Parameter Tuning Research Articles

Unmasking The Enigma:Exploring Personality Traits Using Social Media Text

In this paper personality is an important parameter as it differentiates various individuals from another. Predicting personality has many applications in real world. The main objective of this project is to take textual data as input from the user and then run the trained machine learning model on this data to predict his four personality traits which are Introversion vs Extroversion, Sensing vs Intuition, Thinking vs Feeling, Judging vs Perceiving.The main objective is to build an application where users can answer the questions which are processed and analyzed to output his personality traits. The output is a string of four characters where each character determines a personality trait, total of sixteen personality types are possible.The Machine learning model Random Forest Classifier is used to classify the text and output four personality traits. Processing of large textual data is to be done using Natural Language Processing (NLP) techniques with the help of NLTK libraries to process and categorize the data. In order to increase the performance of the model hyper parameter tuning along with cross fold validations is done. Keywords: Random Forest, Natural Language Processing, Text

Emle-engine: A Flexible Electrostatic Machine Learning Embedding Package for Multiscale Molecular Dynamics Simulations.

We present in this work the emle-engine package (https://github.com/chemle/emle-engine)─the implementation of a new machine learning embedding scheme for hybrid machine learning potential/molecular-mechanics (ML/MM) dynamics simulations. The package is based on an embedding scheme that uses a physics-based model of the electronic density and induction with a handful of tunable parameters derived from in vacuo properties of the subsystem to be embedded. This scheme is completely independent of the in vacuo potential and requires only the positions of the atoms of the machine learning subsystem and the positions and partial charges of the molecular mechanics environment. These characteristics allow emle-engine to be employed in existing QM/MM software. We demonstrate that the implemented electrostatic machine learning embedding scheme (named EMLE) is stable in enhanced sampling molecular dynamics simulations. Through the calculation of free energy surfaces of alanine dipeptide in water with two different ML options for the in vacuo potential and three embedding models, we test the performance of EMLE. When compared to the reference DFT/MM surface, the EMLE embedding is clearly superior to the MM one based on fixed partial charges. The configurational dependence of the electronic density and the inclusion of the induction energy introduced by the EMLE model leads to a systematic reduction in the average error of the free energy surface when compared to MM embedding. By enabling the usage of EMLE embedding in practical ML/MM simulations, emle-engine will make it possible to accurately model systems and processes that feature significant variations in the charge distribution of the ML subsystem and/or the interacting environment.

Understanding the Phase of Responsivity and Noise Sources in Frequency-Domain Multiplexed Readout of Transition Edge Sensor Bolometers

AbstractCosmic microwave background (CMB) experiments have deployed focal planes with $$\mathcal {O}(10^{4})$$ O ( 10 4 ) transition edge sensor (TES) bolometers cooled to sub-Kelvin temperatures by multiplexing the readout of many TES channels onto a single pair of wires. Digital Frequency-domain Multiplexing (DfMux) is a multiplexing technique used in many CMB polarization experiments, such as the Simons Array, SPT-3 G, and EBEX. The DfMux system studied here uses LC filters with resonant frequencies ranging from 1.5 to 4.5 MHz connected to an array of TESs. Each detector has an amplitude-modulated carrier tone at the resonant frequency of its accompanying LC resonator. The signal is recovered via quadrature demodulation where the in-phase (I) component of the demodulated current is in phase with the complex admittance of the circuit and the quadrature (Q) component is orthogonal to I. Observed excess current noise in the Q component is consistent with fluctuations in the resonant frequency. This noise has been shown to be non-orthogonal to the phase of the detector’s responsivity. We present a detailed analysis of the phase of responsivity of the TES and noise sources in our DfMux readout system. Further, we investigate how modifications to the TES operating resistance and bias frequency can affect the phase of noise relative to the phase of the detector responsivity, using data from Simons Array to evaluate our predictions. We find that both the phase of responsivity and phase of noise are functions of the two tuning parameters, which can be purposefully selected to maximize signal-to-noise (SNR) ratio.

TEXT CLASSIFICATION USING ADAPTIVE BOOSTING ALGORITHM WITH OPTIMIZATION OF PARAMETERS TUNING ON CABLE NEWS NETWORK (CNN) ARTICLES

The development of the era encourages advances in communication and information technology. This resulted in the exchange of information being faster because it is connected to the internet. One platform that provides online news articles is Cabel News Network (CNN), which has been broadcasting news on its website since 1995. The number of Cabel News Network news articles continues to increase, so news articles are categorized to make it easier for readers to find articles according to the category they want. Classification is a technique for determining the class of an object based on its characteristics, where the class label is known beforehand. One of the algorithms for classification is adaptive boosting (AdaBoost). The AdaBoost algorithm performs classification by building several weighted decision trees (stumps), then the class determination is based on the number of stumps with the highest weight. The AdaBoost algorithm can be combined with parameter tuning to avoid overfitting or underfitting resulting from a weak set of stumps. Therefore, this study implements the AdaBoost algorithm with parameter tuning on CNN news article classification. The data used in this study is CNN news article data from 2011 to 2022 sourced from the Kaggle page. The data is categorized into six classes, namely business, entertainment, health, news, politics, and sports. This study uses two evaluation metrics, namely the accuracy value and the confusion matrix to measure the performance of the AdaBoost algorithm. The accuracy value obtained is 0,78763, the precision value is 0.91, the recall value is 0.85, and the F1 score value is 0.88.

Robust control strategies applicable to DC–DC converter with reliability assessment: A review

AbstractThis paper provides an overview of various control strategies for DC–DC converters along with a brief discussion on various performance indices for evaluating the reliability of designed controller. DC–DC converters are emerging as the fastest‐growing interfacing devices in the world because of their emergent applications in almost all domains of engineering. Notably, the non‐linear behavior of DC–DC converters offers a perplexing problem in designing a robust controller. A robust controller must ensure proper closed‐loop stability with desired performance and reliable control for output voltage regulation in the event of various possible perturbations. This creates the requirement for practically implementable non‐linear controllers that can effectively overlook the drawbacks of linear controllers. The prominence of this composition is based on the expansion in tuning techniques of proportional‐integral‐derivative controller gain parameters using linear strategies and heading towards non‐linear strategies by reviewing 196 research papers. The modification in non‐linear control strategies in terms of their fundamental features associated with key benefits and limitations are comprehensively reviewed. This study mostly revolves around non‐linear internal model control (IMC) strategies for parameter tuning of IMC‐based controllers for various order of plants with an improved degree of freedom.

Improving replicability in single-cell RNA-Seq cell type discovery with Dune

BackgroundSingle-cell transcriptome sequencing (scRNA-Seq) has allowed new types of investigations at unprecedented levels of resolution. Among the primary goals of scRNA-Seq is the classification of cells into distinct types. Many approaches build on existing clustering literature to develop tools specific to single-cell. However, almost all of these methods rely on heuristics or user-supplied parameters to control the number of clusters. This affects both the resolution of the clusters within the original dataset as well as their replicability across datasets. While many recommendations exist, in general, there is little assurance that any given set of parameters will represent an optimal choice in the trade-off between cluster resolution and replicability. For instance, another set of parameters may result in more clusters that are also more replicable.ResultsHere, we propose Dune, a new method for optimizing the trade-off between the resolution of the clusters and their replicability. Our method takes as input a set of clustering results—or partitions—on a single dataset and iteratively merges clusters within each partitions in order to maximize their concordance between partitions. As demonstrated on multiple datasets from different platforms, Dune outperforms existing techniques, that rely on hierarchical merging for reducing the number of clusters, in terms of replicability of the resultant merged clusters as well as concordance with ground truth. Dune is available as an R package on Bioconductor: https://www.bioconductor.org/packages/release/bioc/html/Dune.html.ConclusionsCluster refinement by Dune helps improve the robustness of any clustering analysis and reduces the reliance on tuning parameters. This method provides an objective approach for borrowing information across multiple clusterings to generate replicable clusters most likely to represent common biological features across multiple datasets.

Adaptive Network-Based Fuzzy Inference System–Proportional–Integral–Derivative Controller Based on FPGA and Its Application in Radiofrequency Ablation Temperature Control

The radiofrequency ablation temperature system is characterised by its time-varying, non-linear, and hysteretic nature. The application of PID controllers to the control of radiofrequency ablation temperature systems has a number of challenges, including overshoot, dependence on high-precision mathematical models, and difficulty in parameter tuning. Therefore, in order to improve the effectiveness of radiofrequency ablation temperature control, an adaptive network-based fuzzy inference system combined with an incremental PID controller was used to optimise the shortcomings of the PID controller in radiofrequency ablation temperature control. At the same time, the learning rate at the time of updating the consequence parameters was set by segmentation to solve the problem of poor control accuracy when the ANFIS-PID controller is implemented based on FPGA fixed-point decimals. Based on FPGA-in-the-loop simulation experiments and ex vivo experiments, the effectiveness of the ANFIS-PID controller in the temperature control of radiofrequency ablation was verified and compared with the PID controller under the same conditions. The experimental results show that the ANFIS-PID controller has a superior performance in terms of tracking capability and stability compared with the PID controller.

A model for the noninvasive, habitat-inclusive estimation of upper limit abundance for synanthropes, exemplified by M. fascicularis.

Accurately estimating population sizes for free-ranging animals through noninvasive methods, such as camera trap images, remains particularly limited by small datasets. To overcome this, we developed a flexible model for estimating upper limit populations and exemplified it by studying a group-living synanthrope, the long-tailed macaque (Macaca fascicularis). Habitat preference maps, based on environmental and GPS data, were generated with a maximum entropy model and combined with data obtained from camera traps, line transect distance sampling, and direct sightings to produce an expected number of individuals. The mapping between habitat preference and number of individuals was optimized through a tunable parameter ρ (inquisitiveness) that accounts for repeated observations of individuals. Benchmarking against published data highlights the high accuracy of the model. Overall, this approach combines citizen science with scientific observations and reveals the long-tailed macaque populations to be (up to 80%) smaller than expected. The model's flexibility makes it suitable for many species, providing a scalable, noninvasive tool for wildlife conservation.

A mechanical manipulated electromechanical coupling design with stretchable electret film: mechanical sensing, energy harvesting, and actuation

Electrical and mechanical energy converts around the nature, and electromechanical coupling effect is applied in various conditions such as mechanical sensing, electrical actuation, and self-powering. During the energy type conversion, electromechanical parameters are among the key issues, such as enlarging the sensitivity and range of mechanical sensing, and energy harvesting efficiency. In this work, a mechanical manipulated approach with stretchable electret is proposed to continuously manipulate the electromechanical parameters. An electromechanical coupling demonstration with pre-stretched electret films and non-contact electrodes are applied, verifying high and regulable electromechanical coupling parameters, and it is advantaged from large deformable and overload permissible capability. This mechanical manipulation approach proposes a new possibility on simplifying the structural and mechanical design of various electromechanical devices, and further enhancing the general applicability with certain geometry and material with ultra-high and tunable electromechanical coupling parameters.

Multi-view Stable Feature Selection with Adaptive Optimization of View Weights

The feature selection problem in multi-view data has garnered widespread attention and research in recent years, leading to the development of numerous feature selection algorithms tailored for multi-view data. However, existing methods often focus solely on known data, overlooking the potential distribution information of unknown data. Additionally, these methods inevitably introduce a large number of parameters to fully utilize the information from different views in multi-view data, thereby reducing the efficiency of model training. To address these issues comprehensively, we propose a novel framework called Multi-view Stable Feature Selection with Adaptive Optimization of View Weights (MvSFS-AOW). Specifically, the framework first employs the Multi-view Stable Feature Selection (MvSFS) algorithm to evaluate and select features from different views. Subsequently, it dynamically adjusts view weights using the Adaptive Optimization of View Weights (AOW) algorithm to achieve optimal generalization performance. By incorporating unknown data into the training process, we enhance the reliability of the framework in practical applications. Furthermore, our framework achieves competitive performance without requiring extensive parameter tuning. Experimental results demonstrate that the proposed framework achieves promising classification and clustering performance on multiple datasets, surpassing other state-of-the-art algorithms. Code for this paper available on: https://github.com/boredcui/MvSFS-AOW.

Systematic Control Design for Adaptive Rejection of Narrow-Band Disturbances Based on a Novel Resonant Generalized Extended State Observer

In this paper, the adaptive rejection of multiple narrow-band disturbances with time-varying frequencies is investigated. A systematic control approach based on a resonant generalized extended state observer (RGESO) is proposed. The frequency response of the RGESO is shaped into a desired form to reject the total disturbance, including the wide-band uncertainties and the narrow-band disturbances. The state-space realization of the RGESO is also provided to estimate both the states and the total disturbance. To handle the time-varying frequencies of the narrow-band disturbances, a direct frequency estimation approach based on the Levenberg–Marquardt method and an online supervision strategy is presented to improve convergence and reliability, and the estimations are then used to adapt the RGESO parameters. Parameter tuning and stability robustness analysis methods are also developed for the proposed scheme to facilitate its application. Finally, extensive simulations are conducted to demonstrate the effectiveness of our method.

Refining the reconstruction-based Monte Carlo methods for solving breakage population balance equation

This work presents a class of refinement reconstruction-based Monte Carlo methods dedicated to solving a generic breakage population balance equation. The focus is on enhancing the precision and computational speed of the existing reconstruction scheme by incorporating two refinement strategies, including effective-breakage ratios and population scaling rules with local coarsening/refinement based on adaptive subdomain meshes. A comparative analysis of the performances of the proposed methods is conducted for the benchmark breakage case with a known analytical solution. The results indicate that the accuracy can be significantly increased by a stepwise-varying breakage ratio and the local refinement scaling rule, whereas the cost can be reduced by properly tuning parameters related to the refinement strategies. In addition, multi-objective optimization is introduced to achieve optimal simulation with the least systematic and statistical errors and CPU time, offering further insights into the combination of conditions required for the Monte Carlo methods to yield the best possible efficiency.

Prediction of bubble departing diameter in pool boiling of mixtures by ANN using modified ReLU

In this research, bubble departure diameter in pool boiling have been measured in aqueous amine and ethylene glycol solutions for various concentrations. The experimental data have been compared with major existing predictive correlations. It is shown that the effect and identity of the independent variables on bubble diameter proposed in the previous studies are inconsistent. The predictions of different correlations have on average a deviation of about 40% from the experimental data. This is mainly due to the complicated interactions between bubbles on the heterogeneous boiling medium, which provides a complex condition. This complexity limits any mathematical modelling of the forces acting on the developing bubbles. Particularly in liquid solutions, where mass transfer by back diffusion through micro-sub-layers adds further complexity. In this work, the classical artificial neural network, ANN, with rectified linear unit, ReLU, activating function, AF, has been modified. This modification is based on adding a numerical matrix to each layer to adjust the slope of AF for each neuron independently. The addition of this parameter, together with the adjustment of the bias matrix, makes the activation function more flexible than the classical ReLU. To find the tuning parameters, a genetic algorithm was implemented instead of the back-propagation technique. It is shown that the predictions of the trained ANN with modified ReLU AF agree within an absolute average error of 10%, which is equal to the total uncertainty of the measurements. Prediction of bubble departing diameter in boiling phenomena is a key parameter for accurate design, operation and optimisation in many industrial systems.

Parameter optimization of type II fuzzy sliding mode control for bridge crane systems based on improved grey wolf algorithm

AbstractBridge cranes are complex nonlinear dynamic systems with underactuated characteristics, making it challenging for controllers to man age the spatial swing of the load effectively. Additionally, uncertainties both within and outside the system adversely impact control performance. To address these issues, a Type‐II fuzzy sliding mode controller has proven effective in enhancing the anti‐swing control performance of the payload. However, due to the intricate parameter adjustment optimization problem and potential challenges in dealing with nonlinearity and uncertainty, especially in complex dynamic systems, this paper proposes a grey wolf algorithm based on a dynamic spiral hunting mechanism. This enhancement endows the algorithm with improved convergence speed and higher robustness, enabling more effective parameter tuning for the second‐order fractional‐order sliding mode controller (FSMC). The proposed algorithm demonstrates superior convergence speed and solution accuracy performance through testing and comparison. Finally, simulation verification under two conditions of the bridge crane system validates the effectiveness of the proposed approach.

Utilizing machine learning to predict employee turnover in high-stress sectors

This study investigates the application of machine learning techniques to predict employee turnover in high-stress sectors. The primary objective is to enhance retention strategies by accurately identifying potential turnover risks. The research utilizes a comprehensive dataset comprising various factors, including employee demographics, job satisfaction, performance metrics, and stress levels. Multiple machine learning algorithms, such as logistic regression, decision trees, random forests, and neural networks, are employed to build predictive models. The methodology involves data preprocessing, feature selection, model training, and evaluation. Cross-validation and hyper parameter tuning are performed to ensure the robustness and accuracy of the models. The performance of each algorithm is assessed using metrics such as accuracy, precision, recall, and the area under the receiver operating characteristic curve (AUC-ROC). Key findings reveal that machine learning models can effectively predict employee turnover, with random forests and neural networks demonstrating superior performance. Significant predictors of turnover include job satisfaction, stress levels, and performance ratings. The study concludes that integrating machine learning models into human resource practices can provide valuable insights for preemptive interventions, ultimately reducing turnover rates in high-stress environments. Future research should explore the integration of real-time data and the potential of deep learning techniques to further enhance predictive accuracy. Additionally, the ethical implications of using predictive models in HR decisions warrant careful consideration to ensure fairness and transparency. Keywords: Machine Learning (ML), Employee Turnover, Predictive Analytics, Human Resources (HR), High-Stress Sectors, Decision Trees, Random Forests, Extreme Gradient Boosting (XGBoost), Personalized Retention Strategies, Business Intelligence (BI) Tools, Data Quality, Ethical Considerations, Data Privacy, Natural Language Processing (NLP), Deep Learning, Real-time Data Analysis, Employee Engagement, Work-Life Balance, Organizational Performance, Data-Driven Insights.

New PID parameter tuning based on improved dung beetle optimization algorithm

New <scp>PID</scp> parameter tuning based on improved dung beetle optimization algorithm

Learning from the present for the future: The Jülich LOFAR Long-term Archive

The Forschungszentrum Jülich has been hosting the German part of the LOFAR archive since 2013. It is Germany’s most extensive radio astronomy archive, currently storing nearly 22 petabytes (PB) of data. Future radio telescopes are expected to require a dramatic increase in long-term data storage. Here, we take stock of the current data management of the Jülich LOFAR Data Archive, describe the ingestion, the storage system, the export to the long-term archive, and the request chain. We analysed the data availability over the last 10 years and searched for the underlying data access pattern and the energy consumption of the process. We determine hardware-related limiting factors, such as network bandwidth and cache pool availability and performance, and software aspects, e.g. workflow adjustment and parameter tuning, as the main data storage bottlenecks. By contrast, the challenge in providing the data from the archive for the users lies in retrieving the data from the tape archive and staging them. Building on this analysis, we suggest how to avoid/mitigate these problems in the future and define the requirements for future even more extensive long-term data archives.

An adaptive PID controller for path following of autonomous underwater vehicle based on Soft Actor–Critic

In recent years, autonomous underwater vehicles (AUVs) have witnessed rapid development, and its motion control has garnered increasing attention. Meanwhile, in industries, PID controllers are still wildly employed by most AUVs due to their simplicity, ease of deployment, and a certain level of robustness. However, they are facing significant challenges in parameter tuning, especially when dealing with various control missions and changing external environments. Deep reinforcement learning, as a data-driven approach, has gradually made its impact in AUV control. However, its lack of interpretability has hindered its deployment in relevant experiments. To address these issues, this paper proposed an adaptive PID controller for path following of AUVs based on the Soft Actor–Critic (SAC). This controller combines the interpretability of PID with the intelligence of reinforcement learning. A simulation platform was established and compared with other typical control methods, demonstrating the superiority of the proposed controller. Finally, the feasibility of the proposed SAC-PID controller was validated by lake trials. The results showed that the SAC-PID controller significantly outperformed the PID and Proximal Policy Optimization (PPO) PID controllers in multiple dimensions, such as control precision and convergence speed.

Augmenting the Stability of Automatic Voltage Regulators through Sophisticated Fractional-Order Controllers

The transition from traditional to renewable energy sources is a critical issue in current energy-generation systems, which aims to address climate change and the increased demand for energy. This shift, however, imposes additional burdens on control systems to maintain power system stability and quality within predefined limits. Addressing these challenges, this paper proposes an innovative Modified Hybrid Fractional-Order (MHFO) automatic voltage regulator (AVR) equipped with a fractional-order tilt integral and proportional derivative with a filter plus a second-order derivative with a filter FOTI-PDND2N2 controller. This advanced controller combines the benefits of a (FOTI) controller, known for enhancing dynamic performance and steady-state response, with a (PDND2N2) controller to improve system robustness and adaptability. The proposed MHFO controller stands out with its nine tunable parameters, providing more extensive control options than the conventional three-parameter PID controller and the five-parameter FOPID controller. Furthermore, a recent optimization approach using a growth optimizer (GO) has been formulated and applied to optimally adjust the MHFO controller’s parameters simultaneously. The performance of the proposed AVR based on the MHFO-GO controller is scrutinized by contrasting it with various established and developed optimization algorithms. The comparative study shows that the AVR based on the MHFO-GO controller surpasses other AVR controllers from the stability, robustness, and dynamic response speed points of view.

Design and Fabrication of Fibrous Spindle‐Like Constructs Using a Melt Electrohydrodynamic Writing Process

AbstractAdvanced manufacturing of 3D‐structured materials enables the production of biomimetic muscle tissues. While models of muscle tissue exist, current approaches possess a limited ability to capture essential elements of the muscle tissue microarchitecture. Therefore, this paper aims to engineer the intrinsically complex muscle spindle‐like ellipsoid geometry using a polymer melt‐based electrohydrodynamic (EHD) printing system. EHD systems have conventionally reported fiber deposition in a layerwise fashion. However, without mitigation, the observed fiber sagging and residual charge phenomena for the melt electrowriting (MEW) process limit the ability to produce layered fibrous 3D constructs with in‐plane fiber alignment. However, in this work, fiber sagging and residual charge phenomena are leveraged as part of the design intent to deposit nonoverlapping suspended fibers between two stationary walls toward spindle‐like construct fabrication. Specifically, herein the structural and mechanical properties of the MEW‐enabled spindle‐like constructs are analyzed as a function of the process and design parameters that govern control over fiber sagging and residual charge. The results indicate that the collector speed and wall‐to‐wall distance are the key parameters for tuning the spindle morphology. Moreover, cycle number and fiber diameter are identified as effective parameters for tuning the spindle mechanical properties.