Outlier Removal Research Articles

Persistence-based clustering with outlier-removing filtration

This article describes a non-parametric clustering algorithm with an outlier removal step. Our method is based on tools from topological data analysis: we define a new filtration on metric spaces which is a variant of the Vietoris–Rips filtration that adds information about the points' nearest neighbor to the persistence diagram. We prove a stability theorem for this filtration, and evaluate our method on point cloud and graph datasets, showing that it can compete with state-of-the-art methods while being non-parametric.

Machine learning based analysis and detection of trend outliers for electromyographic neuromuscular monitoring.

Neuromuscular monitoring is frequently plagued by artefacts, which along with the frequent unawareness of the principles of this subtype of monitoring by many clinicians, tends to lead to a cynical attitute by clinicians towards these monitors. As such, the present study aims to derive a feature set and evaluate its discriminative performance for the purpose of Train-of-Four Ratio (TOF-R) outlier analysis during continuous intraoperative EMG-based neuromuscular monitoring. Patient data was sourced from two devices: (1) Datex-Ohmeda Electromyography (EMG) E-NMT: a dataset derived from a prospective observational trial including 136 patients (21,891 TOF-R observations), further subdivided in two based on the type of features included; and (2) TetraGraph: a clinical case repository dataset of 388 patients (97,838 TOF-R observations). The two datasets were combined to create a synthetic set, which included shared features across the two. This process led to the training of four distinct models. The models showed an adequate bias/variance balance, suggesting no overfitting or underfitting. Models 1 and 2 consistently outperformed the others, with the former achieving an F1 score of 0.41 (0.31, 0.50) and an average precision score (95% CI) of 0.48 (0.35, 0.60). A random forest model analysis indicated that engineered TOF-R features were proportionally more influential in model performance than basic features. Engineered TOF-R trend features and the resulting Cost-Sensitive Logistic Regression (CSLR) models provide useful insights and serve as a potential first step towards the automated removal of outliers for neuromuscular monitoring devices. NCT04518761 (clinicaltrials.gov), registered on 19 August 2020.

Enhancing Stellar Temperature Estimation through Machine Learning and Multifaceted Data Exploration

This paper employs machine learning to estimate stellar temperatures using photometric data, focusing on the GAIA ESA Archive Data Release 3 data set. The study underscores the effectiveness of neural networks in deciphering intricate relationships within the data. Notably, the addition of metallicity improves model accuracy in characterizing stellar properties. The study also investigates outlier removal techniques, specifically favoring the Isolation Forest method, showcasing its efficacy in refining model performance. Automated machine learning, facilitated by PyCaret Regressor, emerges as a valuable tool, identifying top-performing models and highlighting feature importance. The implications of this research extend beyond the specifics of stellar temperature estimation. In contemplating future directions, this study suggests the exploration of diverse data sources to ensure balanced distributions of stellar temperatures and the potential incorporation of deep learning architectures for heightened accuracy in addressing astrophysical inquiries.

A novel outlier detection method based on Bayesian change point analysis and Hampel identifier for GNSS coordinate time series

The identification and removal of outliers in time series are important problems in numerous fields. In this paper, a novel method (BCP-HI) is proposed to enhance the accuracy of outlier detection in GNSS coordinate time series by combining Bayesian change point (BCP) analysis and the Hampel identifier (HI). By using BCP, change points (cps) in the time series are lidentified, and so the time series is divided into subsegments that have properties of a normal distribution. In each of these separated segments, outliers are detected using HI. Each data element identified as an outlier is corrected by a median filter of window size (w) to obtain the corrected signal. The BCP-HI method was tested on both simulated and real GNSS coordinate time series. Outliers from three different synthetic test datasets with different sampling frequencies and outlier amplitudes were detected with approximately 98% accuracy after processing. After this process, Signal-to-Noise Ratio (SNR) increased from 0.0084 to 10.8714 dB and Root Mean Square (RMS) decreased from 24 to 23 mm. Similarly, for real GNSS data, approximately 98% accuracy was achieved, with an increase in SNR from 0.0003 to 4.4082 dB and a decrease in RMS from 7.6 to 6.6 mm observed. In addition, the output signals after BCP-HI were examined graphically using Lomb–Scargle periodograms and it was observed that clearer power spectrum distributions emerged. When the input and output signals were examined using the Kolmogorov–Smirnov (KS) test, they were found to be statistically similar. These results indicate that the BCP-HI algorithm effectively removes outliers, and enhances processing accuracy and reliability, and improves signal quality.

Emulating the interstellar medium chemistry with neural operators

Context. The study of galaxy formation and evolution critically depends on our understanding of the complex photo-chemical processes that govern the evolution and thermodynamics of the interstellar medium (ISM). In a computational sense, resolving the chemistry is among the weightiest tasks in cosmological and astrophysical simulations. Aims. Astrophysical simulations can include photo-chemical models that allow for a wide range of densities (n), abundances of different species (ni/n) and temperature (T), and plausible evolution scenarios of the ISM under the action of a radiation field (F) with different spectral shapes and intensities. The evolution of such a non-equilibrium photo-chemical network relies on implicit, precise, computationally costly, ordinary differential equations (ODE) solvers. Here, we aim to substitute such procedural solvers with fast, pre-trained emulators based on neural operators. Methods. We emulated a non-equilibrium chemical network up to H2 formation (9 species, 52 reactions) by adopting the DeepONet formalism, namely: by splitting the ODE solver operator that maps the initial conditions and time evolution into a tensor product of two neural networks (named branch and trunk). We used KROME to generate a training set, spanning −2 < log(n/cm−3) ≤ 3.5, log(20) ≤ log(T/K) ≤ 5.5, −6 ≤ log(ni/n) < 0, and adopting an incident radiation field, F, sampled in 10 energy bins with a continuity prior. We separately trained the solver for T and each ni for ≃4.34 GPUhrs. Results. Compared with the reference solutions obtained by KROME for single-zone models, the typical precision obtained is of the order of 10−2, that is, it is 10 times better when using a training that is 40 times less costly, with respect to previous emulators that only considered a fixed F. DeepONet also performs well for T and ni outside the range of the training sample. Furthermore, the emulator aptly reproduces the ion and temperature profiles of photo dissociation regions as well; namely, by giving errors that are comparable to the typical difference between various photo-ionization codes. The present model achieves a speed-up of a factor of 128× with respect to stiff ODE solvers. Conclusions. Our neural emulator represents a significant leap forward in the modelling of ISM chemistry, offering a good balance of precision, versatility, and computational efficiency. Nevertheless, further work is required to address the challenges represented by the extrapolation beyond the training time domain and the removal of potential outliers.

Digital Twin Implementation for Three-Dimensional Rotordynamic Response via Physics-Informed LSTM Neural Networks

Abstract The rotating assemblies of critical machinery are complex dynamical systems and rotordynamic model response prediction inaccuracy risks machinery failure leading to high production losses. Jeffcott, Euler beam, and high-fidelity 3D solid finite element models are frequently utilized for rotordynamic analyses. Even though the 3D rotor has the higher accuracy, beam models are still widely used in industrial applications. To improve prediction accuracy of the lower-fidelity Jeffcott and beam models, a rotordynamics physics-informed neural network (R-PINN) is proposed. This models physics-informed long short-term memory (LSTM) neural networks that utilize partial or limited measured data, by incorporating physical laws. This approach enables the creation of a Digital Twin, which can produce additional data and help remove noise and outliers. In the current study, two R-PINNs are introduced to validate the superior capability of the model for both low- and high-fidelity physics. Random noise of 10% is introduced into the measured data produced by the Digital Twin to replicate real-world noisy measurements. The result shows that both low- and high-fidelity physics R-PINNs can achieve high accuracy even with high noise data, thereby increasing the robustness of the model. The results clearly demonstrate the ability of the proposed R-PINN algorithm to enhance an Euler beam model's predicted response to the level of accuracy of a 3D solid element model's predicted response, the latter acting as a surrogate for test measurements in an actual application.

A 3D reconstruction-based method using unmanned aerial vehicles for the representation and analysis of road sections

Due to the fast growth of cities worldwide, roads are increasing daily, and pavement maintenance has become very heavy and costly. Despite all efforts made under the pavement management system to keep the road surface in good shape, several road sections need to be in better condition, which presents a danger for drivers and pedestrians. This paper proposes a novel pavement 3D reconstruction and segmentation approach using the structure from motion technique, unmanned aerial vehicle, and digital camera. The method consists of the 3D modeling of the road by using images taken from different perspectives and the structure from motion technique. In this method, points cloud is sampled and cleaned using statistical outlier removal and noise filters. After that, duplicated and isolated points are eliminated to retain only significant data. The normal road plane is estimated using the principal component analysis technique and the remaining points. This plan presents a root mean square less than 0.85 cm. Finally, distances from those points to the normal plane are calculated and clustered to segment the road into distressed and non-distressed areas. The proposed approach presents a similarity rate to the survey measurement passed 95%. It has demonstrated promising results and has the potential for further improvement by optimizing various steps.

FOOR: Be Careful for Outlier-Score Outliers When Using Unsupervised Outlier Ensembles

Outlier detection is a very important tool in analyzing patterns and detecting unexpected events in social systems. However, the process of outlier detection could be fraught with uncertainty, with difficulties in determining the veracity of an object’s outlier score. We propose a framework for outlier-score outlier removal (FOOR). FOOR is a selection method, which aims to remove inaccurate outlier scores prior to data processing by ensemble techniques, to improve the accuracy of all ensembles. FOOR has rigorously tested with 30 real-world datasets and seven state-of-the-art ensembles over 25 different base detectors. Simulated experiments showed that FOOR significantly improves the existing techniques, with an average (AVG) of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$+$</tex-math> </inline-formula> 0.05 AUC (from 0.81 to 0.86 AUC). Thus, we recommend FOOR as the new standard for outlier-score preprocessing before ensembles.

Utilizing Nearest-Neighbor Clustering for Addressing Imbalanced Datasets in Bioengineering.

Imbalance classification is common in scenarios like fault diagnosis, intrusion detection, and medical diagnosis, where obtaining abnormal data is difficult. This article addresses a one-class problem, implementing and refining the One-Class Nearest-Neighbor (OCNN) algorithm. The original inter-quartile range mechanism is replaced with the K-means with outlier removal (KMOR) algorithm for efficient outlier identification in the target class. Parameters are optimized by treating these outliers as non-target-class samples. A new algorithm, the Location-based Nearest-Neighbor (LBNN) algorithm, clusters one-class training data using KMOR and calculates the farthest distance and percentile for each test data point to determine if it belongs to the target class. Experiments cover parameter studies, validation on eight standard imbalanced datasets from KEEL, and three applications on real medical imbalanced datasets. Results show superior performance in precision, recall, and G-means compared to traditional classification models, making it effective for handling imbalanced data challenges.

Assessment of the Improvement in Observation Precision of GNSS, SLR, VLBI, and DORIS Inputs from ITRF2014 to ITRF2020 Using TRF Stacking Methods

International terrestrial reference frame (ITRF) input data, generated by Global Navigation Satellite Systems (GNSS), Satellite Laser Ranging (SLR), Very Long Baseline Interferometry (VLBI), and Doppler Orbitography and Radiopositioning integrated by satellite (DORIS) combination centers (CCs), are considered to be relatively high-quality and accurate solutions. Every few years, these input data are submitted to the three ITRS combination centers, namely Institut Géographique National (IGN), Deutsches Geodätisches Forschungsinstitut at the Technische Universität München (DGFI-TUM), and Jet Propulsion Laboratory (JPL), to establish a multi-technique combined terrestrial reference frame (TRF). Generally, these solutions have undergone three rounds of outlier removal: the first at the technique analysis centers during solution generations and the second during the technique-specific combination by the CCs; ITRS CCs then perform a third round of outlier removal and preprocessing during the multi-technique combination of TRFs. However, since the primary objective of CCs is to release the final TRF product, they do not emphasize the publication of analytical preprocessing results, such as the outlier rejection rate. In this paper, our specific focus is on assessing the precision improvement of ITRF input data from 2014 to 2020, which includes evaluating the accuracy of coordinates, the datum accuracy, and the precision of the polar motions, for all four techniques. To achieve the above-mentioned objectives, we independently propose a TRF stacking approach to establish single technical reference frameworks, using software developed by us that is different from the ITRF generation. As a result, roughly 0.5% or less of the SLR observations are identified as outliers, while the ratio of DORIS, GNSS, and VLBI observations are below 1%, around 2%, and ranging from 1% to 1.2%, respectively. It is shown that the consistency between the SLR scale and ITRF has improved, increasing from around −5 mm in ITRF2014 datasets to approximately −1 mm in ITRF2020 datasets. The scale velocity derived from fitting the VLBI scale parameter series with all epochs in ITRF2020 datasets differs by approximately 0.21 mm/year from the velocity obtained by fitting the data up to 2013.75 because of the scale drift of VLBI around 2013. The decreasing standard deviations of the polar motion parameter (XPO, YPO) offsets between Stacking TRFs and 14C04 (20C04) indicate an improvement in the precision of polar motion observations for all four techniques. From the perspective of the weighted root mean square (WRMS) in station coordinates, since the inception of the technique, the station coordinate WRMS of DORIS decreased from 30 mm to 5 mm for X and Y components, and 25 mm to 5 mm for the Z component; SLR WRMS decreased from 20 mm to better than 10 mm (X, Y and Z); GNSS WRMS decreased from 4 mm to 1.5 mm (X and Y) and 5 mm to 2 mm (Z); while VLBI showed no significant change.

SWOT Level 2 Lake Single-Pass Product: The L2_HR_LakeSP Data Preliminary Analysis for Water Level Monitoring

The Surface Water and Ocean Topography (SWOT) mission, launched in December 2022, aims to address the crucial environmental goal of water monitoring to support preparedness for extreme events and facilitate adaptation to climate change on global and local scales. This mission will provide a comprehensive inventory of worldwide water resources, lakes, reservoir storage, and river dynamics. In this work, we carried out a preliminary assessment of SWOT’s Lake product Level 2 version 1.1, also known as “L2_HR_LakeSP”. The analysis was performed across six diverse lakes on three continents, revealing an average median bias of 0.08 m with respect to the considered reference, after suitable outlier removal. An overall precision of 0.22 m was found, combined with an average correlation of 68% between SWOT and reference time series. Moreover, the accuracy varied in the considered six lakes, since biases up to some decimeters were found for some of them; they could be due to residual inconsistencies between the vertical reference frame of SWOT and that of the considered reference. In summary, the first analysis of the “L2_HR_LakeSP” product, Version 1.1, demonstrated the promising potential of SWOT for monitoring seasonal variations in water levels. Nevertheless, notable anomalies were found in the water masks, particularly in higher latitudes, suggesting potential difficulties in accurately delineating water bodies in those regions. Additionally, a discernible reduction in accuracy was observed towards the end of the monitoring period. These preliminary findings indicate some issues that should be addressed in future investigations about the quality and potential of SWOT’s lake products for advancing our understanding of global water dynamics.

Strategies for Imputing Missing Values and Removing Outliers in the Dataset for Machine Learning-Based Construction Cost Prediction

Accurately predicting construction costs during the initial planning stages is crucial for the successful completion of construction projects. Recent advancements have introduced various machine learning-based methods to enhance cost estimation precision. However, the accumulation of authentic construction cost data is not straightforward, and existing datasets frequently exhibit a notable presence of missing values, posing challenges to precise cost predictions. This study aims to analyze diverse substitution methods for addressing missing values in construction cost data. Additionally, it seeks to evaluate the performance of machine learning models in cost prediction through the removal of conditional outliers. The primary goal is to identify and propose optimal strategies for handling missing value in construction cost records, ultimately improving the reliability of cost predictions. According to the analysis results, among single imputation methods, median imputation emerges as the most suitable, while among multiple imputation methods, lasso regression imputation produces the most superior outcomes. This research contributes to enhancing the trustworthiness of construction cost predictions by presenting a pragmatic approach to managing missing data in construction cost performance records, thereby facilitating more precise project planning and execution.

Predictive Neural Network Modeling for Almond Harvest Dust Control.

This study introduces a neural network-based approach to predict dust emissions, specifically PM2.5 particles, during almond harvesting in California. Using a feedforward neural network (FNN), this research predicted PM2.5 emissions by analyzing key operational parameters of an advanced almond harvester. Preprocessing steps like outlier removal and normalization were employed to refine the dataset for training. The network's architecture was designed with two hidden layers and optimized using tanh activation and MSE loss functions through the Adam algorithm, striking a balance between model complexity and predictive accuracy. The model was trained on extensive field data from an almond pickup system, including variables like brush speed, angular velocity, and harvester forward speed. The results demonstrate a notable predictive accuracy of the FNN model, with a mean squared error (MSE) of 0.02 and a mean absolute error (MAE) of 0.01, indicating high precision in forecasting PM2.5 levels. By integrating machine learning with agricultural practices, this research provides a significant tool for environmental management in almond production, offering a method to reduce harmful emissions while maintaining operational efficiency. This model presents a solution for the almond industry and sets a precedent for applying predictive analytics in sustainable agriculture.

Three-dimensional reconstruction of aerosols based on spectral radiance matching

This paper evaluates a three-dimensional scene construction algorithm that assigns CALIPSO-observed profiles to receiving pixels in MODIS imagery through spectral radiance matching (SRM), so that a limited number of vertically resolved aerosol observations can be expanded into a regionally continuous three-dimensional aerosol structure. The SRM algorithm was tested using moderate resolution imaging spectro-radiometer (MODIS) and cloud-aerosol lidar and infrared pathfinder satellite observation (CALIPSO) data for the period of April 10–25, 2015. First, the spectral radiance channel and the surface reflectance channel were selected for combination tests to select the best channel combination 01-04-07-26 for the application of SRM algorithm in the field of aerosol three-dimensional reconstruction. The average aerosol matching rate for this channel combination can reach 90.57%, 81.64%, 78.6%, 74.12%, and 65.96% at extended distances of 5, 20, 30, 50, and 100 km, respectively. In addition, the combination of the average correct match rate and the reconstructed aerosol structure revealed that the most suitable extension distance is up to 100 km. Then, for the reconstruction profile validation, lidar sites under four weather conditions (clean sky, aerosol loading, thin clouds, and thick clouds) were selected for validation. The comparison results reveal that the average relative errors are about 30%, 60%, 30%–60%, and 70% for these four conditions, respectively. Overall, the average relative error of the reconstructed profiles within 100 km of the orbital track of CALIPSO (except under thick cloud conditions) is less than 60%, which is within a reasonable error envelop. There are also some limitations that can be improved later by adding more constraints: 1) it is difficult to reconstruct some short-time, small-scale aerosol or cloud features; 2) it is difficult to reconstruct some hierarchical structures under thick clouds due to signal attenuation caused by thick clouds; 3) there is an underestimation of the attenuated backscatter coefficient under clouds. Finally, for aerosol optical depth (AOD), the SRM algorithm largely underestimates AOD. The reconstructed AOD is in good agreement with aerosol robotic network (AERONET), with R2=0.8231 after removal of outliers.

PointCVaR: Risk-Optimized Outlier Removal for Robust 3D Point Cloud Classification

With the growth of 3D sensing technology, the deep learning system for 3D point clouds has become increasingly important, especially in applications such as autonomous vehicles where safety is a primary concern. However, there are growing concerns about the reliability of these systems when they encounter noisy point clouds, either occurring naturally or introduced with malicious intent. This paper highlights the challenges of point cloud classification posed by various forms of noise, from simple background noise to malicious adversarial/backdoor attacks that can intentionally skew model predictions. While there's an urgent need for optimized point cloud denoising, current point outlier removal approaches, an essential step for denoising, rely heavily on handcrafted strategies and are not adapted for higher-level tasks, such as classification. To address this issue, we introduce an innovative point outlier cleansing method that harnesses the power of downstream classification models. Using gradient-based attribution analysis, we define a novel concept: point risk. Drawing inspiration from tail risk minimization in finance, we recast the outlier removal process as an optimization problem, named PointCVaR. Extensive experiments show that our proposed technique not only robustly filters diverse point cloud outliers but also consistently and significantly enhances existing robust methods for point cloud classification. A notable feature of our approach is its effectiveness in defending against the latest threat of backdoor attacks in point clouds.

Graph Context Transformation Learning for Progressive Correspondence Pruning

Most of existing correspondence pruning methods only concentrate on gathering the context information as much as possible while neglecting effective ways to utilize such information. In order to tackle this dilemma, in this paper we propose Graph Context Transformation Network (GCT-Net) enhancing context information to conduct consensus guidance for progressive correspondence pruning. Specifically, we design the Graph Context Enhance Transformer which first generates the graph network and then transforms it into multi-branch graph contexts. Moreover, it employs self-attention and cross-attention to magnify characteristics of each graph context for emphasizing the unique as well as shared essential information. To further apply the recalibrated graph contexts to the global domain, we propose the Graph Context Guidance Transformer. This module adopts a confident-based sampling strategy to temporarily screen high-confidence vertices for guiding accurate classification by searching global consensus between screened vertices and remaining ones. The extensive experimental results on outlier removal and relative pose estimation clearly demonstrate the superior performance of GCT-Net compared to state-of-the-art methods across outdoor and indoor datasets.

Influence of VF and SOR-Filtering Methods on Tree Height Inversion Using Unmanned Aerial Vehicle LiDAR Data

Forests, as the main body of the terrestrial ecosystem, have long been focal points for accurate structural parameter extraction. Among these parameters, tree height is a fundamental measurement factor that plays an important role in monitoring forest structure and biomass. The emergence of unmanned aerial vehicle light detection and ranging (UAV-LiDAR) technology has provided a strong guarantee of the acquisition of forest tree height parameters. However, UAV-LiDAR point cloud data have problems such as a large volume and data redundancy, and different point cloud data processing methods have different effects. Based on voxel filtering (VF) and statistical outlier removal (SOR)point cloud data processing experimental analysis, this study explored the influence of different filtering methods on the forest tree height inversion efficiency and accuracy. First, the point cloud data processed by VF is significantly better than that of SOR in terms of point cloud number, file size, running time, etc. The number of point clouds for VF decreased by an average of 96.91% compared with the original point clouds. Second, the VF tree height inversion accuracy was better than the tree height inversion data using SOR. The average accuracy of VF was 96.24%, while that of SOR was 94.17%. In summary, VF can effectively reduce data redundancy and improve tree height inversion accuracy.

Enhancing Breast Cancer Classification: A Deep Learning Approach with Outlier Detection With Egret Swarm Optimization Algorithm and Feature Selection Integration – A Systematic Review

Cancer, the second leading cause of death worldwide, encompasses a spectrum of diseases characterized by uncontrolled cell proliferation. Effective therapy relies on screening, early diagnosis, and recurrence prediction. Timely identification of lesions is critical for improving patient survival rates, and accurate classification of benign cancers can prevent unnecessary treatment. Breast cancer (BC) classification into benign or malignant categories is a subject of extensive review. This review discusses a novel approach using deep neural networks (DNN) and genetic algorithms (GA) based on Egret Swarm Optimization (ESO) to address this challenge. Initially, an Enhanced Linear Discriminant Analysis (ILDA) technique is applied for data dimensionality reduction. Subsequently, DNN controls outlier detection to identify anomalies in the cancer dataset. GA is then employed for feature selection, efficiently locating optimal features in complex search spaces, crucial for cancer diagnosis models, especially non-invasive ones. Finally, ESO algorithm, integrating the aggressive and sit-and-wait tactics of Great Egret and Snowy Egret, respectively, is utilized to classify malignancy. In ths review, The proposed method is evaluated using BC datasets, with trials conducted before and after outlier removal. Experimental results demonstrate the effectiveness and accuracy of the proposed strategy. The findings of this study hold promise for enhancing BC diagnosis and guiding patient treatment decisions.

Abstract P347: Variability in Predictive Error of Waist Circumference Models: A Stratified Analysis Based on Body Mass Index, Waist Circumference, and Their Ratio

Introduction: Waist circumference (WC) independently predicts cardiovascular risks, yet clinical measurements are sporadic. Some models have been developed for WC prediction, with body mass index (BMI) as a key predictor. While these models are generally efficient, variability in error across distinct subpopulations is possible. This study aimed to assess the prediction error (PE) variability of WC models across different strata of BMI, WC, and their ratio, and to evaluate the role of BMI for WC prediction. Hypothesis: 1) PE varies across BMI, WC, and their interrelated strata. 2) BMI is a key feature for WC prediction. Methods: Data from 30,298 individuals from the National Health and Nutrition Examination Survey (2007-2018) were analyzed. After outlier removal and kNN-imputation for missing values, machine learning predictive models for WC were developed. The base model included age, gender, ethnicity, height, weight, and BMI. A non-BMI model was also formulated, incorporating demographics and clinical variables selected via a feature-selection process. Root Mean Square Error (RMSE) across BMI, WC, and ratio of BMI-decile to WC-decile categories were assessed. Results: The base model's RMSE for the entire cohort was 5.70. Stratification by BMI revealed higher error rates for obesity classes II (5.88) and III (12.28), contrasting with lower RMSEs in participants with normal BMI (3.95), overweight (4.63), and obesity class I (5.41). Individuals with high WC had an increased RMSE (6.28) compared to those with low WC (5.04). Those with a BMI-decile/WC-decile below one had elevated error (6.98) compared to those with ratios equal to (4.60) or above one (5.52). Across all categories, the base model outperformed the non-BMI model (Figure 1). Conclusion: WC model PE demonstrated variability across BMI, WC, and their respective ratios, underscoring the predictive significance of BMI for WC. Considering population-specific variations in predictive WC modeling is important within clinical and research sets.

Revolutionizing the used car market: Predicting prices with XGBoost

In contemporary dynamic automobile industry, informed decision-making is crucial for success. Accurate sales predictions are essential for optimizing operations and resource allocation in the used car market. This article explores the application of machine learning, specifically the eXtreme Gradient Boosting(XGBoost) algorithm, in predicting used car prices in the UK. The dataset consists of over 110,000 data points from various brands, including features like model, year, mileage, and more. Data preprocessing, including outlier removal and feature scaling, is performed to prepare the dataset. XGBoost, a powerful machine learning algorithm, is selected for price prediction and compared with traditional regression models. The results show that eXtreme Gradient Boosting(XGBoost) outperforms other models in terms of R2 score, Mean Absolute Error, and Mean Square Error. Its accuracy and efficiency make it a valuable tool for both buyers and sellers in the used car market. However, challenges such as data quality, feature engineering complexity, and model interpretability remain. Future research directions include improving data quality, exploring advanced feature engineering techniques, and integrating external data sources. This study highlights the potential of machine learning in enhancing decision-making in the dynamic used car market.