Performance Issues Research Articles

Deep Encodings vs. Linguistic Features in Lexical Complexity Prediction

AbstractIn this work, we present a novel approach to lexical complexity prediction (LCP) that combines diverse linguistic features with encodings from deep neural networks. We explore the integration of 23 handcrafted linguistic features with embeddings from two well-known language models: BERT and XLM-RoBERTa. Our method concatenates these features before inputting them into various machine learning algorithms, including SVM, Random Forest, and fine-tuned transformer models. We evaluate our approach using two datasets: CompLex for English (a high-resource language) and CLexIS2 for Spanish (a relatively low-resource language in ), allowing us to study performance issues from a cross-lingual perspective. Our experiments involve different combinations of linguistic features with encodings from pretrained deep learning models, testing both token-based and sequence-related encodings. The results demonstrate the effectiveness of our hybrid approach. For the English CompLex corpus, our best model achieved a mean absolute error (MAE) of 0.0683, representing a 29.2% improvement over using linguistic features alone (MAE 0.0965). On the Spanish CLexIS2 corpus, we achieved an MAE of 0.1323, a 19.4. These findings show that handcrafted linguistic features play a fundamental role in achieving higher performance, particularly when combined with deep learning approaches. Our work suggests that hybrid approaches should be considered over full end-to-end solutions for LCP tasks, especially in multilingual contexts.

Performance issue monitoring, identification and diagnosis of SaaS software: a survey

Performance issue monitoring, identification and diagnosis of SaaS software: a survey

Reliability of plastid and mitochondrial localisation prediction declines rapidly with the evolutionary distance to the training set increasing.

Mitochondria and plastids import thousands of proteins. Their experimental localisation remains a frequent task, but can be resource-intensive and sometimes impossible. Hence, hundreds of studies make use of algorithms that predict a localisation based on a protein's sequence. Their reliability across evolutionary diverse species is unknown. Here, we evaluate the performance of common algorithms (TargetP, Localizer and WoLFPSORT) for four photosynthetic eukaryotes (Arabidopsis thaliana, Zea mays, Physcomitrium patens, and Chlamydomonas reinhardtii) for which experimental plastid and mitochondrial proteome data is available, and 171 eukaryotes using orthology inferences. The match between predictions and experimental data ranges from 75% to as low as 2%. Results worsen as the evolutionary distance between training and query species increases, especially for plant mitochondria for which performance borders on random sampling. Specificity, sensitivity and precision analyses highlight cross-organelle errors and uncover the evolutionary divergence of organelles as the main driver of current performance issues. The results encourage to train the next generation of neural networks on an evolutionary more diverse set of organelle proteins for optimizing performance and reliability.

What automatic speech recognition can and cannot do for conversational speech transcription

Transcripts are vital in any research involving conversation. Most transcription is conducted manually, by experts; a process which can take many times longer than the conversation itself. Recently, there has been interest in using automatic speech recognition (ASR) to automate transcription, driven by the wide availability of ASR platforms such as OpenAI’s Whisper. However as studies typically focus on metrics such as the word error rate, there is a lack of detail about ASR transcript quality and the practicalities of ASR use in research. In this paper we review six state-of-the-art ASR technologies, three commercial and three open-source. We assess their capabilities as automatic transcription tools. We find that the commercial ASR systems mostly capture an accurate representation of what was said, and overlapping speech is handled well. Unlike prior work, we show that commercial ASR also preserves the location, but not necessarily the spelling of a large majority of non-lexical tokens: short words such as uh-hum which play vital roles in conversation. We show that the open-source ASR systems produce substantially more errors than their commercial counterparts. However, we highlight how the cost and privacy advantages of open-source ASR may outweigh performance issues in certain applications. We discuss practical considerations for ASR deployment in research, concluding that present ASR technology cannot yet replace the trained transcriber. However, a high-quality initial transcript generated by ASR can provide a good starting point and may be further refined by manual correction. We make all ASR-generated transcripts available for future research in the supplementary material.

Eco-friendly, stable, and high-performance biochar prepared by a twice-modification scheme: Saccharification of raw materials & thermal air oxidation of biochar

Organic pollutants, such as phenolic compounds, pose significant risks to both the environment and human health. While biochar is an effective adsorbent for removing these pollutants, its dissolved solid (DS) components can lead to the loss of functional groups, structural disintegration, unstable performance, and environmental issues. This study introduces a twice-modification scheme designed to produce a biochar (BC-M) that combines high stability with superior performance. The process begins with the preparation of a stable biochar from cellulase-treated lignocellulose. This precursor biochar is then subjected to thermal air oxidation to enhance its oxygen-containing functional groups, thereby improving its adsorption capabilities. A mathematical model was developed to explore the relationship between different thermal air oxidation conditions and the properties of BC-M, aiming to optimize both adsorption capacity and DS. The model's multi-objective optimization indicated the optimal modification conditions. Compared to unmodified biochar (BC-O), BC-M showed significant improvements: its specific surface area increased by 63.6%, pore volume by 139%, and functional groups by 50%–1271%. Notably, the DS of BC-M was reduced to just 1.08 mg/L, representing a 97.5% reduction from BC-O, with a minimal mass loss of only 0.78 ± 0.45% during modification. BC-M also demonstrated a remarkable enhancement in the adsorption of phenolic compounds, with a capacity 21%–2408% higher than BC-O. Furthermore, calculations indicated that BC-M could reduce carbon emissions by 0.70 t CO2/yr/t, outperforming activated carbon in this regard. This study offers valuable insights into biochar modification, providing a low-cost, high-stability, and high-efficiency alternative for environmental cleanup.

Speech Enhancement Algorithm Based on Microphone Array and Lightweight CRN for Hearing Aid

To address the performance and computational complexity issues in speech enhancement for hearing aids, a speech enhancement algorithm based on a microphone array and a lightweight two-stage convolutional recurrent network (CRN) is proposed. The algorithm consists of two main modules: a beamforming module and a post-filtering module. The beamforming module utilizes directional features and a complex time-frequency long short-term memory (CFT-LSTM) network to extract local representations and perform spatial filtering. The post-filtering module uses analogous encoding and two symmetric decoding structures, with stacked CFT-LSTM blocks in between. It further reduces residual noise and improves filtering performance by passing spatial information through an inter-channel masking module. Experimental results show that this algorithm outperforms existing methods on the generated hearing aid dataset and the CHIME-3 dataset, with fewer parameters and lower model complexity, making it suitable for hearing aid scenarios with limited computational resources.

DIGITAL TRANSFORMATION: IS THERE ANY LINK WITH MARKET PERFORMANCE OF PHARMACEUTICAL MICRO, SMALL AND MEDIUM-SIZED ENTERPRISES IN KENYA

The business-operating environment in Kenya and especially in the City County of Nairobi has caused a large number of pharmaceutical Micro, Small, and Medium-Sized Enterprises report profit warnings and thus experience performance issues. Data from the World Bank for the previous two years indicate pharmaceutical MSMEs operating in Kenya have seen a decline in profits and stagnation. The study therefore sought to establish the effect of digital transformation elements, including e-procurement, e-marketing, e-payment, and telemedicine, on the market performance of pharmaceutical Micro, Small, and Medium Enterprises (MSMEs) in Nairobi City County, Kenya. The research examines the limited understanding of how these digital technologies influence market performance, particularly in the pharmaceutical MSME sector. The study employs a cross-sectional descriptive research design, surveying 122 MSMEs from an initial target population of 175, selected through proportionate and simple random sampling. The questionnaire was tested for validity and reliability using Cronbach's alpha with a threshold of 0.7. Quantitative data were analyzed through multiple regression and correlation analysis to determine the strength of the relationships between variables. The results reveal that e-procurement, e-marketing, e-payment, and telemedicine significantly and positively affect market performance. The findings suggest that these digital strategies enhance procurement efficiency, marketing outreach, financial transactions, and healthcare service delivery, ultimately improving the market performance of pharmaceutical MSMEs.

Flight-Based Control Allocation: Towards Human–Autonomy Teaming in Air Traffic Control

It is widely recognized that airspace capacity must increase over the coming years. It is also commonly accepted that meeting this challenge while balancing concerns around safety, efficiency, and workforce issues will drive greater reliance on automation. However, if automation is not properly developed and deployed, it represents something of a double-edged sword, and has been linked to several human–machine system performance issues. In this article, we argue that human–automation function and task allocation may not be the way forward, as it invokes serialized interactions that ultimately push the human into a problematic supervisory role. In contrast, we propose a flight-based allocation strategy in which a human controller and digital colleague each have full control authority over different flights in the airspace, thereby creating a parallel system. In an exploratory human-in-the-loop simulation exercise involving six operational en route controllers, it was found that the proposed system was considered acceptable after the users gained experience with it during simulation trials. However, almost all controllers did not follow the initial flight allocations, suggesting that allocation schemes need to remain flexible and/or be based on criteria capturing interactions between flights. In addition, the limited capability of and feedback from the automation contributed to this result. To advance this concept, future work should focus on substantiating flight-centric complexity in driving flight allocation schemes, increasing automation capabilities, and facilitating common ground between humans and automation.

The Role of Grit in Developing Intrapreneurship Through the Mediation of Career Adaptability Among Generation-Z University Students

Since the latter half of the last decade, there has been a rapid transformation in workforce demographics as Generation Z enters the workplace dynamic and takes over the responsibilities of retiring Baby Boomers. However, as Generation Z employees gradually enter the workforce, organizational human resource practices have yet to evolve to address the performance issues this generation poses, resulting in higher employee turnover ratios among Generation Z employees. Despite this problem, Generation Z has a potential characteristic that will benefit organizations: intrapreneurship. This research examines the influence of grit on intrapreneurship as a valuable construct for Generation Z through the mediation of career adaptability. This research uses quantitative methods with mediation analysis. The results of this research show no direct influence of grit on intrapreneurship. When career adaptability is included as a moderator variable, the analysis results show a significance value below 0.05, which shows the role of career adaptability as a full mediator between grit and intrapreneurship. Keywords: Z Generations, intrapreneurship, grit, career adaptability

Reinforcement learning in sentiment analysis: a review and future directions

Sentiment analysis in natural language processing (NLP) is used to understand the polarity of human emotions (e.g., positive and negative) and preferences (e.g., price and quality). Reinforcement learning (RL) enables a decision maker (or agent) to observe the operating environment (or the current state) and select the optimal action to receive feedback signals (or reward) from the operating environment. Deep reinforcement learning (DRL) extends RL with deep neural networks (i.e., main and target networks) to capture the state information of inputs and address the curse of dimensionality issue of RL. In sentiment analysis, RL and DRL reduce the need for a large labeled dataset and linguistic resources, increasing scalability and preserving the context and order of logical partitions. Through enhancement, the RL and DRL algorithms identify negations, enhance the quality of the generated responses, predict the logical partitions, remove the irrelevant aspects, and ultimately capture the correct sentiment polarity. This paper presents a review of RL and DRL models and algorithms with their objectives, applications, datasets, performance, and open issues in sentiment analysis.

DREAM: Debugging and Repairing AutoML Pipelines

Deep Learning models have become an integrated component of modern software systems. In response to the challenge of model design, researchers proposed Automated Machine Learning (AutoML) systems, which automatically search for model architecture and hyperparameters for a given task. Like other software systems, existing AutoML systems have shortcomings in their design. We identify two common and severe shortcomings in AutoML, performance issue (i.e., searching for the desired model takes an unreasonably long time) and ineffective search issue (i.e., AutoML systems are not able to find an accurate enough model). After analyzing the workflow of AutoML, we observe that existing AutoML systems overlook potential opportunities in search space, search method, and search feedback, which results in performance and ineffective search issues. Based on our analysis, we design and implement DREAM, an automatic and general-purpose tool to alleviate and repair the shortcomings of AutoML pipelines and conduct effective model searches for diverse tasks. It monitors the process of AutoML to collect detailed feedback and automatically repairs shortcomings by expanding search space and leveraging a feedback-driven search strategy. Our evaluation results show that DREAM can be applied on two state-of-the-art AutoML pipelines and effectively and efficiently repair their shortcomings.

Identifying Performance Issues in Cloud Service Systems Based on Relational-Temporal Features

Cloud systems, typically comprised of various components ( e.g. , microservices), are susceptible to performance issues, which may cause service-level agreement violations and financial losses. Identifying performance issues is thus of paramount importance for cloud vendors. In current practice, crucial metrics, i.e. , key performance indicators (KPIs), are monitored periodically to provide insight into the operational status of components. Identifying performance issues is often formulated as an anomaly detection problem, which is tackled by analyzing each metric independently. However, this approach overlooks the complex dependencies existing among cloud components. Some graph neural network-based methods take both temporal and relational information into account, however, the correlation violations in the metrics that serve as indicators of underlying performance issues are difficult for them to identify. Furthermore, a large volume of components in a cloud system results in a vast array of noisy metrics. This complexity renders it impractical for engineers to fully comprehend the correlations, making it challenging to identify performance issues accurately. To address these limitations, we propose Identifying Performance Issues based on Relational-Temporal Features (ISOLATE ), a learning-based approach that leverages both the relational and temporal features of metrics to identify performance issues. In particular, it adopts a graph neural network with attention to characterizing the relations among metrics and extracts long-term and multi-scale temporal patterns using a GRU and a convolution network, respectively. The learned graph attention weights can be further used to localize the correlation-violated metrics. Moreover, to relieve the impact of noisy data, ISOLATE utilizes a positive unlabeled learning strategy that tags pseudo labels based on a small portion of confirmed negative examples. Extensive evaluation on both public and industrial datasets shows that ISOLATE outperforms all baseline models with 0.945 F1-score and 0.920 Hit rate@3. The ablation study also proves the effectiveness of the relational-temporal features and the PU-learning strategy. Furthermore, we share the success stories of leveraging ISOLATE to identify performance issues in Huawei Cloud, which demonstrates its superiority in practice.

DC-SoC: Optimizing a Blockchain Data Dissemination Model Based on Density Clustering and Social Mechanisms

Due to its partially decentralized and highly scalable features, the consortium blockchain has currently overtaken other blockchain technologies as the one most frequently used and studied across various industries. However, performance issues such as low transaction efficiency and redundant communication processes continue to hinder the development of consortium blockchains. In the Hyperledger Fabric consortium blockchain system, transaction efficiency is largely influenced by the consensus protocol and broadcast protocol. This paper introduces a novel consortium blockchain network model, DC-SoC, focused on optimizing broadcast protocols. By incorporating the concept of density clustering, a stable propagation structure is established for the blockchain network, thus optimizing data dissemination in the Gossip protocol. Additionally, the concept of social networks is integrated, using trustworthiness scores and economic incentives to evaluate node security. The experimental results demonstrate that when DC-SoC is applied in a large-scale consortium blockchain environment, it significantly improves communication performance between nodes and ensures transmission reliability.

Structured multi-view k-means clustering

K-means is a very efficient clustering method and many multi-view k-means clustering methods have been proposed for multi-view clustering during the past decade. However, since k-means have trouble uncovering clusters of varying sizes and densities, these methods suffer from the same performance issues as k-means. Improving the clustering performance of multi-view k-means has become a challenging problem. In this paper, we propose a new multi-view k-means clustering method that is able to uncover clusters in arbitrary sizes and densities. The new method simultaneously performs three tasks, i.e., sparse connection probability matrices learning, prototypes aligning, and cluster structure learning. We evaluate the proposed new method by 5 benchmark datasets and compare it with 11 multi-view clustering methods. The experimental results on both synthetic and real-world experiments show the superiority of our proposed method.

Microstructure, mechanical properties and deformation behavior of laser additively repaired 5083 and 6061 Al alloys utilizing AlMgScZr powders

Laser additive repair (LAR), as an efficient repair method, lacks specialized repair materials for Al alloys. In this work, the high-strength AlMgScZr powder was employed to address the scarcity of specialized materials and the issue of inadequate performance in LAR of 5083-H112/6061-T6 Al alloy. The microstructure, mechanical properties and deformation behavior of repaired specimens were studied. The repair zone (RZ) had high strength and high density, and the porosity was as low as 0.12 %. There was good compatibility between the repair material and the base metal (BM), and good metallurgical bonding was achieved at the fusion line. The microstructure and strengthening phase (T-phase) in the heat affected zone (HAZ) of the 5083 repaired parts exhibited negligible changes, there was no deterioration in mechanical properties. The yield strength was 162 MPa, tensile strength was 291 MPa, and elongation was 16.2 %, reaching 94 %, 104 %, and 70 % of the BM, respectively. The mechanical properties are superior in the current research on LAR of Al alloys. The LAR technique showcases its versatility in repairing aging non-strengthening Al alloys. The transition of β′′→β′ (or with B′/U1/U2)→β of the nano-reinforced phase resulted in deteriorative mechanical properties of HAZ in the 6061 repair part, consequently, the tensile strength of 6061 repair part was only 63.8 % of the strength of BM. After solution aging treatment, the β′′ phase in HAZ re-precipitated, effectively restoring the strength of 6061 repaired parts. The tensile strength of the repaired parts was increased to 95.2 % of the strength of BM. The present study elucidates the evolution of microstructure and mechanical properties during LAR process of Al alloys, offering valuable insights for future applications of this technology on Al alloys.

The Role of Excess Charge Mitigation in Electromagnetic Hygiene: An Integrative review.

The electromagnetic characteristics of many environments have changed significantly in recent decades. This is in large part due to the increased presence of equipment that emits electromagnetic radiation and materials that may often readily gain excess charge. The presence of excess charge can often increase risk of infection from pathogens, and likelihood of individuals experiencing compromised performance, respiratory problems and other adverse health issues from increased uptake of particulate matter. It is proposed that adopting improved electromagnetic hygiene measures, including optimized humidity levels, to reduce the presence of inappropriate levels of electric charge can help reduce the likelihood of ill health, infection and poor performance arising from contaminant inhalation and deposition, plus reduce the likelihood of medical devices and other electronic devices getting damaged and/or having their data compromised. It is suggested that such measures should be more widely adopted within clinical practice guidelines and water, sanitation and hygiene programs.

Dual-Functional SnO2-CNT-PANI||PP||SnO2 Separator for Enhanced Lithium-Sulfur Battery Stability and Capacity

Lithium-sulfur batteries (LSBs) provide high capacity and energy density for advanced energy storage but encounter performance and safety issues due to lithium polysulfides (LiPSs) dissolution and dendrite formation. This study introduces a dual-functional SnO2-CNT-PANI||PP||SnO2 separator (DF-SNCPS) aimed at improving the stability and safety of LSBs. The separator includes a SnO2-CNT-PANI layer on the sulfur cathode side for enhanced electrical conductivity and polysulfide adsorption and a SnO2 nanoparticle (nano-SnO2) layer on the lithium anode side to regulate lithium-ion concentration and inhibit dendrite growth. Density functional theory (DFT) and the phase field method have verified the relevant material's catalytic performance and dendrite suppression ability. These methods have provided a comprehensive understanding of the underlying mechanisms, providing auxiliary explanations for their role in catalyzing polysulfide conversion and suppressing lithium dendrite growth. This innovative design mitigates the shuttle effect and enhances cyclic stability and sulfur utilization, achieving a notable cyclic performance with a capacity retention of 0.05% per cycle over 800 cycles at 1C. The synergistic effects of the modified separator present a viable pathway for LSB commercialization.

Multiple lubricated joints with long and short bearings in multibody mechanical systems - Modeling, simulation, and performance analysis

This study examines the effect of multiple lubricated imperfect long and short bearings on the performance of a multibody mechanical system. Unlike the typical assumption of a single perfect or imperfect lubricated joint due to modeling difficulties, this research considers the practical impacts of clearances in multiple joints. While unlubricated joints generally cause significant performance issues, lubricants reduce these effects, creating more localized peaks. The findings show that as the number of lubricated joints increases, both the magnitude and frequency of these peaks rise. In systems with multiple journal bearings, torque peaks become more noticeable due to the additional degrees of freedom introduced by the clearances. These degrees of freedom amplify acceleration, leading to higher lubricant reaction forces, which in turn require greater motor torque peaks to maintain the system's kinematics. Shorter lubricated joints exhibit more severe peaks than longer ones, mainly due to side leakage causing axial pressure variation and reduced damping capacity. The study highlights the need to replace idealized joints with imperfect ones for more accurate modeling of practical systems.

How to manage performance issues in nursing practice.

Addressing performance issues in nursing practice is a crucial but sensitive aspect of the role of nurse leaders and managers. This article explains how they can undertake this process by using a supportive and proactive approach that can foster a positive work environment, improve patient care and empower nurses to thrive in practice. Those involved in addressing a nurse's performance issues must have the knowledge and skills required to do so effectively with sensitivity and care, while working within their level of competence. • Where performance issues have been identified, early intervention is essential to foster constructive dialogue, preserve relationships and maintain a motivated and productive healthcare team. • The focus should be on understanding what aspects of the nurse's performance require improvement, what has been learned from the situation and how this can be applied to the nurse's future practice. • It is useful to work with the nurse to develop a collaborative action plan for improvement that establishes clear goals, timelines and resource allocation. REFLECTIVE ACTIVITY: 'How to' articles can help to update your practice and ensure it remains evidence-based. Apply this article to your practice. Reflect on and write a short account of: • A time you addressed performance issues with a colleague or team member, or how someone helped you to improve your practice. • How you could use this information to develop your knowledge and skills in addressing performance issues.

The effect of data complexity on classifier performance

The research area of Software Defect Prediction (SDP) is both extensive and popular, and is often treated as a classification problem. Improvements in classification, pre-processing and tuning techniques, (together with many factors which can influence model performance) have encouraged this trend. However, no matter the effort in these areas, it seems that there is a ceiling in the performance of the classification models used in SDP. In this paper, the issue of classifier performance is analysed from the perspective of data complexity. Specifically, data complexity metrics are calculated using the Unified Bug Dataset, a collection of well-known SDP datasets, and then checked for correlation with the defect prediction performance of machine learning classifiers (in particular, the classifiers C5.0, Naive Bayes, Artificial Neural Networks, Random Forests, and Support Vector Machines). In this work, different domains of competence and incompetence are identified for the classifiers. Similarities and differences between the classifiers and the performance metrics are found and the Unified Bug Dataset is analysed from the perspective of data complexity. We found that certain classifiers work best in certain situations and that all data complexity metrics can be problematic, although certain classifiers did excel in some situations.