Adaptive Technique Research Articles

Synthesis, Characterization, and Optimization of Novel Furan-ring Fused Chalcones via Radical Cyclization of α,β-Unsaturated Ketones and Cyclic Ketone

Abstract: Novel Furan-ring Fused Chalcones (FFC) were synthesized using a radical cyclization reaction of α,β-unsaturated ketones with cyclic ketone as the model reaction to attain this goal. In this study, traditional and microwave-assisted methods for the efficient and cost-effective synthesis of furan-ring fused chalcones in mild reaction conditions are compared and optimized. The goal is to develop a reliable and adaptable synthetic technique that may be used to produce these useful chalcone derivatives quickly and effectively. The optimal experimental conditions for these reactions were carefully determined using two independent methodologies: conventional (Method A) and microwave (Method B). The results indicated that the proposed method B could be used effectively in the future to synthesize novel furans with short reaction times and acceptable yields (87-94 %), and products were purified by column chromatography and preparative thin layer chromatography (PTLC). All new compounds were characterized by 1H-NMR, 13C-NMR, LC-MS, and elemental analyses.

Enhancing Contextual Masking in Reversible Linguistic Steganography with Ensemble Methods

Various cybercrimes can be prevented by text authentication that is responsible for preserving digital identities and contents. Digital signatures come in handy as a way of authenticating texts, which is an extensively used method. One approach to this problem is linguistic steganography, which allows hiding the signature in other words within the text and thereby facilitating efficient data management. However, it should be noted that there is a danger that these kinds of changes may result in inappropriate decisions being taken by automated computing systems not to mention change their final outputs (unseen). As such, many people are becoming more concerned with the possibility of reversing steganography so that it becomes possible to eliminate any distortions made during the process. This paper uses Contextual masking instead of masking randomly with BERT model. The goal behind this research was developing a natural language text specific Reversible Steganographic System. Our model uses pre-trained BERT as a transformer based masked language model and reversibly embeds messages through predictive word substitution. To quantify predictive uncertainty, we introduce an adaptive steganographic technique using Bayesian deep learning. This experiment shows us how our proposed system balances imperceptibility with capacity while maintaining near semantics at all times. Also, we integrate ensemble methods instead of Monte Carlo to balance the imperceptibility.

Efficient Simulation of the Laser-Based Powder Bed Fusion Process Demonstrated on Open Lattice Materials Fabrication

Strut-based or open lattice materials are a category of advanced materials used in medical and aerospace applications due to their properties, such as high strength-to-weight ratio and energy absorption capability. The most prominent method for the fabrication of lattice materials is the Laser-based Powder Bed Fusion (L-PBF) additive manufacturing (AM) process, due to its ability to produce parts of complex geometries. The current work presents an efficient meso-scale finite element (FE) modeling methodology of the L-PBF process demonstrated in the fabrication of body-centered cubic (BCC) lattice materials. The modeling efficiency is gained through an adaptive mesh refinement technique, which results in accurate and efficient prediction of the temperature field during the process evolution. To examine the efficiency of the modeling method, the computational time is compared with that of a conventional FE simulation, based on the element and birth technique. The temperature history difference between the two approaches is minor but the adaptive mesh modeling requires only a small portion of the simulation time of the conventional model. In addition, the computational results present a good correlation with the available experimental measurements for various process parameters validating the presented efficient method.

Gradient-based adaptive neural network technique for two-dimensional local fractional elliptic PDEs

Abstract This paper introduces gradient-based adaptive neural networks to solve local fractional elliptic partial differential equations. The impact of physics-informed neural networks helps to approximate elliptic partial differential equations governed by the physical process. The proposed technique employs learning the behaviour of complex systems based on input-output data, and automatic differentiation ensures accurate computation of gradient. The method computes the singularity-embedded local fractional partial derivative model on a Hausdorff metric, which otherwise halts the computation by available approximating numerical methods. This is possible because the new network is capable of updating the weight associated with loss terms depending on the solution domain and requirement of solution behaviour. The semi-positive definite character of the neural tangent kernel achieves the convergence of gradient-based adaptive neural networks. The importance of hyperparameters, namely the number of neurons and the learning rate, is shown by considering a stationary anomalous diffusion-convection model on a rectangular domain. The proposed method showcases the network’s ability to approximate solutions of various local fractional elliptic partial differential equations with varying fractal parameters.

Domain-Adaptive and Context-Aware Fall Detection Based on Coarse-Fine Network Learning

Accurate fall detection among older adults is crucial for minimizing injuries and fatalities. However, existing fall detection systems face challenges due to the rarity and variability of falls, compounded by limitations in real-world datasets. To address this, a novel fall detection approach integrating domain adaptation and context-awareness within a Coarse-Fine Network Learning framework is proposed. The model combines high-level semantic understanding with low-level spatial details to achieve robust fall detection across diverse environments. Domain adaptation techniques like transfer learning and domain-specific fine-tuning are introduced to enhance model generalization and adaptability. Additionally, context-aware features, including environmental cues and behavioral patterns, reduce false alarms. Extensive experimentation on real- world datasets demonstrates the superior performance of the model, outperforming traditional approaches. The framework holds promise for deployment in healthcare settings, contributing to improved safety for older adults worldwide. The interpretability of the model's predictions enhances its usability in practical applications.

Object Detection in Diverse Environment

This project focuses on the creation of a custom dataset for object detection, employing meticulous annotation techniques and diverse data sources. The dataset is evaluated using various object detection models, including YOLOv8, on both the custom dataset and established benchmarks. Through comparative analysis, the quality, diversity, and generalizability of the custom dataset are assessed. Additionally, the impact of dataset quality on model performance is examined, highlighting areas for improvement. Furthermore, the integration of Generative Adversarial Networks (GANs) to augment dataset diversity is explored. Future work involves refining dataset collection methodologies, exploring novel architectures, and integrating domain adaptation techniques to enhance model robustness and applicability in real-world scenarios.

Study on Water Entry of a 3D Torpedo Based on the Improved Smoothed Particle Hydrodynamics Method

The water entry of a torpedo is a complex nonlinear problem, involving transient impact, free surface deformation, droplet splashing, and fluid–structure coupling, which poses severe challenges to traditional mesh methods. The meshless smoothed particle hydrodynamics (SPH) method shows unique advantages in capturing the complex features of the water entry of the torpedo at different entry angles. However, it still suffers from some inherent shortcomings, such as low surface discretization accuracy, poor discretization flexibility, and low calculation efficiency. In this study, an improved adaptive SPH algorithm is proposed to investigate the water entry of the torpedo accurately and efficiently. This method integrates meshless point generation and adaptive techniques simultaneously. The numerical results demonstrate that when the torpedo vertically enters the water at different velocities, the induced impact loads acting on the head of the torpedo fluctuate significantly with two peak values in the initial stage and thereafter stabilize in a later stage. The impact load acting on the torpedo, the entry depth of the torpedo, the splash height of the droplets, and the size of the cavity generated around the torpedo increase with the increment in the entry velocity. When the torpedo enters the water at different entry angles under the same initial entry velocity, both the vertical and the horizontal movements of the torpedo are observed, which results in more complex variations in parameters along the x- and y-axes. The findings and the corresponding numerical method in this study can provide a certain basis for the future designs of the entry trajectory and the structural bearing capacity of torpedoes.

Strengthening adaptation in coastal Bangladesh: community-based approaches for sustainable agriculture and water management

The coastal region of Bangladesh is significantly influenced by soil and water salinity, which is further exacerbated by the increasing frequency of tropical cyclones and rising sea levels. Understanding the extent of salinity and its challenges is crucial for promoting sustainable agriculture and ensuring access to safe drinking water. Using quantitative (soil and water parameters) and qualitative (focus group discussion and key informant interview) data, we investigated (i) soil and water salinity and soil nutrient contents; and (ii) adaptive practices in agriculture and drinking water management in three sub-districts (Assasuni, Dacope and Morrelganj) in the southwestern coastal region of Bangladesh. Results show that soil salinity levels did not significantly differ among the sub-districts, with Assasuni having slightly higher soil salinity (8.24 dS m-1) compared to Dacope (8.08 dS m-1) and Morrelganj (7.96 dS m-1). Significant differences were observed in the salinity level of pond and canal water among the sub-districts, with Assasuni having the highest levels of salinity in both pond (13.98 dS m-1) and canal water (77.85 dS m-1), compared to other sub-districts. Soil and water salinity were the major challenges reported by the respondents; however, climate-induced stresses (e.g., untimely precipitation) and outbreaks of pests during droughts have been identified as prominent issues in sustainable agriculture. Rainwater harvesting has been identified as a viable adaptive technique in drinking water management, offering a feasible solution to address water and soil salinity. The study underscores the importance of implementing adaptive practices (e.g., rainwater harvesting) to address water scarcity and salinity issues in the coastal region and promote resilient agricultural systems.

Underwater image enhancement via color correction and multi-feature image fusion

Abstract The light attenuation underwater causes the actual underwater images to suffer from color cast, low contrast, and weak illumination. To address these issues, an effective fusion-based method is proposed, which realizes color correction, brightness adjustment, contrast, and detail enhancement of underwater images. Concretely, we first design an adaptive color correction method via dominant color channel judgment and lower color channel compensation. Then, we detect the brightness of each input image and propose a gamma correction function based on the gradient of the cumulative histogram to adjust the brightness of the low-light images. Subsequently, global histogram stretching and adaptive fractional differentiation techniques are employed to process the brightness-adjusted image, and then the global contrast-enhanced version and detail-enhanced version are generated respectively. To integrate the advantages of both versions, a channel fusion method based on the Lab color space is used to fuse the luminance and color of the two versions separately. The experimental results demonstrate the effectiveness of the proposed method in improving the color and illumination of underwater images, as well as enhancing the clarity of images. Moreover, the testing results on multiple datasets validate the excellent stability of this method.

GazeSwitch: Automatic Eye-Head Mode Switching for Optimised Hands-Free Pointing

This paper contributes GazeSwitch, an ML-based technique that optimises the real-time switching between eye and head modes for fast and precise hands-free pointing. GazeSwitch reduces false positives from natural head movements and efficiently detects head gestures for input, resulting in an effective hands-free and adaptive technique for interaction. We conducted two user studies to evaluate its performance and user experience. Comparative analyses with baseline switching techniques, Eye+Head Pinpointing (manual) and BimodalGaze (threshold-based) revealed several trade-offs. We found that GazeSwitch provides a natural and effortless experience but trades off control and stability compared to manual mode switching, and requires less head movement compared to BimodalGaze. This work demonstrates the effectiveness of machine learning approach to learn and adapt to patterns in head movement, allowing us to better leverage the synergistic relation between eye and head input modalities for interaction in mixed and extended reality.

Multi-step Transfer Learning in Natural Language Processing for the Health Domain

The restricted access to data in healthcare facilities due to patient privacy and confidentiality policies has led to the application of general natural language processing (NLP) techniques advancing relatively slowly in the health domain. Additionally, because clinical data is unique to various institutions and laboratories, there are not enough standards and conventions for data annotation. In places without robust death registration systems, the cause of death (COD) is determined through a verbal autopsy (VA) report. A non-clinician field agent completes a VA report using a set of standardized questions as guide to identify the symptoms of a COD. The narrative text of the VA report is used as a case study to examine the difficulties of applying NLP techniques to the healthcare domain. This paper presents a framework that leverages knowledge across multiple domains via two domain adaptation techniques: feature extraction and fine-tuning. These techniques aim to improve VA text representations for COD classification tasks in the health domain. The framework is motivated by multi-step learning, where a final learning task is realized via a sequence of intermediate learning tasks. The framework builds upon the strengths of the Bidirectional Encoder Representations from Transformers (BERT) and Embeddings from Language Models (ELMo) models pretrained on the general English and biomedical domains. These models are employed to extract features from the VA narratives. Our results demonstrate improved performance when initializing the learning of BERT embeddings with ELMo embeddings. The benefit of incorporating character-level information for learning word embeddings in the English domain, coupled with word-level information for learning word embeddings in the biomedical domain, is also evident.

Sensor Fault Reconstruction Using Robustly Adaptive Unknown-Input Observers.

Sensors are a key component in industrial automation systems. A fault or malfunction in sensors may degrade control system performance. An engineering system model is usually disturbed by input uncertainties, which brings a challenge for monitoring, diagnosis, and control. In this study, a novel estimation technique, called adaptive unknown-input observer, is proposed to simultaneously reconstruct sensor faults as well as system states. Specifically, the unknown input observer is used to decouple partial disturbances, the un-decoupled disturbances are attenuated by the optimization using linear matrix inequalities, and the adaptive technique is explored to track sensor faults. As a result, a robust reconstruction of the sensor fault as well as system states is then achieved. Furthermore, the proposed robustly adaptive fault reconstruction technique is extended to Lipschitz nonlinear systems subjected to sensor faults and unknown input uncertainties. Finally, the effectiveness of the algorithms is demonstrated using an aircraft system model and robotic arm and comparison studies.

Fragmentation of colliding liquid rims

We present direct numerical simulations of the splashing process between two cylindrical liquid rims. This belongs to a class of impact and collision problems with a wide range of applications in science and engineering, and motivated here by splashing of breaking ocean waves. Interfacial perturbations with a truncated white noise frequency profile are introduced to the rims before their collision, whose subsequent morphological development is simulated by solving the two-phase incompressible Navier–Stokes equation with the adaptive mesh refinement technique, within the Basilisk software environment. We first derive analytical solutions predicting the unsteady interfacial and velocity profiles of the expanding sheet forming between the two rims, and develop scaling laws for the evolution of the lamella rim under capillary deceleration. We then analyse the formation and growth of transverse ligaments ejected from the lamella rims, which we find to originate from the initial corrugated geometry of the perturbed rim surface. Novel scaling models are proposed for predicting the decay of the ligament number density due to the ongoing ligament merging phenomenon, and found to agree well with the numerical results presented here. The role of the mechanism in breaking waves is discussed further and necessary next steps in the problem are identified.

Development of UroSAM: a machine learning model to automatically identify kidney stone composition from endoscopic video.

Introduction Chemical composition analysis is important in prevention counseling for kidney stone disease. Advances in laser technology have made dusting techniques more prevalent, but this offers no consistent way to collect enough material to send for chemical analysis, leading many to forgo this test. We developed a novel machine learning (ML) model to effectively assess stone composition based on intraoperative endoscopic video data. Methods Two endourologists performed ureteroscopy for kidney stones ≥ 10mm. Representative videos were recorded intraoperatively. Individual frames were extracted from the videos and the stone was outlined by human tracing. An ML model, UroSAM, was built and trained to automatically identify kidney stones in the images and predict the majority stone composition: calcium oxalate monohydrate (COM), dihydrate (COD), calcium phosphate (CAP), or uric acid (UA). UroSAM was built on top of the publicly available Segment Anything Model (SAM) and incorporated a U-Net convolutional neural network (CNN). Discussion A total of 78 ureteroscopy videos were collected; 50 were used for the model after exclusions (32 COM, 8 COD, 8 CAP, 2 UA). The ML model segmented the images with 94.77% precision. Dice coefficient (0.9135) and Intersection over Union (0.8496) confirmed good segmentation performance of the ML model. A video-wise evaluation demonstrated 60% correct classification of stone composition. Subgroup analysis showed correct classification in 84.4% of COM videos. A post-hoc adaptive threshold technique was used to mitigate biasing of the model towards COM due to data imbalance - this improved the overall correct classification to 62% while improving the classification of COD, CAP, and UA videos. Conclusions This study demonstrates the successful development of UroSAM, an ML model that precisely identifies kidney stones from natural endoscopic video data. More high quality video data will improve the performance of the model in classifying the majority stone composition.

Multi‐armed bandit based online model selection for concept‐drift adaptation

AbstractEnsemble methods are among the most effective concept‐drift adaptation techniques due to their high learning performance and flexibility. However, they are computationally expensive and pose a challenge in applications involving high‐speed data streams. In this paper, we present a computationally efficient heterogeneous classifier ensemble entitled OMS‐MAB which uses online model selection for concept‐drift adaptation by posing it as a non‐stationary multi‐armed bandit (MAB) problem. We use a MAB to select a single adaptive learner within the ensemble for learning and prediction while systematically exploring promising alternatives. Each ensemble member is made drift resistant using explicit drift detection and is represented as an arm of the MAB. An exploration factor controls the trade‐off between predictive performance and computational resource requirements, eliminating the need to continuously train and evaluate all the ensemble members. A rigorous evaluation on 20 benchmark datasets and 9 algorithms indicates that the accuracy of OMS‐MAB is statistically at par with state‐of‐the‐art (SOTA) ensembles. Moreover, it offers a significant reduction in execution time and model size in comparison to several SOTA ensemble methods, making it a promising ensemble for resource constrained stream‐mining problems.

Investigating the Dynamic Behavior of Integer and Noninteger Order System of Predation with Holling’s Response

The paper’s primary objective is to examine the dynamic behavior of an integer and noninteger predator–prey system with a Holling type IV functional response in the Caputo sense. Our focus is on understanding how harvesting influences the stability, equilibria, bifurcations, and limit cycles within this system. We employ qualitative and quantitative analysis methods rooted in bifurcation theory, dynamical theory, and numerical simulation. We also delve into studying the boundedness of solutions and investigating the stability and existence of equilibrium points within the system. Leveraging Sotomayor’s theorem, we establish the presence of both the saddle-node and transcritical bifurcations. The analysis of the Hopf bifurcation is carried out using the normal form theorem. The model under consideration is extended to the fractional reaction–diffusion model which captures non-local and long-range effects more accurately than integer-order derivatives. This makes fractional reaction–diffusion systems suitable for modeling phenomena with anomalous diffusion or memory effects, improving the fidelity of simulations in turn. An adaptable numerical technique for solving this class of differential equations is also suggested. Through simulation results, we observe that one of the Lyapunov exponents has a negative value, indicating the potential for the emergence of a stable-limit cycle via bifurcation as well as chaotic and complex spatiotemporal distributions. We supplement our analytical investigations with numerical simulations to provide a comprehensive understanding of the system’s behavior. It was discovered that both the prey and predator populations will continue to coexist and be permanent, regardless of the choice of fractional parameter.

Synchronization of coupled neural networks with multiple switching topologies via adaptive control techniques

Synchronization of coupled neural networks with multiple switching topologies via adaptive control techniques

Emotion Detection Through Electrocardiogram Signal Classification in an IOT Environment with Deep Neural Networks

An ECG detects the health and rhythm of the heart by measuring the electric activity of the heart. It has also been demonstrated that a person's emotions may influence the electrical activity of the heart. As a result, studying the electrical behaviour of the heart may simply determine a person's cardiac state and emotional wellness. IoT is a new technology that is quickly gaining acceptance throughout the world. Anybody, at any time, from anywhere, may connect to any network or service because to the extraordinary power and capacity of IoT. IoT-enabled devices have revolutionized the medical business by providing new capabilities such as remote patient monitoring and self-monitoring. This research proposed an IoT-based ECG monitoring system that employs a heart rate sensor to generate data and an intelligent hybrid classification algorithm to categorize the data. ECG monitoring has become a widely used method for detecting cardiac problems. The following are the primary contributions of this paper: To begin, this paper describes WISE (Wearable IoT-cloud-based health monitoring system), a one-of-a-kind system for real-time personal health monitoring. WISE makes use of the BASN (body area sensor network) technology to provide real-time health monitoring. WISE rapidly transfers data from the BASN to the cloud, and a lightweight wearable LCD may be included to enable quick access to real-time data. This model can address the issue of class imbalance in the ECG dataset, assisting in the development of an IoT-based smart and accurate healthcare system. Pre-processing, feature extraction, and classification are the three steps in any classification technique, whether it is emotion classification or heart health classification. Sensors are used to collect an ECG signal from a person's outside body. The provided ECG signal is first pre-processed using the Butterworth Filtering Method, which effectively reduces noise from the signal. Following pre-processing, the Adaptive Discrete Wavelet Transform technique is used to anticipate the signal's attributes. Lastly, a decision making classification approach based on relational weights is used to determine if the ECG signal is normal or abnormal.

Intelligent in-process enhancement technique for machining efficiency in CNC machine tools based on spindle power

In the realm of mechanical machining, adaptive machining techniques offer an efficient method. However, existing adaptive machining technologies cannot automatically identify the type of workpiece to be machined, nor can they set different control targets based on tool load-bearing capacity. This often requires manual and cumbersome operations, making the application of adaptive machining technology difficult to popularize. To address these challenges, this paper proposes an intelligent and efficient adaptive machining method. Specifically, a workpiece automatic classification algorithm based on machine tool spindle power is introduced. This algorithm classifies the machined workpieces according to the collected spindle power data, enhancing the convenience of adaptive machining technology. Also based on spindle power data, a time-series segmentation method is proposed, dividing spindle power into different subsequences according to the tools used in machining, which enhances the accuracy of adaptive machining technology. Furthermore, an adaptive feed rate control algorithm is designed based on fuzzy control theory, realizing intelligent and efficient adaptive machining of workpieces to improve machining efficiency. Finally, experiments are conducted to validate the effectiveness of the intelligent and efficient adaptive machining method. The proposed method effectively addresses the drawbacks of existing adaptive machining method, which require complex parameter settings. It simplifies the operational process, enhances the intelligence, and improves the practical applicability of adaptive machining techniques. This represents an innovative contribution to the field of adaptive machining techniques.

A Look at Genomic Selection Techniques for Climate Change Adaptation and Production in Livestock

Livestock production profoundly intersects with global climate dynamics, contributing to greenhouse gas emissions and confronting vulnerability to climate impacts. In addressing these challenges, imperative adjustments are requisite to fortify the climate robustness of livestock systems. Notably, the prevalent reliance on commercial breeds with limited genetic diversity exposes production strategies to disruption, especially if these breeds are confined to environments that may lose economic viability under future climate scenarios. Consequently, understanding the adaptability of animal populations to forthcoming environmental conditions is paramount for sustaining livestock production. Assessing the genetic underpinnings of climate adaptation necessitates the exploration of tailored genomic selection techniques encompassing both production traits, presumed to have moderate heritability, and adaptation traits, presumed to have low heritability. Through a nuanced examination of genomic selection dynamics, insights into the genetic mechanisms fostering resilience in livestock populations amidst shifting environmental contexts are garnered. Employ genomic analysis to pinpoint genetic markers associated with traits like heat tolerance, disease resistance, and feed efficiency in livestock. Collaborate across disciplines to develop tailored breeding programs integrating these markers, and validate their effectiveness through rigorous field trials and ongoing monitoring to enhance livestock resilience and productivity in varied climatic conditions. Elucidating these mechanisms and their application in breeding programs offers a comprehensive understanding of how genetic advancements can enhance both production efficiency and climate resilience in livestock. This discourse aims to bridge the chasm between scientific inquiry and pragmatic implementation, thereby facilitating informed decision-making in livestock breeding strategies tailored to mitigate the ramifications of climate change.