation of electric vehicles, especially in extreme high-temperature environments. This paper numerically investigates the effects of a cooling plate and the blade battery parameters on maximum battery temperature, maximum temperature difference, and cooling water pressure drop. Additionally, the energy efficiency of these solutions under various cooling demands is analyzed. The numerical results show that increasing the channel number and changing the flow direction does not significantly improve the cooling performance of the cooling plate. Moreover, the effect of cooling water temperature on the maximum temperature difference in blade batteries is negligible. Furthermore, increasing the cooling water mass flow rate and the rotational speed of the cooling fan is preferred when Tmax − Ta > 6 K, while reducing the cooling water temperature is more energy-efficient when Tmax − Ta < 6 K. These results are expected to offer theoretical guidance and data support for designing cooling systems for blade batteries in extreme high-temperature environments.

Investigation into Recycling of Iron Oxide Scale on H13 Steel by Low‐Temperature Carbothermal Reduction Method

The current method for recovering iron oxide scale in the steel industry is not economically optimal, especially for high‐alloy scales found in alloyed steel. This study focuses on iron oxide scale containing valuable metals like chromium (Cr), molybdenum (Mo), and vanadium (V). The required carbon addition is calculated based on the iron and chromium oxides in the scale. The effects of varying carbon additions and reduction temperatures on reduction efficiency are thoroughly examined. Kinetic studies show that as temperature and carbon increase, the rate‐limiting step shifts from interfacial diffusion to interfacial reaction. Reduction experiments assess carbon utilization, metallization rate, deoxidation rate, and removal of harmful elements. Results show that high temperatures hinder sulfur (S) and phosphorus (P) removal, and excess carbon reduces carbon utilization efficiency. Optimal conditions are a carbon ratio of 0.2174 and a temperature of 1150 °C. The carbothermic reduction product requires further refinement through conventional ladle slag systems to meet the quality standards for metallic materials. Over 65% of alloying elements are recovered, though phosphorus content remains slightly higher than in finished alloy steel. The materials from this study are suitable as high‐quality intermediates for alloy steel production.

Soil Microbial and Metabolomic Shifts Induced by Phosphate-Solubilizing Bacterial Inoculation in Torreya grandis Seedlings

Phosphorus is crucial for plant growth and development, but excess fertilizer not absorbed by plants often binds with metal ions like iron and manganese, forming insoluble compounds that contribute to soil environmental pollution. This study investigates the impact of Burkholderia sp., a phosphate-solubilizing bacterium utilized as a biofertilizer, on the fertility of T. grandis soil, alongside the associated shifts in soil metabolites and their relationship with microbial communities after inoculation. The soil microbial community structures and metabolite profiles were analyzed via amplicon sequencing and high-resolution untargeted metabolomics. The inoculation of phosphate-solubilizing bacteria led to a significant (p < 0.05) enhancement in total phosphorus, potassium, and nitrogen concentrations in the soil, with a marked increase in available phosphorus in bulk soil (p < 0.05). Moreover, the microbial community structure exhibited significant shifts, particularly in the abundance of bacterial phyla such as Acidobacteria, Chloroflexi, Proteobacteria, and the fungal phylum Ascomycota. Metabolomic analysis revealed distinct metabolites, including fatty acids, hormones, amino acids, and drug-related compounds. Key microbial taxa such as Chloroflexi, Proteobacteria, Acidobacteria, Verrucomicrobia, Mucoromycota, and Ascomycota indirectly contributed to soil phosphorus metabolism by influencing these differential metabolites. In conclusion, the application of phosphate-solubilizing bacteria offers an innovative approach to improving soil quality in T. grandis, promoting phosphorus utilization efficiency, and enhancing soil ecosystem health by optimizing microbial communities and metabolite compositions.

The design of intelligent highway transportation system in smart city based on the internet of things.

The design of intelligent expressway transportation system based on the Internet of Things is studied to improve the safety, travel experience, and operation management of expressway. The characteristics of the Internet of Things and cloud computing technology and its application on the expressway are analyzed, and the system design requirements of expressway intelligent transportation are understood. Besides, the overall architecture of the system is studied and designed. The IaaS layer, PaaS layer, and SaaS layer of the cloud platform are designed and deployed. The intelligent information system can make expressway highly informative. The simulation experiments reveal that the system only needs 120 milliseconds in accident processing time, which is far lower than the intelligent transportation system that only uses edge computing technology (201 milliseconds) and the intelligent transportation system that only uses cloud computing technology (443 milliseconds). Meanwhile, the accident response time is only 12s, which is also superior to other models. In terms of cost-effectiveness, the monthly cost of the system is 7004 yuan, with a CPU utilization rate of 53%, demonstrating good cost-effectiveness and resource utilization efficiency. In addition, compared with the existing system, the average traffic congestion time has been reduced by 25%, the traffic accident rate has been reduced by 18%, and the accident rate has been reduced by 27%. The intelligent traffic system design of expressway, expressway safety, travel service, and operation management is effectively improved by researching the intelligent traffic system design of expressway.

Research on Microgrid Optimal Scheduling Based on an Improved Honey Badger Algorithm

As global energy demands continue to grow and environmental protection pressures increase, microgrids have garnered widespread attention due to their ability to effectively integrate distributed energy sources, improve energy utilization efficiency, and enhance grid stability. Due to the complexity of internal structure, variety of energy sources, and uncertainty of load demand, the optimal scheduling problem of microgrids becomes extremely complicated. Traditional optimization methods often perform poorly in complex and dynamic microgrid environments, and it is assumed that the complexity is low or that more simplification is needed, which leads to poor convergence and local optimality when dealing with uncertainty and nonlinear problems, making intelligent optimization algorithms a crucial solution to this problem. To address the shortcomings of the traditional honey badger algorithm, such as the slow convergence speed and a tendency to fall into local optima in complex microgrid optimal scheduling problems, this paper proposes a multi-strategy improved honey badger algorithm. During the population initialization phase, a combined opposition-based learning strategy is introduced to enhance the algorithm’s exploration and exploitation capabilities. Additionally, the introduction of variable spiral factors and a linearly decreasing strategy for parameters improves the overall efficiency of the algorithm and reduces the risk of local optima. To further enhance population diversity, a hunger search strategy is employed, providing stronger adaptability and global search capabilities in varying environments. The improved honey badger algorithm is then applied to solve the multi-objective optimal scheduling problem in grid-connected microgrid modes. The simulation results indicate that the improved honey badger algorithm effectively enhances the economic and environmental benefits of microgrid operations, improving system operational stability.

An MPPT method using phasor particle swarm optimization for PV‐based generation system under varying irradiance conditions

AbstractThe dependency of photovoltaic (PV) systems‐based generation systems on constantly varying temperatures and incident sunlight affect the non‐linear behaviour of PV panels. Therefore, tracking the maximum power output of the PV panels for efficient utilization becomes a challenge, resulting in power losses and the creation of intense heating spots in the shaded areas of the PV modules when the PV modules receive varying degrees of insolation, that is, under partial shading conditions (PSCs). The inclusion of bypass diodes in parallel to each PV module mitigates this problem to an extent but leads to the formation of several peaks in the P‐V characteristics. As a result, maximum power point tracking (MPPT) to deliver maximum power at the load becomes a limitation for conventional optimization algorithms as they are normally based on hill climbing algorithm and get stuck at local maxima of the P‐V curve. Therefore, metaheuristic algorithms are used thereby eliminating the possibility of getting trapped at the local optima. However, particle swarm optimization (PSO) suffers from delayed convergence, more iterations to reach the optimal point, and random parameter selection. Hence, this study employs an improved version of PSO called Phasor‐PSO (P‐PSO) in an MPPT controller. The proposed algorithm is parameter‐less which results in reduced computational complexity and thus provides quick decision in achieving the maximum power point (MPP). The hardware‐in‐loop real‐time analysis demonstrates the supremacy of P‐PSO over PSO in faster convergence, higher efficiency, and reduced power losses during tracking under various PSCs. The P‐PSO based MPPT method will find application in grid connected and stand‐alone solar PV system with better efficiency of power transfer.

Products Distribution and Reaction Kinetics from Co-Pyrolysis of Pinewood and Polypropylene under Non-Catalytic and Catalytic Reaction Conditions

Abstract Co-pyrolysis technology offers vital pathways for efficient utilization of plastics and biomass resources to help reduce the environmental problems and energy resources issues. The pyrolysis characteristics of pinewood and polypropylene (PP) mixtures were analyzed using TGA. The results showed a decrease in first peak of the mixture with increase in PP in the mixture, while the second peak increased with an increase in PP in the mixture. The addition of a catalyst decreased the DTG peak heights. The reduction in the first peak with different catalysts was in the order: CaO/ZSM-5 >CaO >ZSM-5, while for the second peak showed: CaO >CaO/ZSM-5 >ZSM-5. The activation energy, calculated by Flynn–Wall–Ozawa (FWO), Kissinger–Akahira–Sunose (KAS), and Friedman models, revealed that ZSM-5 reduced the activation energy, whereas CaO/ZSM-5 increased the activation energy, as compared to no catalyst case. Increase of co-pyrolysis temperature reduced the yield of aldehydes, ketones, acids and esters, but increased the yield of hydrocarbons. The addition of CaO reduced the yield of ketones, phenols, esters, and acids, while it increased the yield of alcohols. The addition of ZSM-5 also decreased the yield of ketones, phenols, acids and hydrocarbons, but increased the yield of furans and alcohols. The addition of CaO/ZSM-5 specifically reduced the yield of aldehydes and alcohols. The results show the important role of the specific catalysts examined on the resulting products distribution for the same reaction condition.

Hepatic Transcriptomics of Broilers with Low and High Feed Conversion in Response to Caloric Restriction

Background: In broiler chickens, the efficient utilization of macro- and micronutrients is influenced by various metabolic pathways that are closely linked to feed efficiency (FE), a critical metric in poultry industry, with residual feed intake (RFI) as the preferred proxy. Feed restriction is considered an approach to address the underlying molecular mechanisms of feed conversion. We hypothesized that broiler chickens with divergent RFI subjected to quantitative feed restriction differ in their pattern of molecular pathways for efficient nutrient utilization in liver as post-absorptive tissue. Methods: Cobb 500FF broiler chickens divergent for RFI (n = 112) were feed-restricted from day 9 until market weight at day 33–37 post-hatch. Based on a previous trial, feed restriction levels were set at 92% (low-RFI birds) and 80% (high-RFI birds) relative to the control groups. Transcriptomic analyses of the liver were conducted. Results: Due to the interaction of the RFI group and feeding regimen, a total of 140 to 507 differentially expressed genes were identified for the respective contrasts, with implications for hepatic metabolism and cellular stress response. Although the broilers did not realize their full growth potential under restrictive feeding (12.4% reduced body weight vs. controls, p = 0.094), the gene expression patterns indicate a lower susceptibility to blood coagulation (KNG1, FGG, and FGB), suggesting that controlled and mild feed restriction could lead to health benefits in less feed-efficient broilers. Moreover, FE traits are shown to be linked to cellular detoxification processes (MGST3 and CYP2AC2) and triacylglycerol syntheses (MOGAT1 and LPIN1). Conclusions: Divergent transcriptional profiles between broiler groups under varied caloric conditions indicate potential for optimizing nutritional management strategies.

Кластерне управління проєктами: підходи, виклики та перспективи

The article examines cluster project management as a modern innovative approach to organizing and coordinating projects. It establishes that cluster management is based on the consolidation of interrelated projects into clusters for the joint use of resources, enhancing the efficiency of innovative potential utilization, and stimulating innovation. Cluster management allows for resource optimization, improved communication, the development of synergy between projects, and decision-making flexibility. The article analyzes the main types of clusters, including geographical, sectoral, vertical, and horizontal, and their impact on the success of project management and the productivity of individual projects. It also discusses the key advantages of cluster management, including resource optimization, improved coordination among participants, increased overall productivity, and flexibility. The study reveals that the cluster approach helps reduce costs, avoid duplication of efforts, promotes rapid market response, and allows organizations to effectively adapt to market challenges. In addition to the advantages, the article highlights the challenges organizations face when applying the cluster approach. These include the need to adapt to external changes, competition among cluster participants, and intellectual property protection, which can pose a threat to innovation. A new approach to cluster project management is proposed, which integrates vertical and horizontal clusters with modern digital technologies, such as monitoring and management tools. This approach enhances the effectiveness, resilience, and innovativeness of projects in the context of globalization and digitalization. The article also explores successful examples of clusters, such as the Silicon Valley technology cluster and the California wine cluster, which illustrate the benefits of collaboration within a cluster system. The article emphasizes the importance of flexibility and continuous updating of management methodologies to ensure the success of clusters in today's competitive environment.

Self-sacrificing and self-supporting biomass carbon anode–assisted water electrolysis for low-cost hydrogen production

Electrooxidation of renewable and CO2-neutral biomass for low-cost hydrogen production is a promising and green technology. Various biomass platform molecules (BPMs) oxidation assisted hydrogen production technologies have obtained noticeable progress. However, BPMs anodic oxidation is highly dependent on electrocatalysts, and the oxidation mechanism is ambiguous. Meanwhile, the complexity and insolubility of natural biomass severely constrain the efficient utilization of biomass resources. Here, we develop a self-sacrificing and self-supporting carbon anode (SSCA) using waste corncobs. The combined results from multiple characterizations reveal that the structure-property-activity relationship of SSCA in carbon oxidation reaction (COR). Theoretical calculations demonstrate that carbon atoms with a high spin density play a pivotal role in reducing the adsorption energy of the reactive oxygen intermediate (*OH) during the transition from OH- to *OH, thereby promoting COR. Additionally, the HER||COR system allows driving a current density of 400 [Formula: see text] at 1.24 V at 80 °C, with a hydrogen production electric consumption of 2.96 kWh Nm-3 (H2). The strategy provides a ground-breaking perspective on the large-scale utilization of biomass and low-energy water electrolysis for hydrogen production.

RiboSeq.Org: an integrated suite of resources for ribosome profiling data analysis and visualization.

Ribosome profiling (Ribo-Seq) has revolutionisedour understanding of translation, but the increasing complexity and volume of Ribo-Seq data present challenges for its reuse. Here, we formally introduce RiboSeq.Org, an integrated suite of resources designed to facilitate Ribo-Seq data analysis and visualisation within a web browser. RiboSeq.Org comprises several interconnected tools: GWIPS-viz for genome-wide visualisation, Trips-Viz for transcriptome-centric analysis, RiboGalaxy for data processing and the newly developed RiboSeq data portal (RDP) for centralised dataset identification and access. The RDP currently hosts preprocessed datasets corresponding to 14840 sequence libraries (samples) from 969 studies across 96 species, in various file formats along with standardised metadata. RiboSeq.Org addresses key challenges in Ribo-Seq data reuse through standardised sample preprocessing, semi-automated metadata curation and programmatic informationaccess via a REST API and command-line utilities. RiboSeq.Org enhances the accessibility and utility of public Ribo-Seq data, enabling researchers to gain new insights into translational regulation and protein synthesis across diverse organisms and conditions. By providing these integrated, user-friendly resources, RiboSeq.Org aims to lower the barrier to reproducible research in the field of translatomics and promote more efficient utilisation of the wealth of available Ribo-Seq data.

Spatiotemporal Analysis of the Coupling Relationship Between Urban Infrastructure and Land Utilization in a Shrinking City: A Case Study of Hegang, Northeast China

Globally, urbanization is accelerating, with China witnessing a significant 40% rise in urbanization rate over the past 4 decades. However, the dynamic changes in the spatial coupling between infrastructure and utilization intensity during the early, middle, and late stages of urbanization are not clear. The trajectory of development and coupling within the urbanization process is crucial for understanding issues such as urban over-saturation and urban shrinkage. Using Hegang in Northeastern China as an example, we utilized high-resolution remote sensing data, examined the construction intensity of urban land use, analyzed the degree of coupling with utilization efficiency, and clarified the dynamic evolution of the binary relationship system between development and coupling. Results show that Hegang’s construction intensity has seen a continuous rise from 1992 to 2000, with a 200.06% increase over 28 years, while its coupling with utilization efficiency has experienced a significant drop in the 21st century, suggesting a persistent decline in the utilization of buildings and a notable urban shrinkage phenomenon. Considering development status and coupling degree, we delineate a characteristic urbanization state curve for Hegang, reflecting its progression through stages of “Underdeveloped, Highly coupled,” to “Underdeveloped, Weakly coupled”, and finally to “Highly developed, Weakly coupled”, offering insights into its urban development path. This research not only establishes a foundational data groundwork for future land-use planning in Hegang but also presents a replicable template for urbanization path analysis in other cities, contributing to a broader understanding of urban development dynamics.

Soil test crop response nutrient prescription equations for improving soil health and yield sustainability—a long-term study under Alfisols of southern India

IntroductionEnhancing soil health and nutrient levels through fertilizers boosts agricultural productivity and global food security. However, careful fertilizer use is essential to prevent environmental damage and improve crop yields. The soil test crop response (STCR) is a scientific approach to fertilizer recommendation that ensures efficient use, supporting higher crop production while protecting the environment and preserving resources.MethodologyA long-term field experiment on the STCR approach was initiated in 2017 at the Zonal Agriculture Research Station, University of Agricultural Sciences, Bangalore, India. The experiment aimed to study the impact of STCR-based nutrient prescription along with farmyard manure (FYM) for a targeted yield of soybean (Glycine max), sunflower (Helianthus annuus), dry chili (Capsicum annuum), aerobic rice (Oryza sativa L.), foxtail millet (Setaria italica), okra (Abelmoschus esculentus), and kodo millet (Paspalum scrobiculatum) on yield and changes in soil health in comparison with other approaches of fertilizer recommendation.ResultsThe results showed a significant and positive impact of the integrated use of fertilizer with FYM based on the STCR approach on the productivity of all the crops and soil fertility. Significantly higher yields of soybean (23.91 q ha−1), sunflower (27.13 q ha−1), dry chili (16.67 q ha−1), aerobic rice (65.46 q ha−1), foxtail millet (14.07 q ha−1), okra (26.82 t ha−1), and kodo millet (17.10 q ha−1) were observed in the STCR NPK + FYM approach at yield level 1 compared to the general recommended dose and soil fertility rating approach. This approach outperformed the standard recommendations, enhancing nutrient uptake and efficiency across various crops. Utilizing the principal component analysis, the soil quality index effectively reflected the impact of nutrient management on soil properties, with the STCR NPK + FYM treatment at yield level 1 showing the highest correlation with improved soil physical and chemical parameters.DiscussionThe STCR approach led to improved yield, nutrient uptake, utilization efficiency, and soil health, thanks to a balanced fertilization strategy. This strategy was informed by soil tests and included factors like crop-induced nutrient depletion, baseline soil fertility, the efficiency of inherent and added nutrients through fertilizers and farmyard manure, and the success of yield-targeting techniques in meeting the nutritional needs of crops.

Energy management strategy for fuel cell hybrid tractor considering demand power frequency characteristic compensation.

The application of fuel cell tractors is expected to drive technological upgrades and sustainable development in agricultural machinery. However, fuel cell hybrid systems face issues such as slow dynamic response, low efficiency, and short lifespan. This paper proposes an energy management strategy based on signal reconstruction methods, including Empirical Mode Decomposition (EMD) and Variational Mode Decomposition (VMD), to achieve optimal energy utilization and system efficiency based on frequency response characteristics. First, we collected data on the tractor's traction force and operating speed, and calculated the required traction power using a full-machine dynamics model built in MATLAB software. We conducted frequency response characteristic analysis of the fuel cell hybrid system based on EMD and VMD, establishing an energy management controller to sequentially meet the average power demand of the fuel cell under plowing load operations, the instantaneous acceleration power demand of the power battery, and the real-time compensation power demand of the supercapacitor. The results show that the EMD strategy exhibits good stability, while the VMD strategy performs better in terms of hydrogen consumption. Under the VMD strategy, the hybrid system achieves a maximum output efficiency of 55.0% with a total hydrogen consumption of 750g. Compared to the EMD strategy, the maximum efficiency of the system increases by 27.31%, and hydrogen consumption decreases by 3.49%. This study provides a new theoretical foundation and technological route for the application of fuel cell hybrid systems in the field of agricultural machinery.

High‐Throughput Screening of Dual‐Atom Catalysts for Methane Combustion: A Combined Density Functional Theory and Machine‐Learning Study

AbstractCeria‐supported precious metal catalysts have undergone extensive investigation for the catalytic methane combustion. However, it remains a significant challenge to achieve both highly synergistic oxidation activity and efficient atom utilization remains a challenge for commonly used supported nanoparticles and single‐atom catalysts. Dual‐atom catalysts (DACs) emerges as a frontier of advanced catalysts, presenting unique catalytic properties that benefit from the synergy of neighboring metal sites. In this study, 361 ceria‐supported DACs (M1M2/CeO2) encompassing combinations of 19 transition metals are systematically explored. Using high‐throughput density functional theory calculations, the structures, stability as well as activity of M1M2/CeO2 are assessed. Notably, Au1Ga1/CeO2 is identified as a promising DAC exhibiting high activity for methane total oxidation, substantiated by comprehensive DFT‐calculated reaction pathways. Furthermore, employing six machine‐learning algorithms, the structure‐properties relationship is explored within ceria‐based DACs and highlight the importance of oxidation states and atomic radii of doped metals as the descriptors. The trained model by computational dataset exhibits high accuracy and predict a more active Mn1Au1/CeO2 than those screened using only DFT datasets. The high‐throughput strategy demonstrated in this work not only provides insights into the rational design of methane oxidation catalysts, but also paves the way for exploring DACs for diverse applications.

Implementation of AI Transportation Routing in Reverse Logistics to Reduce CO2 Footprint

Purpose: This study explores the implementation of artificial intelligence (AI) in transportation routing within reverse logistics to reduce CO2 emissions. By optimizing routing processes, we aim to enhance the efficiency of logistics operations while minimizing environmental impact. Methodology: we analyze the implications of improved transportation strategies. Furthermore, we discuss the intersection of technological innovation and environmental economics in shaping sustainable practices. Findings: The findings underscore the importance of AI-driven decision-making frameworks (D81) and their role in advancing IT management in logistics. The study on AI-driven transportation routing in reverse logistics highlights significant implications for both businesses and the environment. By optimizing routes, AI reduces fuel consumption and CO₂ emissions, helping companies meet sustainability goals. This leads to cost savings, as AI improves route efficiency and fleet utilization, while also enhancing operational efficiency. AI’s ability to predict and adjust to real-time conditions ensures scalability and adaptability, making it easier to handle increasing returns volumes in e-commerce. Moreover, AI enables better inventory management and resource planning, reducing the need for urgent shipments. It also paves the way for innovations like autonomous vehicles, further transforming reverse logistics. Unique Contribution to Theory, Practice and Policy: The study shows how AI can help businesses comply with environmental regulations, providing a competitive edge. Ultimately, AI integration in reverse logistics contributes to both operational improvements and environmental sustainability, reinforcing corporate social responsibility.

Use of Data Augmentation and Fine-tuned Transfer Learning Models for Classification of Cervical Cancer

Innovative strategies for early and accurate diagnosis of cervical cancer continue to be a global health priority. In this research, we present a powerful framework for improving cervical cancer classification by combining data augmentation methods with tailored transfer learning models. The lack of properly labelled medical data is a major roadblock to developing reliable diagnosis models. The geometric transformations, contrast tweaks, and noise addition that we use to increase the dataset artificially help us overcome this difficulty. The model's robustness is improved as a result of the additional exposure to diverse real-world circumstances provided by this supplemented dataset. We use a transfer learning strategy based on pre-trained convolutional neural networks (CNNs) to harness the potential of deep learning. These networks are well-suited for medical image analysis due to their ability to quickly and accurately identify features in a wide variety of images. Our expanded cervical cancer dataset is used to fine-tune these algorithms for their diagnostic purpose. Our experiments show that utilising both data augmentation and fine-tuned transfer learning considerably raises the accuracy of cervical cancer categorization. In terms of sensitivity, specificity, and overall accuracy, the model performs at a state-of-the-art level. This suggests that it may be useful in the diagnosis and early identification of cervical cancer. Additionally, our method demonstrates the significance of efficient data utilisation and model adaptability in the field of medicine. Our method is scalable and affordable, and it can be easily implemented in healthcare settings since we handle the problem of data scarcity and take advantage of the features of pre-trained deep learning models.

Evaluation of Capital Utilization Efficiency in Agricultural Listed Firms Based on a Three-Stage Data Envelopment Analysis

Rural revitalization is intricately tied to national prosperity, with listed companies in agriculture, forestry, animal husbandry, and fisheries playing pivotal roles in driving this transformative agenda. However, these sectors face critical challenges, notably resource constraints and environmental pressures, which underscore the necessity of evaluating capital utilization efficiency. This study employs a combined three-stage DEA model and Malmquist index to analyze capital utilization efficiency across 85 listed companies within China’s agriculture, forestry, animal husbandry, and fishery industries over the period 2018–2022. The primary objectives are to quantify capital utilization efficiency, identify pathways for industry optimization, advance ecological agricultural sustainability, and offer valuable insights to investors and policymakers. Key findings include: (1) When grouped by sub-sector and adjusted to account for external environmental factors and random disturbances, the initial comprehensive capital utilization efficiency in agricultural companies was found to be significantly overestimated. Subsequent third-stage adjustments revealed decreases in average comprehensive efficiency, technical efficiency, and scale efficiency by 29.79%, 3.03%, and 27.37%, respectively, largely driven by a marked decline in scale efficiency, ultimately diminishing overall efficiency. (2) The capital utilization efficiency in agricultural trading firms remains suboptimal, with scale efficiency posing a critical limitation. High dependency on government subsidies and excess specialized personnel further constrain efficiency improvements. (3) Dynamic analysis using the DEA-Malmquist model indicates low total factor productivity in the agricultural sub-sector, primarily due to inefficiencies in capital management and suboptimal scale allocation. These findings underscore the need for targeted strategies to enhance resource allocation and management, bolster talent development and financial management frameworks, and drive technological research and development, innovation, and efficient capital allocation across the agriculture, forestry, animal husbandry, and fishery sectors.

Construction of Iron-Modified Lignin-Based Nanomicrocapsules for Enhancing the Functionality of Natural Product-Based Pesticides.

To address the issue of low pesticide utilization owing to poor dispersibility, low leaf surface adhesion, and poor transport within plants, this study exploits electrostatic interactions between sodium lignosulfonate (SL) and dodecyltrimethylammonium chloride (DTAC) to induce self-assembly, followed by iron ion (Fe3+) chelation and loading with a natural product-based pesticide, rosin-based triazole derivative (RTD), yielding RTD@SL-DTAC-Fe nanomicrocapsules (NMs). It is worth noting that the presence of Fe3+ enhances the dispersibility of the NMs. The water dispersibility and photostability of RTD are significantly improved after encapsulation, and a stimulus response to laccase is achieved. Leaf-washing experiments confirm the enhanced adhesion of RTD@SL-DTAC-Fe NMs to the surface of rice plant leaves compared to that of free RTD. Fluorescently labeled NMs exhibit bidirectional transport within rice plants, and RTD@SL-DTAC-Fe NMs demonstrates better transport performance than RTD. In vitro and in vivo antifungal tests indicate that encapsulation by NMs significantly enhanced pesticide activity. Field trials demonstrate that NMs exhibited prolonged efficacy compared to RTD. Finally, the safety evaluation confirms the environmental friendliness of the NMs. This study provides valuable insight for optimizing and improving the utilization efficiency and biosafety of natural product-based pesticides.