Ensemble Of Sets Research Articles

Improving Brain Tumor Classification Efficacy through the Application of Feature Selection and Ensemble Classifiers

Accurate brain tumor detection is crucial due to its high mortality rate. However, existing automated methods suffer from limited accuracy and high false-positive rates. In this study, we aimed to improve brain tumor classification by comparing 17 different classifiers organized into six groups: Decision Tree (DT) Model, Support Vector Machine (SVM), Naive Bayes Classifier, Logistic Regression, Generalized Linear Model (GLM) Classifier, and Neural Network. We utilized a dataset of 3,762 Magnetic Resonance Imaging (MRI) scans of brain tumors from Kaggle, with each image having dimensions of 240 × 240 pixels and labeled as tumor or non-tumor. Our approach involved three main steps: extracting visual information using 17 predictor classes, optimizing feature extraction through weight optimization, and comparing different sets of classifier models. We evaluated the models’ performance using the confusion matrix and Receiver Operating Characteristics (ROC) curves. Our results showed that optimizing feature selection and utilizing ensemble classifiers improved the accuracy of brain tumor classification. The DT Model with ensemble classifiers emerged as the best-performing classifier, achieving an accuracy of 98.11% and an AUC of 0.99. Notably, Random Tree (RT) exhibited the highest accuracy within the ensemble classifier set, with a significant increase compared to other models. Our proposed method outperformed the standard approach, demonstrating its potential for enhancing brain tumor detection accuracy. This study contributes to the field by providing a more accurate method for detecting brain tumors, potentially enabling earlier detection and improved patient outcomes. Future research should focus on further improving brain tumor diagnosis and treatment through the application of machine learning techniques.

Comparing explicit and implicit ensemble perception: 3 stimulus variables and 3 presentation modes

Visual scenes are too complex for one to immediately perceive all their details. As suggested by Gestalt psychologists, grouping similar scene elements and perceiving their summary statistics provides one shortcut for evaluating scene gist. Perceiving ensemble statistics overcomes processing, attention, and memory limits, facilitating higher-order scene understanding. Ensemble perception spans simple/complex dimensions (circle size, face emotion), including various statistics (mean, range), and inherently spans space and/or time, when sets are presented scattered across the visual scene, and/or sequentially in rapid series. Furthermore, ensemble perception occurs explicitly, when observers are asked to judge set mean, and also automatically/implicitly, when observers are engaged in an orthogonal task. We now study relationships among these ensemble-perception phenomena, testing explicit and implicit ensemble perception; for sets varying in circle size, line orientation, or disc brightness; and with spatial, temporal or spatio-temporal presentation. Following ensemble set presentation, observers were asked if a test image, or which of two test images, had been present in the set. Confirming previous results, responses reflected implicit mean perception, depending on test image distance from the mean, and on its being within or outside ensemble range. Subsequent experiments asked the same observers to explicitly judge whether test images were larger, more clockwise, or brighter than the set mean, or which of two test images was closer to the mean. Comparing implicit and explicit mean perception, we find that explicit ensemble averaging is more precise than implicit mean perception—for each ensemble variable and presentation mode. Implications are discussed regarding possible separate mechanisms for explicit versus implicit ensemble perception.

Quantifying uncertainty for deep learning based forecasting and flow-reconstruction using neural architecture search ensembles

Classical problems in computational physics such as data-driven forecasting and signal reconstruction from sparse sensors have recently seen an explosion in deep neural network (DNN) based algorithmic approaches. However, most DNN models do not provide uncertainty estimates, which are crucial for establishing the trustworthiness of these techniques in downstream decision making tasks and scenarios. In recent years, ensemble-based methods have achieved significant success for the uncertainty quantification in DNNs on a number of benchmark problems. However, their performance on real-world applications remains under-explored. In this work, we present an automated approach to DNN discovery and demonstrate how this may also be utilized for ensemble-based uncertainty quantification. Specifically, we propose the use of a scalable neural and hyperparameter architecture search for discovering an ensemble of DNN models for complex dynamical systems. We highlight how the proposed method not only discovers high-performing neural network ensembles for our tasks, but also quantifies uncertainty seamlessly. This is achieved by using genetic algorithms and Bayesian optimization for sampling the search space of neural network architectures and hyperparameters. Subsequently, a model selection approach is used to identify candidate models for an ensemble set construction. Afterwards, a variance decomposition approach is used to estimate the uncertainty of the predictions from the ensemble. We demonstrate the feasibility of this framework for two tasks — forecasting from historical data and flow reconstruction from sparse sensors for the sea-surface temperature. We demonstrate superior performance from the ensemble in contrast with individual high-performing models and other benchmarks.

IARMTS: Design of an Interference-Aware Routing Model with Time Synchronization Capabilities for Dense Wireless Sensor Network Deployments

Performance of dense wireless sensor networks is often degraded due to communication interference and time synchronization issues. Existing machine learning & deep learning models that propose bioinspired & pre-emptive packet-analysis solutions for these tasks either have high complexity, or high deployment costs. Moreover, these models cannot be scaled for heterogeneous node & traffic types, which limits their applicability when applied to real-time scenarios. To overcome these issues, this text proposes design of an interference-aware routing model with time synchronization capabilities for dense wireless sensor network deployments. The network initially collects temporal clock states & packet delivery performance of different nodes on heterogeneous traffic scenarios. These traffic patterns are converted into frequency, entropy, Gabor, and Wavelet components. The converted components are used to train an ensemble set of Naïve Bayes (NB), k Nearest Neighbour (kNN), Multilayer Perceptron (MLP), and Support Vector Machine (SVM) classifiers. These classifiers assist in identification of optimal clock deviations and set of routing paths. These routing paths are further fine-tuned via use of a Bacterial Foraging Optimization (BFO) Model, which assists in identification of interference-aware paths. The BFO Model uses a temporal fitness function that fuses throughput, communication delay, energy levels, and packet delivery performance for different set of contextual communications. Due to which, the model is able to showcase lower end-to-end delay, higher throughput, lower energy consumption, and higher packet delivery performance when compared with existing routing methods under high density nodes & heterogeneous network scenarios. The model showcases 99% PDR, 18.3% lower delay, 19.5% higher energy efficiency and 10.4% lower delay levels when compared with existing methods.

RANDOM FOREST MODEL ON BIG DATA FOR FORECASTING THE STABILITY OF THE BANKING SYSTEM OF THE RUSSIAN FEDERATION

The article analyzes current trends in the use of artificial intelligence, cognitive technologies, Big Data and other innovations in order to predict the stability of the Russian banking system. The current trends in the development of banking innovations in the context of digitalization of the economy are summarized. Among the research methods used, the Random Forest artificial intelligence system, formed in the Google Collab cloud service in Python, should be noted. A neural network model for predicting the dynamics of bank profits based on the use of the Random Forest (RF – random forest) machine learning model is proposed. The results of the study indicate that, with the average value of the profit of banks for the period 2010-2021. 1126.0416 billion rubles, the mean absolute error (Mean Absolute Error) amounted to 129.96666 billion rubles, or 11.5%, which can be regarded as a good result in conditions of market uncertainty. The analysis showed that current trends are global in nature and lead to qualitative changes in the flow of business processes between consumers of banking services and credit institutions.
 Purpose: is to form a model that predicts the value of the target variable – the profit of the banking system using the Random Forest deep learning model, which is an ensemble of “decision trees”. The scientific novelty is to put forward and prove the hypothesis that with the help of the formed artificial intelligence system “Random Forest”, a profit forecast for commercial banks of the Russian Federation can be obtained.
 Method or methodology of the work. The work used such research methods as: monographic, analytical, artificial intelligence system – “Random Forest”, as well as comparative analysis. Calculations were performed in XL tables, Google Collab cloud service in Python.
 Results. The developed AI-system “Random Forest”, designed to predict the profits of banks, is presented. It is of practical importance to use the results of the Random Forest neural network to predict the development of credit institutions, which makes it possible to solve a major national economic problem - ensuring the stability of the banking system of the Russian Federation. A neural network model for predicting the dynamics of bank profits based on the use of the Random Forest (RF) machine learning model is proposed, which is a machine learning algorithm that consists in using an ensemble (set) of decision trees.
 The scope of the results: the financial sector, the banking sector, forecasting and ensuring the sustainable activity of credit institutions in the Russian Federation.

Diverging Trends in Rain‐On‐Snow Over High Mountain Asia

AbstractRain‐on‐snow (ROS) over snow‐dominated regions such as High Mountain Asia (HMA) modulates snowmelt and runoff and is key contributor in influencing water availability and hazards (e.g., floods and landslides). We studied the trends in ROS in HMA over the past two decades from 2001 to 2018 using the land surface model Noah‐MP driven by an ensemble precipitation data set. Our results show that changes in precipitation phase and rainfall are altering ROS. Because of the strong physical heterogeneity and atmospheric dynamics of HMA, ROS characteristics and trends are region‐dependent and ROS occurs predominantly over the Indus, Ganges‐Brahmaputra, and northwestern basins. In the Indus, ROS representing ∼5% of the annual precipitation and ∼20% of the annual snowmelt, has an increasing trend. This is contrary to the Ganges‐Brahmaputra characterized by decreasing ROS trends, where it represents ∼11% of the annual precipitation and ∼60% of the annual snowmelt. In the northwestern basins, ROS has bidirectional trends due to elevation patterns and trends in rainfall, and it constitutes ∼5 to ∼10% of the annual precipitation. Increasing trends in ROS over Indus contribute to reducing the snowpack in late summer, with concerns of reduced water availability and increased groundwater exploitation. Similarly, because of its high amount and contribution to snowmelt, the decreasing ROS trends in the Ganges‐Brahmaputra will have consequences of decreased recharge from the headwaters and exacerbated use of groundwater unless increasing trends in rainfall compensate for the decreasing snowmelt. These results provide new insights on ROS‐driven changes in the hydrological cycle over HMA.

Stochastic (re)constructions of non-stationary material structures: Using ensemble averaged correlation functions and non-uniform phase distributions

Depending on the spatial distribution of mass or phases within the studied materials, they can be classified into two distinct property types: stationary and non-stationary. Stochastic reconstruction with simulated annealing and correlation functions is a powerful technique to produce material structural images from limited input data, but they usually deal with statistically homogeneous images only. In this work we show that the alteration of the initial phase distributions before starting annealing can produce non-stationary structures. The degree of such non-stationarity depends on numerous parameters of the annealing, including maximum correlation length, the choice of correlation function weights, annealing cooling schedule, in a convoluted manner—this makes it impossible to segregate the influence of each parameter. In general, we have shown that it is possible to choose such parameters in order to suppress the inhomogeneity of the resulting replicas. The degree of stationarity of the stochastic reconstructions had a significant influence on the physical properties of the reconstructed binary structures—computed full permeability tensors showed different degree of anisotropy and off-diagonal terms values depending on the reconstruction parameters. The proposed approach to produce non-stationary structures from ensemble averaged set of correlation functions opens numerous ways to attack theoretical and practical problems with natural and artificial materials with statistically inhomogeneous structure.

Cosmological inference from an emulator based halo model. II. Joint analysis of galaxy-galaxy weak lensing and galaxy clustering from HSC-Y1 and SDSS

We present high-fidelity cosmology results from a blinded joint analysis of galaxy-galaxy weak lensing (ΔΣ) and projected galaxy clustering (wp) measured from the Hyper Suprime-Cam Year-1 (HSC-Y1) data and spectroscopic Sloan Digital Sky Survey (SDSS) galaxy catalogs in the redshift range 0.15<z<0.7. We define luminosity-limited samples of SDSS galaxies to serve as the tracers of wp in three spectroscopic redshift bins, and as the lens samples for ΔΣ. For the ΔΣ measurements, we select a single sample of 4×106 source galaxies over 140 deg2 from HSC-Y1 with photometric redshifts (photo z) greater than 0.75, enabling a better handle of photo-z errors by comparing the ΔΣ amplitudes for the three lens redshift bins. The deep, high-quality HSC-Y1 data enable significant detections of the ΔΣ signals, with integrated signal-to-noise ratio S/N∼15 in the range 3≤R/[h−1 Mpc]≤30 for the three lens samples, despite the small area coverage. For cosmological parameter inference, we use an input galaxy-halo connection model built on the dark emulator package (which uses an ensemble set of high-resolution N-body simulations and enables fast, accurate computation of the clustering observables) with a halo occupation distribution that includes nuisance parameters to marginalize over modeling uncertainties. We model the ΔΣ and wp measurements on scales from R≃3 and 2 h−1 Mpc, respectively, up to 30 h−1 Mpc (therefore excluding the baryon acoustic oscillations information) assuming a flat ΛCDM cosmology, marginalizing over about 20 nuisance parameters and demonstrating the robustness of our results to them. With various tests using mock catalogs described in Miyatake et al. [preceding paper, Phys. Rev. D 106, 083519 (2022)], we show that any bias in the clustering amplitude S8≡σ8(Ωm/0.3)0.5 due to uncertainties in the galaxy-halo connection is less than ∼50% of the statistical uncertainty of S8, unless the assembly biaseffect is unexpectedly large. Our best-fit models have S8=0.795−0.042+0.049 (mode and 68% credible interval) for the flat ΛCDM model; we find tighter constraints on the quantity S8(α=0.17)≡σ8(Ωm/0.3)0.17=0.745−0.031+0.039.23 MoreReceived 29 November 2021Accepted 22 August 2022DOI:https://doi.org/10.1103/PhysRevD.106.083520© 2022 American Physical SocietyPhysics Subject Headings (PhySH)Research AreasCosmologyDark energyGravitational lensesGravitation, Cosmology & Astrophysics

An uncertainty-aware deep learning architecture with outlier mitigation for prostate gland segmentation in radiotherapy treatment planning.

Task automation is essential for efficient and consistent image segmentation in radiation oncology. We report on a deep learning architecture, comprising a U-Net and a variational autoencoder (VAE) for automatic contouring of the prostate gland incorporating interobserver variation for radiotherapy treatment planning. The U-Net/VAE generates an ensemble set of segmentations for each image CT slice. A novel outlier mitigation (OM) technique was implemented to enhance the model segmentation accuracy. The primary source dataset (source_prim) consisted of 19 200 CT slices (from 300 patient planning CT image datasets) with manually contoured prostate glands. A smaller secondary source dataset (source_sec) comprised 640 CT slices (from 10 patient CT datasets), where prostate glands were segmented by 5 independent physicians on each dataset to account for interobserver variability. Data augmentation via random rotation (<5 degrees), cropping, and horizontal flipping was applied to each dataset to increase sample size by a factor of 100. A probabilistic hierarchical U-Net with VAE was implemented and pretrained using the augmented source_prim dataset for 30 epochs. Model parameters of the U-Net/VAE were fine-tuned using the augmented source_sec dataset for 100 epochs. After the first round of training, outlier contours in the training dataset were automatically detected and replaced by the most accurate contours (based on Dice similarity coefficient, DSC) generated by the model. The U-Net/OM-VAE was retrained using the revised training dataset. Metrics for comparison included DSC, Hausdorff distance (HD, mm), normalized cross-correlation (NCC) coefficient, and center-of-mass (COM) distance (mm). Results for U-Net/OM-VAE with outliers replaced in the training dataset versus U-Net/VAE without OM were as follows: DSC=0.82±0.01 versus 0.80±0.02 (p=0.019), HD=9.18±1.22 versus 10.18±1.35mm (p=0.043), NCC=0.59±0.07 versus 0.62±0.06, and COM=3.36±0.81 versus 4.77±0.96mm over the average of 15 contours. For the average of 15 highest accuracy contours, values were as follows: DSC=0.90±0.02 versus 0.85±0.02, HD=5.47±0.02 versus 7.54±1.36mm, and COM=1.03±0.58 versus 1.46±0.68mm (p<0.03 for all metrics). Results for the U-Net/OM-VAE with outliers removed were as follows: DSC=0.78±0.01, HD=10.65±1.95mm, NCC=0.46±0.10, COM=4.17±0.79mm for the average of 15 contours, and DSC=0.88±0.02, HD=7.00±1.17mm, COM=1.58±0.63mm for the average of 15 highest accuracy contours. All metrics for U-Net/VAE trained on the source_prim and source_sec datasets via pretraining, followed by fine-tuning, show statistically significant improvement over that trained on the source_sec dataset only. Finally, all metrics for U-Net/VAE with or without OM showed statistically significant improvement over those for the standard U-Net. A VAE combined with a hierarchical U-Net and an OM strategy (U-Net/OM-VAE) demonstrates promise toward capturing interobserver variability and produces accurate prostate auto-contours for radiotherapy planning. The availability of multiple contours for each CT slice enables clinicians to determine trade-offs in selecting the "best fitting" contour on each CT slice. Mitigation of outlier contours in the training dataset improves prediction accuracy, but one must be wary of reduction in variability in the training dataset.

Multimedia Vision Improvement and Simulation in Consideration of Virtual Reality Reconstruction Algorithms

Due to the large noise and many discrete points of the image in the traditional image reconstruction process, the reconstruction quality of the image deviates greatly from the actual target. In this study, the virtual reality reconstruction algorithm is applied to multimedia vision, the virtual reality image is corrected by using the binocular offset positioning method, the denoising process is performed in the image reconstruction process, and the high-pass filter matrix is used to improve the image reproduction. At the same time, the three-dimensional reconstruction algorithm is used to perform correlation retrieval, the ensemble point set and the discrete point set are obtained, the maximum and minimum reconstruction degree areas are clarified, and the deviation reconstruction and peak relocation can be performed. Finally, the experimental test results show that the algorithm in this study can enhance the authenticity of image reconstruction, improve the accuracy of image corner detection, and effectively reduce the noise interference in the process of reconstructing the image.

Climate change-induced reduction in agricultural land suitability of West-Africa's inland valley landscapes

CONTEXTAlthough rice production has increased significantly in the last decade in West Africa, the region is far from being rice self-sufficient. Inland valleys (IVs) with their relatively higher water content and soil fertility compared to the surrounding uplands are the main rice-growing agroecosystem. They are being promoted by governments and development agencies as future food baskets of the region. However, West Africa's crop production is estimated to be negatively affected by climate change due to the strong dependence of its agriculture on rainfall. OBJECTIVEThe main objective of the study is to apply a set of machine learning models to quantify the extent of climate change impact on land suitability for rice using the presence of rice-only data in IVs along with bioclimatic indicators. METHODSWe used a spatially explicit modeling approach based on correlative Ecological Niche Modeling. We deployed 4 algorithms (Boosted Regression Trees, Generalized Linear Model, Maximum Entropy, and Random Forest) for 4-time periods (the 2030s, 2050s, 2070s, and 2080s) of the 4 Representative Concentration Pathways (RCP2.6, RCP4.5, RCP6.0, and RCP8) from an ensemble set of 32 spatially downscaled and bias-corrected Global Circulation Models climate data. RESULTS AND CONCLUSIONSThe overall trend showed a decrease in suitable areas compared to the baseline as a function of changes in temperature and precipitation by the order of 22–33% area loss under the lowest reduction scenarios and more than 50% in extreme cases. Isothermality or how large the day to night temperatures oscillate relative to the annual oscillations has a large impact on area losses while precipitation increase accounts for most of the areas with no change in suitability. Strong adaptation measures along with technological advancement and adoption will be needed to cope with the adverse effects of climate change on inland valley rice areas in the sub-region. SIGNIFICANCEThe demand for rice in West Africa is huge. For the rice self-sufficiency agenda of the region, “where” and “how much” land resources are available is key and requires long-term, informed planning. Farmers can only adapt when they switch to improved breeds, providing that they are suited for the new conditions. Our results stress the need for land use planning that considers potential climate change impacts to define the best areas and growing systems to produce rice under multiple future climate change uncertainties.

Full-shape cosmology analysis of the SDSS-III BOSS galaxy power spectrum using an emulator-based halo model: A 5% determination of σ8

We present the results obtained from the full-shape cosmology analysis of the redshift-space power spectra for 4 galaxy samples of the SDSS-III BOSS DR12 galaxy catalog over $0.2 < z < 0.75$. For the theoretical template, we use an emulator that was built from an ensemble set of $N$-body simulations, which enables fast and accurate computation of the redshift-space power spectrum of halos. Combining with the halo occupation distribution to model the halo-galaxy connection, we can compute the redshift-space power spectrum of BOSS-like galaxies in the flat $\Lambda$CDM cosmology. In our cosmology inference, we use the power spectrum monopole, quadrupole and hexadecapole and include 7 nuisance parameters to model uncertainties in the halo-galaxy connection for each galaxy sample, but do not use any information on the abundance of galaxies. We demonstrate a validation of our analysis pipeline using the mock catalogs of BOSS-like galaxies, generated using different recipes of the halo-galaxy connection and including the assembly bias effect. Assuming weak priors on cosmological parameters, except for $\Omega_{\rm b}h^2$ and $n_{\rm s}$, we show that our model well reproduces the BOSS power spectra. Including the power spectrum information up to $k_{\rm max}=0.25\,h{\rm Mpc}^{-1}$, we find $\Omega_\mathrm{m}=0.301^{+0.012}_{-0.011}$, $H_0=68.2 \pm 1.4~\mathrm{km\,s}^{-1}\mathrm{Mpc}^{-1}$, and $\sigma_8=0.786^{+0.036}_{-0.037}$, for the mode and 68\% credible interval, after marginalization over nuisance parameters. We find little improvement in the cosmological parameters beyond a maximum wavelength $k_{\rm max}\simeq 0.2\,h\,{\rm Mpc}^{-1}$ due to the shot noise domination and marginalization of the halo-galaxy connection parameters. Our results are consistent with the Planck CMB results within $1\sigma$ statistical uncertainties.

Regional opening strategies with commuter testing and containment of new SARS-CoV-2 variants in Germany

BackgroundDespite the vaccination process in Germany, a large share of the population is still susceptible to SARS-CoV-2. In addition, we face the spread of novel variants. Until we overcome the pandemic, reasonable mitigation and opening strategies are crucial to balance public health and economic interests.MethodsWe model the spread of SARS-CoV-2 over the German counties by a graph-SIR-type, metapopulation model with particular focus on commuter testing. We account for political interventions by varying contact reduction values in private and public locations such as homes, schools, workplaces, and other. We consider different levels of lockdown strictness, commuter testing strategies, or the delay of intervention implementation. We conduct numerical simulations to assess the effectiveness of the different intervention strategies after one month. The virus dynamics in the regions (German counties) are initialized randomly with incidences between 75 and 150 weekly new cases per 100,000 inhabitants (red zones) or below (green zones) and consider 25 different initial scenarios of randomly distributed red zones (between 2 and 20% of all counties). To account for uncertainty, we consider an ensemble set of 500 Monte Carlo runs for each scenario.ResultsWe find that the strength of the lockdown in regions with out of control virus dynamics is most important to avoid the spread into neighboring regions. With very strict lockdowns in red zones, commuter testing rates of twice a week can substantially contribute to the safety of adjacent regions. In contrast, the negative effect of less strict interventions can be overcome by high commuter testing rates. A further key contributor is the potential delay of the intervention implementation. In order to keep the spread of the virus under control, strict regional lockdowns with minimum delay and commuter testing of at least twice a week are advisable. If less strict interventions are in favor, substantially increased testing rates are needed to avoid overall higher infection dynamics.ConclusionsOur results indicate that local containment of outbreaks and maintenance of low overall incidence is possible even in densely populated and highly connected regions such as Germany or Western Europe. While we demonstrate this on data from Germany, similar patterns of mobility likely exist in many countries and our results are, hence, generalizable to a certain extent.

Transfer learning based novel ensemble classifier for COVID-19 detection from chest CT-scans

Coronavirus Disease 2019 (COVID-19) is a deadly infection that affects the respiratory organs in humans as well as animals. By 2020, this disease turned out to be a pandemic affecting millions of individuals across the globe. Conducting rapid tests for a large number of suspects preventing the spread of the virus has become a challenge. In the recent past, several deep learning based approaches have been developed for automating the process of detecting COVID-19 infection from Lung Computerized Tomography (CT) scan images. However, most of them rely on a single model prediction for the final decision which may or may not be accurate. In this paper, we propose a novel ensemble approach that aggregates the strength of multiple deep neural network architectures before arriving at the final decision. We use various pre-trained models such as VGG16, VGG19, InceptionV3, ResNet50, ResNet50V2, InceptionResNetV2, Xception, and MobileNet and fine-tune them using Lung CT Scan images. All these trained models are further used to create a strong ensemble classifier that makes the final prediction. Our experiments exhibit that the proposed ensemble approach is superior to existing ensemble approaches and set state-of-the-art results for detecting COVID-19 infection from lung CT scan images.

A Combined Control Systems and Machine Learning Approach to Forecasting Iceberg Flux off Newfoundland

Icebergs have long been a threat to shipping in the NW Atlantic and the iceberg season of February to late summer is monitored closely by the International Ice Patrol. However, reliable predictions of the severity of a season several months in advance would be useful for planning monitoring strategies and also for shipping companies in designing optimal routes across the North Atlantic for specific years. A seasonal forecast model of the build-up of seasonal iceberg numbers has recently become available, beginning to enable this longer-term planning of marine operations. Here we discuss extension of this control systems model to include more recent years within the trial ensemble sample set and also increasing the number of measures of the iceberg season that are considered within the forecast. These new measures include the seasonal iceberg total, the rate of change of the seasonal increase, the number of peaks in iceberg numbers experienced within a given season, and the timing of the peak(s). They are predicted by a range of machine learning tools. The skill levels of the new measures are tested, as is the impact of the extensions to the existing seasonal forecast model. We present a forecast for the 2021 iceberg season, predicting a medium iceberg year.

GBoost: A novel Grading-AdaBoost ensemble approach for automatic identification of erythemato-squamous disease

Ensemble learning is one of the powerful machine learning approaches that is generally used to strengthen models by combining the performances of several weak learners. It holds a great potential for solving umpteen problems in healthcare domain by enabling health systems to use data analytically for identifying best practices that improves healthcare and additionally reduces the cost too. The main focus of the present work is the automatic identification of erythemato-squamous disease (ESD) with higher accuracy performance, thereby, an ESD prediction system has been proposed using ensemble approach. The present study introduces GBoost (GB) ensemble framework that is based on grading approach with AdaBoost scheme for analysis and prediction of erythemato-squamous disease (ESD). The experiments were performed using dermatology dataset. The ESD prediction system uses imputation and filter approaches for data preprocessing and includes two phases for building models. In the first phase, models have been built using individual classifiers without using any ensemble technique whereas the second phase includes the GB ensemble along with dynamic base-classifiers and static meta-classifier for model building. At the end, the best classifier from phase one (without using GB ensemble framework) has been compared with the best GB ensemble set (using GB ensemble framework) from phase 2 to obtain the overall best model for ESD prediction. The proposed ESD prediction system using GB ensemble framework has achieved an accuracy of 99.45% which is higher than all the previous works on this dataset. The use of ensemble learning in this study exhibits a remarkable performance in the automatic identification of Erythemato-sequamous disease (ESD) with augmented accuracy.

Mechanical unfolding of ensemble biomolecular structures by shear force.

Mechanical unfolding of biomolecular structures has been exclusively performed at the single-molecule level by single-molecule force spectroscopy (SMFS) techniques. Here we transformed sophisticated mechanical investigations on individual molecules into a simple platform suitable for molecular ensembles. By using shear flow inside a homogenizer tip, DNA secondary structures such as i-motifs are unfolded by shear force up to 50 pN at a 77 796 s−1 shear rate. We found that the larger the molecules, the higher the exerted shear forces. This shear force approach revealed affinity between ligands and i-motif structures. It also demonstrated a mechano-click reaction in which a Cu(i) catalyzed azide–alkyne cycloaddition was modulated by shear force. We anticipate that this ensemble force spectroscopy method can investigate intra- and inter-molecular interactions with the throughput, accuracy, and robustness unparalleled to those of SMFS methods.

A generic approach to decipher the mechanistic pathway of heterogeneous protein aggregation kinetics.

Amyloid formation is a generic property of many protein/polypeptide chains. A broad spectrum of proteins, despite having diversity in the inherent precursor sequence and heterogeneity present in the mechanism of aggregation produces a common cross β-spine structure that is often associated with several human diseases. However, a general modeling framework to interpret amyloid formation remains elusive. Herein, we propose a data-driven mathematical modeling approach that elucidates the most probable interaction network for the aggregation of a group of proteins (α-synuclein, Aβ42, Myb, and TTR proteins) by considering an ensemble set of network models, which include most of the mechanistic complexities and heterogeneities related to amyloidogenesis. The best-fitting model efficiently quantifies various timescales involved in the process of amyloidogenesis and explains the mechanistic basis of the monomer concentration dependency of amyloid-forming kinetics. Moreover, the present model reconciles several mutant studies and inhibitor experiments for the respective proteins, making experimentally feasible non-intuitive predictions, and provides further insights about how to fine-tune the various microscopic events related to amyloid formation kinetics. This might have an application to formulate better therapeutic measures in the future to counter unwanted amyloidogenesis. Importantly, the theoretical method used here is quite general and can be extended for any amyloid-forming protein.

Power System Event Classification and Localization Using a Convolutional Neural Network

Detection and timely identification of power system disturbances are essential for situation awareness and reliable electricity grid operation. Because records of actual events in the system are limited, ensemble simulation-based events are needed to provide adequate data for building event-detection models through deep learning; e.g., a convolutional neural network (CNN). An ensemble numerical simulation-based training data set have been generated through dynamic simulations performed on the Polish system with various types of faults in different locations. Such data augmentation is proven to be able to provide adequate data for deep learning. The synchronous generators’ frequency signals are used and encoded into images for developing and evaluating CNN models for classification of fault types and locations. With a time-domain stacked image set as the benchmark, two different time-series encoding approaches, i.e., wavelet decomposition-based frequency-domain stacking and polar coordinate system-based Gramian Angular Field (GAF) stacking, are also adopted to evaluate and compare the CNN model performance and applicability. The various encoding approaches are suitable for different fault types and spatial zonation. With optimized settings of the developed CNN models, the classification and localization accuracies can go beyond 84 and 91%, respectively.

Wind speed big data forecasting using time-variant multi-resolution ensemble model with clustering auto-encoder

The fluctuation of wind speed brings difficulty for wind energy utilization. Wind speed forecasting can improve safety of wind power integration. To better forecast fluctuating wind speed, a time-variant multi-resolution ensemble model is proposed, which contains three stages. In stage I, the proposed model uses low-resolution and high-resolution datasets to build base forecasting models. In this manner, the additional fluctuation information in the high-resolution big data is used for better forecasting. In stage II, the wind speed is clustered with a clustering auto-encoder algorithm, and an ensemble weight set is obtained by optimizing weights on each cluster respectively. The clustering auto-encoder algorithm can analyze the nonlinear fluctuation, and generate clustering-friendly features. In stage III, by switching ensemble weights over time, the time-variant ensemble forecasting results can be obtained by combining the base models. The time-variant structure can track the changes of fluctuating characteristics over time, further improving forecasting performance. Four groups of real wind speed high-resolution big data with 51,840 sample points and the corresponding low-resolution data with 8640 sample points are utilized for case study. The forecasting results indicate: (a) the proposed model can effectively predict the wind speed. Taking 1-step forecasting as an example, the mean absolute errors of the proposed model for dataset #1, #2, #3 and #4 are 0.6318 m/s, 0.6957 m/s, 0.6121 m/s and 0.6175 m/s, respectively; (b) the proposed model outperforms other benchmark models.