Performance Benchmarks Research Articles

Fake news detection in low-resource languages: A novel hybrid summarization approach

The proliferation of fake news across languages and domains on social media platforms poses a significant societal threat. Current automatic detection methods for low-resource languages (e.g., Swahili, Indonesian and other low-resource languages) face limitations due to two factors: sequential length restrictions in pre-trained language models (PLMs) like multilingual bidirectional encoder representation from transformers (mBERT), and the presence of noisy training data. This work proposes a novel and efficient multilingual fake news detection (MFND) approach that addresses these challenges. Our solution leverages a hybrid extractive and abstractive summarization strategy to extract only the most relevant content from news articles. This significantly reduces data length while preserving crucial information for fake news classification. The pre-processed data is then fed into mBERT for classification. Extensive evaluations on a publicly available multilingual dataset demonstrate the superiority of our approach compared to state-of-the-art (SOTA) methods. Our analysis, both quantitative and qualitative, highlights the strengths of this method, achieving new performance benchmarks and emphasizing the impact of content condensation on model accuracy and efficiency. This framework paves the way for faster, more accurate MFND, fostering more robust information ecosystems.

A Methodology for Speaker Diazaration System Based on LSTM and MFCC Coefficients

Research on Speaker Identification is always difficult. A speaker may be automatically identified using by comparing their voice sample with their previously recorded voice, the machine learning strategy has grown in favor in recent years. Convolutional neural networks (CNN) , deep neural networks (DNN) are some of the machine learning techniques that has employed recently. The article will discuss a successful speaker verification system based on the d-vector to construct a new approach based on speaker diarization. In particular, in this article, we use the concept of LSTM to cluster the speech segments using MFCC coefficients and identify the speakers in the diarization system. The proposed system will be evaluated using benchmark performance metrics, and a comparative study will be made with other models. The need to consider the LSTM neural network using acoustic data and linguistic dialect is considered. LSTM networks could produce reliable speaker segmentation outputs.

Evaluation of Prognostic Scores in Patients with Head Trauma in the Emergency Department

Background: The aim of this study was to investigate the effectiveness of Glasgow coma scale (GCS), GCS-motor component (mGCS), and FOUR (Full Outline of Un-responsiveness) Scores in predicting the prognosis of patients who presented to the emergency department with head trauma. Methods: In this prospective cross-sectional study, was obtained to collected data of patients with head trauma, who presented to the emergency department. Participants’ demographic data, medical history, GCS, FOUR scores, the duration of emergency department stays, as well as 24-hour, 7-day, and 28-day mortality rates were recorded on the case report forms. Results: Data from 302 patients were used to develop a risk score for detecting significant brain pathology via computed tomography (CT) scans. The regression model, incorporating total GCS and sex-based variables, explained 22.5% of variance and accurately classified 91.1% of cases. The model's area under the curve for detecting significant pathology via CT was 0.714. Conclusion: GCS, mGCS, and FOUR scores did not achieve the necessary the diagnostic performance benchmark to be used alone to predict or exclude clinically significant brain injury in patients with head trauma.

Approximate Dynamic Programming for Control of Wave Energy Converters With Implementation and Validation on a Point Absorber Prototype

It is recognized that the application of constrained optimal control for wave energy converters (WECs), represented by model predictive control (MPC), is hindered by its computation burden resulting from online optimization, especially when the model contains nonlinearities. In this article, a novel control solution based on approximate dynamic programming (ADP) is proposed, in which an explicit controller is solved by model sampling and neural network training. In this way, the major computation load is moved offline to enable fast online execution, and the need for wave prediction is eliminated. The effectiveness of ADP is verified by a simulation case study based on a nonlinear WEC system. Using nonlinear MPC as the performance benchmark, over 90% energy efficiency is achieved by ADP under various sea states with constraint satisfaction and dramatically faster computation. The ADP controller is further experimentally validated on a newly constructed WEC prototype through wave tank testing, where successful implementation and satisfactory performance are demonstrated.

A spherical fuzzy-based DIBR II-AROMAN model for sustainability performance benchmarking of wind energy power plants

Wind energy power plants (WEPPs) are renewable energy generation facilities that are increasingly used today. The performance evaluations of these energy plants are generally based on technical performance. Although there are studies in the literature focusing on determining the technical performance of WEPPs, there is a lack of research on evaluating their sustainable performance. The primary motivation of this study is to develop a decision model for identifying the sustainable performance of WEPPs. The model incorporates fuzzy logic and multi-criteria decision-making (MCDM) approaches, incorporating expert opinions and both qualitative and quantitative criteria. To determine the influence levels of experts, spherical fuzzy (SF) sets are utilized. The weighting of criteria is achieved using the SF-defining interrelationships between ranked criteria II (DIBR II) method, which is a novel extension of the DIBR II method based on SF sets. Additionally, an SF-alternative ranking order method accounting for two-step normalization (AROMAN) method is developed for ranking the sustainable performance of WEPPs, building upon the AROMAN method, and adapting it to SF sets. This hybrid model is named the SF-DIBR II-AROMAN hybrid model. The step-by-step procedures of this model are presented in the research, and an algorithm for these procedures is developed. To demonstrate its practicality, a case study is conducted, focusing on WEPPs with a capacity ranging from 25 to 150 megawatts in Çanakkale, Turkey. According to the research results, among the fifteen WEPPs in the Çanakkale region, “Saros WEPP” is identified as having the best one. The robustness of the obtained results is ensured through sensitivity analyses. Based on the results of two sensitivity analysis scenarios, “Saros WEPP” is found to have the highest sustainable performance. The SF-DIBR II-AROMAN hybrid model results are compared with different alternative ranking method results, highlighting its superior aspects. Overall, SF-DIBR II-AROMAN is consistent and robust. The research also provides insights and managerial implications, as well as explains the benefits of the proposed model for evaluating WEPPs’ sustainable performance.

Sustainability risk assessment in manufacturing: A Life Cycle Assessment-based Failure Mode and Effects Analysis approach

This study introduces a methodology integrating Life Cycle Assessment (LCA) with Failure Mode and Effects Analysis (FMEA) to identify and assess environmental sustainability risks in manufacturing processes in alignment with the Corporate Sustainability Reporting Directive (CSRD) and European Sustainability Reporting Standards (ESRS). In contrast to existing top-down approaches at sector and company level, this methodology specifically addresses the challenge of transitory sustainability risks in manufacturing, driven by rapid changes in regulatory frameworks and market expectations, by providing a robust and adaptable framework for comprehensive, bottom-up risk assessment directly on the manufacturing shop floor. Through a detailed literature review, the paper identifies gaps in current risk assessment methods and proposes a novel approach that enhances traditional FMEA by incorporating LCA results, thereby facilitating a comprehensive, quantitative analysis of sustainability risks. The practical application of this methodology is demonstrated through a case study, underscoring its effectiveness in real-world settings and potential for broader applications in sustainable manufacturing. The results presented in the paper also include quantitative sustainability performance benchmarks for the EU28 industrial sector, offering a strategic tool for manufacturers to improve sustainability practices, align operations with environmental standards, and mitigate transitory sustainability risks. This research contributes to the sustainable evolution of manufacturing, providing a methodological foundation for understanding and managing sustainability risks.

A Detailed Historical and Statistical Analysis of the Influence of Hardware Artifacts on SPEC Integer Benchmark Performance

A Detailed Historical and Statistical Analysis of the Influence of Hardware Artifacts on SPEC Integer Benchmark Performance

Performance Benchmarking of Redpanda an event streaming platform using SBK and Open Messaging Frameworks

Redpanda is an event-driven, publish/subscribe based messaging system. The two main performance benchmarking frameworks, Storage Benchmark Kit (SBK) and open messaging benchmark are used here to test the performance benchmarking of redpanda’s event processing system. SBK is a software framework for measuring any storage device or client's performance benchmarking and open messaging benchmark framework which supports benchmarking for a wide variety of messaging platforms. Well formulated fine tuning system for benchmarks is still an open research problem. In this paper, SBK and open messaging frameworks have been experimentally compared with respect to their messaging capabilities (Message sending and receiving throughput) on redpanda’s event processing system. A number of experimental test scenarios have been conducted on data streaming workload, the performance evaluation results revealed that SBK gives maximum throughput performance compared to open messaging benchmark.

Evaluation of the Quality of Pharmaceutical Services Based on Minimum Service Standards at Pandanarang Boyolali Regional General Hospital in 2023

Minimum Pharmaceutical Service Standards are very important because they are performance benchmarks that are used as guidelines for pharmaceutical personnel in providing pharmaceutical services. This research aims to determine the suitability of SPM for hospitals in the pharmaceutical sector on the indicators of length of time for prescription service, customer satisfaction, and suitability of prescriptions with the formulary as well as influencing factors at the Pandanarang Boyolali Regional General Hospital in order to help the hospital in improving the quality of better pharmaceutical services. This research uses descriptive analytical research conducted using quantitative research methods. The sample for this research was 300 outpatient patients/prescriptions for indicators of prescription suitability and waiting time, 150 respondents for indicators of customer satisfaction at Pandanarang Boyolali Regional Hospital. The results of the study showed that in the indicator of the length of time for prescription service, the average time for non-concocted prescriptions was 16.33 ± 15,114 minutes and for concocted prescriptions it was 29.85 ± 27,958 minutes, for the patient satisfaction indicator the results were 94.03%, as well as for the indicator The conformity of the prescription with the formulary was 94.72%. Based on the research results obtained, it can be concluded that according to the Decree of the Minister of Health of the Republic of Indonesia Number 129/MENKES/SK/II/2008 concerning Minimum Service Standards for Hospitals in the Pharmaceutical Sector which has been determined by the Ministry of Health of the Republic of Indonesia that the Pandanarang Boyolali Regional General Hospital is in the indicator The conformity of prescriptions with the formulary does not meet standards, the average waiting time for prescription services meets standards, and the level of customer satisfaction with pharmacy services in pharmaceutical installations meets standards

Optimizing latent graph representations of surgical scenes for unseen domain generalization.

Advances in deep learning have resulted in effective models for surgical video analysis; however, these models often fail to generalize across medical centers due to domain shift caused by variations in surgical workflow, camera setups, and patient demographics. Recently, object-centric learning has emerged as a promising approach for improved surgical scene understanding, capturing and disentangling visual and semantic properties of surgical tools and anatomy to improve downstream task performance. In this work, we conduct a multicentric performance benchmark of object-centric approaches, focusing on critical view of safety assessment in laparoscopic cholecystectomy, then propose an improved approach for unseen domain generalization. We evaluate four object-centric approaches for domain generalization, establishing baseline performance. Next, leveraging the disentangled nature of object-centric representations, we dissect one of these methods through a series of ablations (e.g., ignoring either visual or semantic features for downstream classification). Finally, based on the results of these ablations, we develop an optimized method specifically tailored for domain generalization, LG-DG, that includes a novel disentanglement loss function. Our optimized approach, LG-DG, achieves an improvement of 9.28% over the best baseline approach. More broadly, we show that object-centric approaches are highly effective for domain generalization thanks to their modular approach to representation learning. We investigate the use of object-centric methods for unseen domain generalization, identify method-agnostic factors critical for performance, and present an optimized approach that substantially outperforms existing methods.

Advancing Deep Learning: A Comprehensive Comparative

This manuscript conducts an in-depth comparative analysis of three seminal works in the realm of Convolutional Neural Network (CNN) accelerators using Field-Programmable Gate Arrays (FPGAs). It meticulously evaluates the architectural designs, performance benchmarks, and innovative methodologies employed in each study. The analysis delves into the intricacies of these works, spotlighting their distinct and innovative contributions to the advancement of FPGA-based CNN acceleration. By critically appraising the strengths and limitations of each approach, this paper synthesizes key insights, shedding light on prevalent trends and identifying persistent challenges within this rapidly evolving field. It discusses how each work navigates the trade-offs between efficiency, flexibility, and computational power, which are inherent in FPGA-based systems. Moreover, this comparative study serves as a comprehensive resource that encapsulates the state-of-the-art in FPGA-accelerated CNNs. It not only benchmarks the current technological landscape but also charts potential pathways for future research endeavors. By amalgamating diverse perspectives and methodologies from leading research, the paper aims to inspire and guide ongoing and future investigations, driving innovation in the optimization and application of FPGA-based CNN accelerators. This endeavor is particularly timely, given the escalating demand for high-performance, energy-efficient computational models in domains ranging from autonomous systems to real-time data analytics.

Optimizing Energy Arbitrage: Benchmark Models for LFP Battery Dynamic Activation Costs in Reactive Balancing Market

This study introduces a novel benchmark model for lithium iron phosphate (LFP) batteries in reactive energy imbalance markets, filling a notable gap by incorporating comprehensive operational parameters and market dynamics that are overlooked by conventional models. Addressing the absence of a holistic benchmark for energy-storage systems in electricity markets, this research focuses on the integration of LFP batteries, considering their unique characteristics and market responsiveness. Regression and regularization techniques, coupled with temporal cross-validation, were employed to ensure model robustness and accuracy in predicting energy trading outcomes. This methodological approach allows for a nuanced analysis of battery degradation, power capacity, energy content, and real-time market prices. The model, validated using Belgium’s system imbalance market data from the 2020–2023 period, incorporates both capital and operational expenditures to assess the economic and operational viability of LFP battery energy-storage systems (BESSs). The findings reveal that considering a broader range of operational parameters in energy arbitrage, beyond just the usual energy prices and round-trip efficiency, significantly influences the cost-effectiveness and performance benchmarking of energy storage solutions. This paper advocates for the strategic use of LFP batteries in energy markets, highlighting their potential to enhance grid stability and energy trading profitability. The proposed benchmark model serves as a critical tool for energy traders, providing a detailed framework for informed decision making in the evolving landscape of energy storage technologies.

Q-Learning-Based Priority Dispatching Rule Preference Model for Non-Permutation Flow Shop

Non-permutation flow shop scheduling (NPFS) is an extension of the traditional permutation flow shop scheduling, with a broader solution space. The effectiveness of reinforcement learning in solving flow shop scheduling problems has been demonstrated through its powerful combinatorial optimization capabilities. However, the design and training of the end-to-end policy network is complex, leading to long online training time and limited adaptability of offline training. To overcome these problems, we introduced a NPFS dynamic decision-making process and then proposed a novel NPFS method that combines the Q-learning algorithm with the priority dispatching rule (PDR) set. The NPFS dynamic decision-making process involves decomposing the entire process into multiple sub-job queues for scheduling. The PDR demonstrates better scheduling performance when applied to smaller job queues. By utilizing the Q-learning algorithm, PDRs with superior performance are assigned to sub-scheduling queues based on the generation process of sub-job sequences, resulting in an optimized NPFS strategy. The limited number of PDRs in the PDR set and the small number of sub-job queues in the NPFS process contribute to the efficiency of Q-learning in solving NPFS problem. Finally, we demonstrate the superiority of the proposed NPFS method by a series of numerical experiments using the machine-based assignment PDR method and NSGA-II algorithms as performance benchmarks.

The USCENTCOM Walking Blood Bank Performance Benchmark and Anticipated Benefit of Universal Low Titer Type O Screening.

The walking blood bank (WBB) is a system for emergency blood acquisition from nearby donors if a patient's blood needs exceed the immediate supply. USCENTCOM medical units will perform a walking blood bank if immediate blood requirements exceed the local supply. A benchmark WBB performance time was needed to provide a training goal for military WBB exercises. An expeditionary WBB performance time benchmark was created from prospective measurements of USCENTCOM medical unit performance times over 9months. The mean total time, and new performance benchmark, for a WBB in USCENTCOM was 41.4min +/- 13.2min. USCENTCOM time from donor arrival to a transfusable unit mean time was 34.4 +/- 12.1min. Expeditionary medical units conducting a WBB should expect to meet or exceed the provided benchmark.

Heating load prediction in buildings using decision tree machine learning method

In recent years, the burgeoning imperative of energy-efficient building management practices has surged dramatically, underscoring an urgent mandate for comprehensive studies that integrate cutting-edge optimization algorithms with precise heating load forecasting techniques. These studies are not merely endeavors; they represent concerted efforts to increase building energy efficiency and address mounting concerns regarding sustainability and resource utilization. In the intricate domain of heating, ventilation, and air conditioning (HVAC) systems, energy optimization challenges are being meticulously confronted through rigorous exploration and the application of innovative problem-solving methodologies. This pioneering study introduces groundbreaking methodologies by seamlessly integrating two state-of-the-art optimization algorithms— the Red Fox Optimization and the Golden Eagle Optimizer— with the Decision Tree model. This fusion is aimed at enhancing the accuracy of heating load predictions and streamlining HVAC system optimization processes, marking a significant leap toward achieving heightened energy efficiency and operational efficacy in building management practices. The study emphasizes the significance of precise heating load prediction in advancing energy efficiency, realizing cost savings, and fostering environmental sustainability in building management. Furthermore, it delves into the multifaceted impact of various building features on heating load, encompassing variables such as glazing area, orientation, height, relative compactness, roof area, surface area, and wall area. These insights furnish actionable intelligence for refined decision-making processes in both building design and operation. Based on the results, the DT single model experienced the weakest performance among the three models, with R2 = 0.975 and RMSE = 1.608. The model DTFO (DT + FOX) achieves an extraordinary R2 value of 0.996 and RMSE value of 0.961 for heating load prediction, surpassing the performance benchmarks set by other models. This achievement holds considerable promise for aiding engineers in crafting energy-efficient buildings, particularly within the swiftly evolving landscape of smart home technologies.

Xputer: bridging data gaps with NMF, XGBoost, and a streamlined GUI experience.

The rapid proliferation of data across diverse fields has accentuated the importance of accurate imputation for missing values. This task is crucial for ensuring data integrity and deriving meaningful insights. In response to this challenge, we present Xputer, a novel imputation tool that adeptly integrates Non-negative Matrix Factorization (NMF) with the predictive strengths of XGBoost. One of Xputer's standout features is its versatility: it supports zero imputation, enables hyperparameter optimization through Optuna, and allows users to define the number of iterations. For enhanced user experience and accessibility, we have equipped Xputer with an intuitive Graphical User Interface (GUI) ensuring ease of handling, even for those less familiar with computational tools. In performance benchmarks, Xputer often outperforms IterativeImputer in terms of imputation accuracy. Furthermore, Xputer autonomously handles a diverse spectrum of data types, including categorical, continuous, and Boolean, eliminating the need for prior preprocessing. Given its blend of performance, flexibility, and user-friendly design, Xputer emerges as a state-of-the-art solution in the realm of data imputation.

Procedural factors associated with successful recanalization in patients with acute ischemic stroke treated with endovascular thrombectomy-a nationwide register-based observational study.

Several studies have addressed technical aspects of endovascular thrombectomy (EVT), but it is not well known how procedural factors contribute to technical success in routine healthcare. The aim was to explore factors associated with technically successful EVT on nationwide scale. We did an observational register-based study assessing factors associated with technical success off anterior circulation EVT in Sweden. The main outcome was successful recanalization defined as modified treatment in cerebral ischemia score 2b-3. The association between baseline and treatment variables and successful recanalization were explored using Chi-square(d) test and univariable logistic regression. Multivariable logistic regression was used to define predictors of successful recanalization. The study included 3211 patients treated during 2015 to 2020. Successful recanalization was achieved in 83.1% (2667) with a gradual improvement in technical outcome over the period. After adjustment for age and occlusion location, thet use of general anesthesia, balloon guide catheter (BGC) and an operator with an overall success rate of >85% were independent predictors of successful recanalization. An overall operator success rate of <80% or 80-85%, and an annual center volume lower than 50 were predicitors of recanalization failure. This study illustrates factors associated with procedural success in endovascular thrombectomy on a nationwide scale including the use of general anesthesia, BGC, annual center volumes >50 cases per year and the overall success rate of the individual operator. It highlights the potential benefit of systematic performance measurements, benchmarking, and continuous training to bring all centers and operators to the highest level of performance.

Real-World Performance Benchmarks in the Reading Ability of Prescription Drug Labels by Community-Dwelling Older Adults

Background: Despite known recommendations regarding standards for print size and the intuitive importance of vision in reading prescription labels, the predictive nature of vision and prescription label readability remains largely undefined. Furthermore, while the importance of vision is recognized, various demographic factors associated with the ability to read prescription labels have not been fully elucidated. Objective: Describe relationships between visual acuity, point size, and readability of prescription labels and provide insight into demographic factors associated with prescription label readability. Methods: Cross-sectional examination of prescription label readability by older, community-dwelling adults. Subjects were evaluated as to demographics, visual acuity, and ability to read test instruments consisting of unaltered prescription label features of five medications dispensed by community pharmacies and two drug samples. Descriptive statistics in conjunction with a logit predictive model were employed for data analysis. Results: Instructions for medication use were most recognizable, identified and correctly read by 95.60% of the study cohort while directions for the use of drug samples were lowest (34.91%). Among prescription label features, auxiliary labels consistently demonstrated poor readability. Level of visual acuity was statistically related to the ability to read prescription labels while identifying prescription label components increased proportionally with point size. Race, gender, and history of a recent eye examination were statistically significant predictors of prescription label reading ability. Visual acuity alone was found to explain approximately 26% of the variablity in ability to read Rx labels. Conclusion: Visual acuity is predictive of the ability to access Rx label information and should be considered a modifiable variable for improving prescription label reading ability amenable by appropriate eye care and spectacle correction.

Long-term variations of air pollutants and public exposure in China during 2000–2020

Since 2000, China has faced severe air pollution challenges，prompting the initiation of comprehensive emission control measures post-2013. The subsequent implementation of these measures has led to remarkable enhancements in air quality. This study aims to enhance our understanding of the long-term trends in fine particulate matter (PM2.5) and gaseous pollutants of ozone (O3) and nitrogen dioxide (NO2) across China from 2000 to 2020. Utilizing the Community Multiscale Air Quality (CMAQ) model, we conducted a nationwide analysis of air quality, systematically quantifying model predictions against observations for pollutants. The CMAQ model effectively captured the trends of air pollutants, meeting recommended performance benchmarks. The findings reveal variations in pollutant concentrations, with initial increases in PM2.5 followed by a decline after 2013. The proportion of the population living in high PM2.5 concentrations (>75 μg/m3) decreased to <5 % after 2015. However, during the period from 2017 to 2020, around 40 % of the population continued to live in regions that did not meet the criteria for Chinese air quality standards (35 μg/m3). From 2000 to 2019, fewer than 20 % of the population met the WHO standard (100 μg/m3) for MDA8 O3. In 2000, 77 % of the population met the NO2 standard (<20 μg/m3), a figure that declined to 60 % between 2005 and 2014, nearly reaching 70 % in 2020. This study offers a comprehensive analysis of the changes in pollutants and public exposure in 2000–2020. It serves as a foundational resource for future efforts in air pollution control and health research.

High‐Power‐Density Hybrid Acid/Alkali Zinc–Air Battery for High‐Efficiency Desalination

AbstractThe electrochemical desalination technique is recognized as a promising solution to alleviate freshwater shortages, challenges yet persists in achieving optimal energy efficiency and cost‐effectiveness. Herein, a hybrid acid/alkali zinc air desalination battery (hAA‐ZADB) capable of concurrent desalination and high‐power density is reported. To improve cathodic efficiency and cost‐effectiveness, an electrocatalyst with dual atomic Fe–Mn sites on porous dodecahedral carbon (Mn‐Fe/p‐DC) is fabricated through a simple direct pyrolysis strategy for oxygen reduction reaction (ORR). The Mn–Fe/p‐DC‐900 electrocatalyst demonstrates exceptional electrocatalytic activity (E1/2 = 0.8 V in 0.5 m H2SO4) for ORR. This innovative hybrid acid/alkali cell design, coupled with advanced electrocatalysts, empowers the hAA‐ZADB system to achieve outstanding performance benchmarks with a high open circuit voltage of 2.22 V, an impressive power density of 375 mW cm−2, and notably elevated energy output of 106.9 kJ mol−1 even at a current density of 100 mA cm−2 during desalination. Distinguishing this work is its additional functionality, evident in a rapid salt removal rate of 3.64 mg cm−2 min−1 during desalination, achieving an impressive 88.67% removal of 0.6 M NaCl. This study highlights the promising potential of employing metallic air batteries for a self‐powered desalination technique applicable to specific scenarios.