Local Neighborhood Research Articles

Mobility heterogeneity of urban freight areas: Geospatial evidence from shared logistics dynamics

Amid rising mobility services in cities, the logistics sector plays a crucial role in envisioning daily services for local neighborhoods. Previous research has primarily identified key freight areas (KFAs) in small-scale regions using traditional census and travel survey data. However, differing from KFAs, other areas with distinct freight patterns remain largely unexamined, and there is a lack of comprehensive indicators to evaluate the travel behaviors of large-scale freight vehicles. This study introduces a data-driven geospatial framework to characterize and understand the patterns exhibited by different freight areas through freight dynamics data collected in cities. To analyze key aspects of freight mobility within spatial units, a set of areal indicators is developed using GPS trajectories of shared freight vehicles in Hong Kong. Descriptive statistics illustrate the inherent differences across various areas and their spatial distributions. A hierarchical clustering approach groups all areas based on proposed indicators, providing a quantitative evaluation of collective spatial patterns. The analysis also explores the non-linear relationships between different categories of freight activities and the local built environments and socioeconomic conditions. The results suggest a strong spatial heterogeneity in the areal mobility profile of Hong Kong. Loading and unloading behaviors are frequently observed in districts such as Eastern Kowloon, Kwai Chung, and the Airport, highlighting prevalent freight trip origins and destinations. In contrast, areas with high transit times are mostly found in suburban regions, particularly on the periphery. This study provides essential insights for area-based planning and management of urban freight activities, with relevant practical implications.

Discovering and exploring conceptions of geography teaching: a teacher education starter pack

Initial teacher education, among other things, significantly influences what conceptions of geography teaching build the steppingstones at the beginning of pre-service teachers’ careers. Using a questionnaire, we conducted a survey on Czech pre-service geography teachers’ conceptions of geography teaching (n = 183). Our findings indicate that the Interactionist conception was prevalent, representing the exploration of interconnections and relations between humans and the environment. Pre-service teachers also emphasized the importance of the Localist conception that builds on exploring geographies of the local neighborhood. A path analysis revealed that the pre-service teachers’ conceptions were not much interconnected, often existing as independent, isolated entities. Our results attempt to provide implications for shaping the professional geography knowledge landscape of future geography teachers through a complex “starter pack” – a balanced representation of a range of conceptions of geography teaching in initial teacher education.

Social clustering reinforces external influence on the majority opinion model

Public opinion is subject to peer interaction via social networks and external pressure from the media, advertising, and other actors. In this paper, we study the interaction between external and peer influence on the stochastic opinion dynamics of a majority vote model. We introduce a model where agents update their opinions based on the combined influence of their local neighbourhood (peers) and an external actor in the transition rates. In the first model, the external influence is only felt by agents non-aligned with the external actor (“push strategy”). In the second model, agents are affected by external influence, independently of their opinions (“nudging strategy”). In both cases, the external influence increases the possible macroscopic outcomes. These outcomes are determined by the chosen influence strategy. We also find that the social network structure affects the opinion dynamics, with social clustering positively reinforcing the external influence whereas degree heterogeneity weakens the external forces. These findings are relevant to businesses and policy making, helping to understand how groups of individuals collectively react to external actors.

Preschool children’s physical activity in the home, childcare and neighbourhood environment: A latent profile analysis using device-based measures

ABSTRACT Based on the socioecological conceptual model, the physical environment within the home, childcare and neighbourhood domains are key factors that influence preschool children’s physical activity; however, the relative importance of each of these domains for preschool children’s physical activity is unclear. We explored the physical environment characteristics within three latent profiles of 115 preschool children aged 2–5 years based on where they accumulated moderate-to-vigorous physical activity (MVPA) across five GPS-derived environmental domains. The three profiles were “Active at home” (n = 41), “Active except close to home” (n = 61), and “Active except in local neighbourhood” (n = 13). Compared to other profiles, “Active at home” had fewer parks and playgrounds within their 500–1600 m neighbourhood. Findings suggest preschool children’s MVPA profiles are reflections of their physical environmental opportunities.

Receptive Field Space for Point Cloud Analysis.

Similar to convolutional neural networks for image processing, existing analysis methods for 3D point clouds often require the designation of a local neighborhood to describe the local features of the point cloud. This local neighborhood is typically manually specified, which makes it impossible for the network to dynamically adjust the receptive field's range. If the range is too large, it tends to overlook local details, and if it is too small, it cannot establish global dependencies. To address this issue, we introduce in this paper a new concept: receptive field space (RFS). With a minor computational cost, we extract features from multiple consecutive receptive field ranges to form this new receptive field space. On this basis, we further propose a receptive field space attention mechanism, enabling the network to adaptively select the most effective receptive field range from RFS, thus equipping the network with the ability to adjust granularity adaptively. Our approach achieved state-of-the-art performance in both point cloud classification, with an overall accuracy (OA) of 94.2%, and part segmentation, achieving an mIoU of 86.0%, demonstrating the effectiveness of our method.

Impact of the magnetic horizon on the interpretation of the Pierre Auger Observatory spectrum and composition data

The flux of ultra-high energy cosmic rays reaching Earth above the ankle energy (5 EeV) can be described as a mixture of nuclei injected by extragalactic sources with very hard spectra and a low rigidity cutoff.Extragalactic magnetic fields existing between the Earth and the closest sources can affect the observed CR spectrum by reducing the flux of low-rigidity particles reaching Earth. We perform a combined fit of the spectrum and distributions of depth of shower maximum measured with the Pierre Auger Observatory including the effect of this magnetic horizon in the propagation of UHECRs in the intergalactic space.We find that, within a specific range of the various experimental and phenomenological systematics, the magnetic horizon effect can be relevant for turbulent magnetic field strengths in the local neighbourhood in which the closest sources lieof order Brms ≃ (50–100) nG (20 Mpc/ds)( 100 kpc/Lcoh)1/2, with ds the typical intersource separation and Lcoh the magnetic field coherence length. When this is the case,the inferred slope of the source spectrum becomes softer and can be closer to the expectations of diffusive shock acceleration, i.e., ∝ E-2.An additional cosmic-ray population with higher source density and softer spectra, presumably also extragalactic and dominating the cosmic-ray flux at EeV energies, is also required to reproduce the overall spectrum and composition results for all energies down to 0.6 EeV.

Simulating the LOcal Web (SLOW)

Context. This is the second paper in a series presenting the results from a 500 h−1Mpc large constrained simulation of the local Universe (SLOW). The initial conditions for this cosmological hydro-dynamical simulation are based on peculiar velocities derived from the CosmicFlows-2 catalog. The simulation follows cooling, star formation, and the evolution of super-massive black holes. This allows one to directly predict observable properties of the intracluster medium (ICM) within galaxy clusters, including X-ray luminosity, temperatures, and the Compton-y signal. Aims. Comparing the properties of observed galaxy clusters within the local Universe with the properties of their simulated counterparts enables us to assess the effectiveness of the initial condition constraints in accurately replicating the mildly nonlinear properties of the largest, collapsed objects within the simulation. Methods. Based on the combination of several, publicly available surveys we compiled a sample of galaxy clusters within the local Universe, of which we were able to cross-identify 46 of them with an associated counterpart within the SLOW simulation. We then derived the probability of the cross identification based on mass, X-ray luminosity, temperature, and Compton-y by comparing it to a random selection. Results. Our set of 46 cross-identified local Universe clusters contains the 13 most massive clusters from the Planck SZ catalog as well as 70% of clusters with M500 larger than 2 × 1014 M⊙. Compared to previous constrained simulations of the local volume, we found in SLOW a much larger amount of replicated galaxy clusters, where their simulation-based mass prediction falls within the uncertainties of the observational mass estimates. Comparing the median observed and simulated masses of our cross-identified sample allows us to independently deduce a hydrostatic mass bias of (1 − b)≈0.87. Conclusions. The SLOW constrained simulation of the local Universe faithfully reproduces numerous fundamental characteristics of a sizable number of galaxy clusters within our local neighborhood, opening a new avenue for studying the formation and evolution of a large set of individual galaxy clusters as well as testing our understanding of physical processes governing the ICM.

Phylogenetic indices and temporal and spatial scales shape the neighborhood effect on seedling survival in a mid-mountain moist evergreen broad-leaved forest, Gaoligong Mountains, Southwestern China.

Density dependence and habitat filtering have been proposed to aid in understanding community assembly and species coexistence. Phylogenetic relatedness between neighbors was used as a proxy for assessing the degree of ecological similarity among species. There are different conclusions regarding the neighborhood effect in previous studies with different phylogenetic indices or at different spatiotemporal scales. However, the effects of density dependence, neighbor phylogenetic relatedness, and habitat filtering on seedling survival with different phylogenetic indices or at different temporal and spatial scales are poorly understood. We monitored 916 seedlings representing 56 woody plant species within a 4-ha forest dynamics plot for 4 years (from 2020 to 2023) in a subtropical mid-mountain moist evergreen broad-leaved forest in the Gaoligong Mountains, Southwestern China. Using generalized linear mixed models, we tested whether and how four phylogenetic indices: total phylogenetic distance (TOTPd), average phylogenetic distance (AVEPd), relative average phylogenetic distance (APd'), and relative nearest taxon phylogenetic distance (NTPd'), three temporals (1, 2, and 3 years), and spatial scales (1, 2, and 4 ha) affect the effect of density dependence, phylogenetic density dependence, and habitat filtering on seedling survival. We found evidence of the effect of phylogenetic density dependence in the 4-ha forest dynamics plot. The effects of density dependence, phylogenetic density dependence, and habitat filtering on seedling survival were influenced by phylogenetic indices and temporal and spatial scales. The effects of phylogenetic density dependence and habitat filtering on seedling survival were more conspicuous only at 1-year intervals, compared with those at 2- and 3-year intervals. We did not detect any effects of neighborhood or habitat factors on seedling survival at small scales (1 and 2 ha), although these effects were more evident at the largest spatial scale (4 ha). These findings highlight that the effects of local neighborhoods and habitats on seedling survival are affected by phylogenetic indices as well as temporal and spatial scales. Our study suggested that phylogenetic index APd', shortest time scale (1 year), and largest spatial scales (4 ha) were suitable for neighborhood studies in a mid-mountain moist evergreen broad-leaved forest in Gaoligong Mountains. Phylogenetic indices and spatiotemporal scales have important impacts on the results of the neighborhood studies.

LPOT: Locality-Preserving Gromov-Wasserstein Discrepancy for Nonrigid Point Set Registration.

The main problems in point registration involve recovering correspondences and estimating transformations, especially in a fully unsupervised way without any feature descriptors. In this work, we propose a robust point matching method using discrete optimal transport (OT), which is a natural and useful approach for assignment tasks, to recover the underlying correspondences and improve the nonrigid registration in the presence of unknown global transformations. Specifically, we cast the registration problem as a joint estimation over local transport couplings and global transformations, observing that the local neighborhood topology structures should be preserved strongly and stably for nonrigid transformations. By solving the Gromov-Wasserstein discrepancy, a smooth assignment matrix from one point set to another can be recovered in a fully unsupervised way. Registration performance can be improved by applying an unsupervised map to guide the transformation estimate under the alternating optimization. Experimental results on several datasets reveal how the presented method is superior to the state-of-the-art methods when facing large data degradations.

Significant effect of image contrast enhancement on weld defect detection

Weld defect inspection is an essential aspect of testing in industries field. From a human viewpoint, a manual inspection can make appropriate justification more difficult and lead to incorrect identification during weld defect detection. Weld defect inspection uses X-radiography testing, which is now mostly outdated. Recently, numerous researchers have utilized X-radiography digital images to inspect the defect. As a result, for error-free inspection, an autonomous weld detection and classification system are required. One of the most difficult issues in the field of image processing, particularly for enhancing image quality, is the issue of contrast variation and luminosity. Enhancement is carried out by adjusting the brightness of the dark or bright intensity to boost segmentation performance and image quality. To equalize contrast variation and luminosity, many different approaches have recently been put forth. In this research, a novel approach called Hybrid Statistical Enhancement (HSE), which is based on a direct strategy using statistical data, is proposed. The HSE method divided each pixel into three groups, the foreground, border, and problematic region, using the mean and standard deviation of a global and local neighborhood (luminosity and contrast). To illustrate the impact of the HSE method on the segmentation or detection stage, the datasets, specifically the weld defect image, were used. Bernsen and Otsu’s methods are the two segmentation techniques utilized. The findings from the objective and visual elements demonstrated that the HSE approach might automatically improve segmentation output while effectively enhancing contrast variation and normalizing luminosity. In comparison to the Homomorphic Filter (HF) and Difference of Gaussian (DoG) approaches, the segmentation results for HSE images had the lowest result according to Misclassification Error (ME). After being applied to the HSE images during the segmentation stage, every quantitative result showed an increase. For example, accuracy increased from 64.171 to 84.964. In summary, the application of the HSE method has resulted in an effective and efficient outcome for background correction as well as improving the quality of images.

Joint Projected Fuzzy Neighborhood Preserving C-means Clustering with Local Adaptive Learning

Graph-based dimensionality reduction approaches are extensively applied for preserving the intrinsic structures of high-dimensional data. To capture discriminative information from data clusters that lack a clear manifold structure, previous studies have combined clustering algorithms with local manifold structure preservation. These methods assume that two neighbor data points share the same cluster label. However, the assumption may not hold for Fuzzy C-Means which may assign two nearby data points to different clusters depending on the distance between data points and cluster centroids. This inconsistency degrades the performance in clustering and dimensionality reduction. To tackle this issue, we propose a novel approach for dimensionality reduction, termed joint Projected Fuzzy Neighborhood Preserving C-means Clustering with Local Adaptive Learning (PFNPCM). PFNPCM learns cluster centroids, sparse membership grades, embedded subspaces, and structured graphs simultaneously and extracts both global and local structures of data. Through clustering data that preserves the local neighborhood structure in a low-dimensional space instead of the original one, PFNPCM can effectively capture both global and local information from the inherent manifold structure and cluster structure. The employment of ℓ2,1-norm as the distance metric in the clustering and the preservation of neighborhood structure further ensure the robustness of clustering results against noise or outliers. To achieve the consistency of the captured global and local discriminative information, PFNPCM adopts a novel fuzzy local similarity measure based on the manifold structure, and the manifold assumption thus holds when partitioning unlabeled samples into subgroups. Massive experiments on diverse well-known datasets demonstrate that PFNPCM is superior to some state-of-the-art methods.

The micromechanics of fracture of zirconium hydrides

Zirconium alloys are susceptible to hydrogen embrittlement and hydride precipitation. The precipitation of hydrides is accompanied by a transformation strain, but the contribution of this strain to the fracture of hydrides is not well-understood. In this study, deformation mechanisms and the micromechanics of fracture of hydrides are investigated by conducting in-situ and interrupted ex-situ tensile experiments on hydrided zirconium specimens. Electron backscatter diffraction (EBSD) is used to measure the macro-EBSD maps that contain the grains located in the gauge section of the specimens’ surfaces. Further, high spatial resolution EBSD is used to determine orientation variations induced by the precipitation of hydrides and by the external mechanical load. The measured grain orientations are mapped into a crystal plasticity finite element (CPFE) model to examine the performance of eight different crack initiation criteria. It is shown that a multiscale approach is essential for studying the fracture of hydrides as the details of hydride morphology, orientation, local grain neighborhood, and hydride-hydride interactions are important in such analysis. It is shown that neglecting the effects of hydride-induced transformation strain leads to inaccuracies in predicting both the location and direction of microcracks within hydrides. Among the examined methods, the combination of resolved shear stress and resolved shear strain on slip systems, i.e., the highest shear energy density, consistently predicts the correct locations of hydride microcracks as well as their propagation direction. Further, it is shown that the significant deformation that takes place within hydrides is the main driving force for the fracture of hydrides.

Room Quest

Abstract—Relocating to a new city can be a daunting task, especially when it comes to finding suitable accommodation. RoomQuest is a web-based platform designed to assist employees and students in this transition by providing a user-friendly interface for finding housing that meets their needs. The platform allows homeowners to register their properties and specify facilities, rules, and regulations, making it easier for users to find accommodation that aligns with their preferences. One of the key features of RoomQuest is its ability to streamline the process of searching for housing, thereby reducing the stress and uncertainty associated with relocating to unfamiliar environments. By providing a centralized platform for property listings, RoomQuest aims to simplify the housing search process and improve the overall experience for users. RoomQuest also offers a range of tools and resources to help users make informed decisions about their accommodation. This includes detailed property listings, photos, and virtual tours, as well as information about local amenities and neighborhood characteristics. Additionally, the platform allows users to communicate directly with homeowners, facilitating the booking process and ensuring a smooth transition to their new location.

Integrated hepatitis e virus monitoring in central Argentina: a six-year analysis of clinical surveillance and wastewater-based epidemiology

Wastewater-based epidemiology (WBE) has gained prominence worldwide as a powerful tool in public health. This study aimed to monitor the circulation of Hepatitis E Virus (HEV) from wastewater samples collected during a six-year period and compare these results with clinical surveillance in the central region of Argentina. From 2017 to 2022, 1008 raw wastewater samples were analyzed, including four wastewater treatment plants from four cities (n=319), and 7 local neighborhood collector sewers in Córdoba city (n=689). Serum and/or stool samples from patients suspected of HEV infection were also analyzed (n=48). HEV molecular detection and viral load quantification were performed by real time RT-qPCR, and genetic characterization by two RT-Nested PCRs (targeting partial ORF-1 and ORF-2 genomic regions), sequencing and phylogenetic analysis.Fifty-three (5.3%) wastewater samples were RNA-HEV positive by real time RT-qPCR, with variations according to the location and year (0.0% - 21.6%). Out of these, ORF-2 genomic region was amplified in 20 samples (37.7%) and ORF-1 partial region in 12 (22.6%), and eighteen sequences were obtained. Throughout the study period, two (4.2%) HEV confirmed infections were reported, and one sequence was obtained. Phylogenetic analyses for both genomic regions showed that all the isolates were genotype HEV-3 clade abchijklm.Our study detected HEV in wastewater over a six-year period, despite a low number of clinical cases, emphasizing WBE as a valuable tool that complements clinical surveillance, by detecting pathogens' presence; identifying their transmission, circulation dynamics and excretion hotspots; and revealing changes in their genomic diversity.

Learning to learn point signature for 3D shape geometry

Point signature is a representation that describes the structural geometry of a point within a neighborhood in 3D shapes. Conventional approaches apply a weight-sharing network, e.g., Graph Neural Network (GNN), to all neighborhoods of all points to directly generate point signatures and gain the generalization ability of the network by extensive training over amounts of samples from scratch. However, such approaches lack the flexibility to rapidly adapt to unseen neighborhood structures and thus cannot generalize well to new point sets. In this paper, we propose a novel meta-learning 3D point signature model, 3D meta point signature (MEPS) network, which is capable of learning robust 3D point signatures. Regarding each point signature learning process as a task, our method obtains an optimized model over the best performance on the distribution of all tasks, generating reliable signatures for new tasks, i.e., signatures of unseen point neighborhoods. Specifically, our MEPS consists of two modules: a base signature learner and a meta signature learner. During training, a base-learner is trained to perform specific signature learning tasks. Meanwhile, a meta-learner is trained to update the base-learner with optimal parameters. During testing, the meta-learner learned with the distribution of all tasks can adaptively change the base-learner parameters to accommodate unseen local neighborhoods. We evaluate our MEPS model on 3D shape correspondence and segmentation. Experimental results demonstrate that our method not only gains significant improvements over the baseline model to achieve state-of-the-art performance, but also is capable of handling unseen 3D geometry. Our implementation is available at https://github.com/hhuang-code/MEPS.

SCGAFusion: A skip-connecting group convolutional attention network for infrared and visible image fusion

The goal of infrared and visible image fusion is to synthesize an image that contains complementary information as much as possible in paired infrared and visible images. However, due to using the convolution operation with local receptive field to extract features, most of the existing deep learning-based fusion methods fail to effectively exploit global contextual information in source images and so obtain a limited fusion performance. To address this issue, this paper proposes an end-to-end skip-connecting group convolutional attention network, termed as SCGAFusion, to fuse infrared and visible images. In SCGAFusion, a group convolutional attention block (GAB) is purposefully developed to promote the extraction and utilization capability of the fusion network for informative features. GAB introduces group convolutional layer to obtain hierarchical features, and employs a residual non-local attention module to capture long-range dependencies between pixels. In this way, it can not only focus more on the important regions or details, but also leverage both local neighborhood and global contextual information. Additionally, a hierarchical feature compensation mechanism based on skip connections is devised to integratedly exploit both local and global features. Experimental results on several public datasets qualitatively and quantitatively demonstrate the advantage of SCGAFusion over other state-of-the-art fusion methods.

Urban density, household overcrowding and the spread of COVID-19 in Australian cities

The UN-Habitat World Cities Report 2020 highlighted that overcrowded housing, not urban density, is the major contributing factor to the spread of COVID-19. The relatively successful ability of densely populated cities such as Seoul, Singapore, Tokyo and New York City to manage virus spread supports this. We hypothesise that, given the complexity of the interaction between people and place, the relative contribution of density and crowding to the spread of infectious diseases may be contingent on local factors. To directly compare the role of urban density and household overcrowding, we examine each in relation to COVID-19 incidence in the three largest cities in Australia, Sydney, Melbourne and Brisbane, as the pandemic unfolded from July 2021 to January 2022. Using ecological models adjusted for spatial autocorrelation and area-level measures of age and socio-economic factors, we assess the association between population density, overcrowding in homes, and COVID-19 infections in local neighbourhoods. Challenging prevailing assumptions, we find evidence for an effect of both density and overcrowding on COVID-19 infections depending on the city and area within cities; that is, depending on the local context. For example, in the southwestern suburbs of Sydney, the case rate decreases by between 0.4 and 6.4 with every one-unit increase in gross density however the case rate increases by between 0.01 and 9.6 with every one-unit increase in total overcrowding. These findings have important implications for developing pandemic response strategies: public health measures that target either density (e.g., lockdowns and restricted range of travel) or overcrowding (e.g., restricting number of people relative to dwelling, mask-wearing indoors, vaccination prioritisation) must be cognisant of the geographically local contexts in which they are implemented.

PointNet-based modeling of systematic distance deviations for improved TLS accuracy

Abstract Terrestrial laser scanners (TLSs) have become indispensable for acquiring highly detailed and accurate 3D representations of the physical world. However, the acquired data is subject to systematic deviations in distance measurements due to external influences, such as distance and incidence angle. This research introduces a calibration approach by applying a deep learning model based on PointNet to predict and correct these systematic distance deviations, incorporating not only the XYZ coordinates but also additional features like intensity, incidence angle, and distances within a local neighbourhood radius of 5 cm. By predicting and subsequently correcting systematic distance deviations, the quality of TLS point clouds can be improved. Hence, our model is designed to complement and build upon the foundation of prior internal TLS calibration. A data set collected under controlled environmental conditions, containing various objects of different materials, served as the basis for training and validation the PointNet based model. In addition our analysis showcase the model’s capability to accurately model systematic distance deviations, outperforming existing methods like gradient boosting trees by capturing the spatial relationships and dependencies within the data more effectively. By defining test data sets, excluded from the training process, we underscore the ongoing effectiveness of our model’s distance measurement calibration, showcasing its ability to improve the accuracy of the TLS point cloud.

A deep autoencoder network connected to geographical random forest for spatially aware geochemical anomaly detection

Machine learning (ML) and deep learning (DL) techniques have recently shown encouraging performance in recognizing metal-vectoring geochemical anomalies within complex Earth systems. However, the generalization of these techniques to detect subtle anomalies may be precluded due to overlooking non-stationary spatial structures and intra-pattern local dependencies contained in geochemical exploration data. Motivated by this, we conceptualize in this paper an innovative algorithm connecting a DL architecture to a spatial ML processor to account for local neighborhood information and spatial non-stationarities in support of spatially aware anomaly detection. A deep autoencoder network (DAN) is trained to abstract deep feature codings (DFCs) of multi-element input data. The encoded DFCs represent the typical performance of a nonlinear Earth system, i.e., multi-element signatures of geochemical background populations developed by different geo-processes. A local version of the random forest algorithm, geographical random forest (GRF), is then connected to the input and code layers of the DAN processor to establish nonlinear and spatially aware regressions between original geochemical signals (dependent variables) and DFCs (independent variables). After contributions of the latter on the former are determined, residuals of GRF regressions are quantified and interpreted as spatially aware anomaly scores related to mineralization. The proposed algorithm (i.e., DAN‒GRF) is implemented in the R language environment and examined in a case study with stream sediment geochemical data pertaining to the Takht-e-Soleyman district, Iran. The high-scored anomalies mapped by DAN‒GRF, compared to those by the stand-alone DAN technique, indicated a stronger spatial correlation with locations of known metal occurrences, which was statistically confirmed by success-rate curves, Student's t‒statistic method, and prediction-area plots. The findings suggested that the proposed algorithm has an enhanced capability to recognize subtle multi-element geochemical anomalies and extract reliable insights into metal exploration targeting.

Intuitionistic fuzzy twin proximal SVM with fuzzy hyperplane and its application in EEG signal classification

Twin support vector machine (TSVM) is a contemporary machine learning technique to tackle classification and regression problems. However, TSVM lacks the ability to differentiate between support vectors and noises since it neglects the positional information of input data samples, and hence it is sensitive to noises. Additionally, it fails to consider uncertainties associated with the data, which reduces its capacity for generalization. To address these drawbacks, we propose a novel fuzzy hyperplane based intuitionistic fuzzy twin proximal support vector machine. The first significant feature of the proposed approach is that it gives an intuitionistic fuzzy number based on the relevance to each data vector. This efficiently reduces the impact of noise and outliers by leveraging the local neighborhood information within the data points by incorporating membership and non-membership weights. Secondly, all the parameters present in the model are fuzzy variables, including the offset term and the elements of the normal vector. The suggested fuzzy hyperplane successfully reflects the inherent ambiguity prevalent in real-world categorization problems by reflecting vagueness in the input data through the use of fuzzy variables. The model’s efficiency is enhanced by solving two systems of linear equations to obtain two non-parallel classifiers rather than solving two quadratic programming problems as in standard TSVM. Utilizing non-linear kernel functions within the feature space enables the method to effectively identify complex patterns or non-linear relationships within the datasets. In order to demonstrate the effectiveness of the suggested approach, extensive computer experiments have been conducted on a set of eighteen benchmark datasets with both linear and non-linear kernels. In addition, rigorous statistical analysis, including Friedman and post-hoc Nemenyi tests, have been employed to assess the significance of the observed performance differences. Furthermore, we also performed numerical experiments utilizing linear, Gaussian, and polynomial kernels to classify electroencephalogram (EEG) signals. The outcomes of the experiment are analyzed in terms of average accuracy, processing time, and F-measure. The results demonstrate that the proposed method outperforms existing methods and achieves better generalization.