Probabilistic Forecasting of Electricity Demand Incorporating Mobility Data

Due to extreme weather conditions and anomalous events such as the COVID-19 pandemic, utilities and grid operators worldwide face unprecedented challenges. These unanticipated changes in trends introduce new uncertainties in conventional short-term electricity demand forecasting (EDF) since its result depends on recent usage as an input variable. In order to quantify the uncertainty of EDF effectively, this paper proposes a comprehensive probabilistic EFD method based on Gaussian process regression (GPR) and kernel density estimation (KDE). GPR is a non-parametric method based on Bayesian theory, which can handle the uncertainties in EDF using limited data. Mobility data is incorporated to manage uncertainty and pattern changes and increase forecasting model scalability. This study first performs a correlation study for feature selection that comprises weather, renewable and non-renewable energy, and mobility data. Then, different kernel functions of GPR are compared, and the optimal function is recommended for real applications. Finally, real data are used to validate the effectiveness of the proposed model and are elaborated with three scenarios. Comparison results with other conventional adopted methods show that the proposed method can achieve high forecasting accuracy with a minimum quantity of data while addressing forecasting uncertainty, thus improving decision-making.

Variations in the Spatial Distribution of Smart Parcel Lockers in the Central Metropolitan Region of Tianjin, China: A Comparative Analysis before and after COVID-19

The COVID-19 pandemic has led to a significant increase in e-commerce, which has prompted residents to shift their purchasing habits from offline to online. As a result, Smart Parcel Lockers (SPLs) have emerged as an accessible end-to-end delivery service that fits into the pandemic strategy of maintaining social distance and no-contact protocols. Although numerous studies have examined SPLs from various perspectives, few have analyzed their spatial distribution from an urban planning perspective, which could enhance the development of other disciplines in this field. To address this gap, we investigate the distribution of SPLs in Tianjin's central urban area before and after the pandemic (i.e., 2019 and 2022) using kernel density estimation, average nearest neighbor analysis, standard deviation elliptic, and geographical detector. Our results show that, in three years, the number of SPLs has increased from 51 to 479, and a majority were installed in residential communities (i.e., 92.2% in 2019, and 97.7% in 2022). We find that SPLs were distributed randomly before the pandemic, but after the pandemic, SPLs agglomerated and followed Tianjin's development pattern. We identify eight influential factors on the spatial distribution of SPLs and discuss their individual and compound effects. Our discussion highlights potential spatial distribution analysis, such as dynamic layout planning, to improve the allocation of SPLs in city planning and city logistics.

Crowd Density Estimation Using Neural Network for COVID-19 and Future Pandemics

The population's vulnerability is exacerbated by the lack of effective treatment drugs and immunity to COVID-19. The only viable strategy for combating this pandemic is social separation. In order to automate the task of monitoring social separation using surveillance footage, this study presents a neural network-based crowd density estimation for COVID-19 and future pandemics. The suggested framework employs the object identification model to distinguish persons in the scene, as well as the deep sort technique to track recognized people with issued IDs. The obtained results of the proposed work are compared in terms of loss values defined by object classification and localization, frames per second (FPS), and mean average precision (mAP). The proposed method yields good results against faster region-based convolutional neural network (RCNN) and single-shot detector (SSD). © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.

Representation Learning with Information Theory to Detect COVID-19 and Its Severity

Successful data representation is a fundamental factor in machine learning based medical imaging analysis. Deep Learning (DL) has taken an essential role in robust representation learning. However, the inability of deep models to generalize to unseen data can quickly overfit intricate patterns. Thereby, the importance of implementing strategies to aid deep models in discovering useful priors from data to learn their intrinsic properties. Our model, which we call a dual role network (DRN), uses a dependency maximization approach based on Least Squared Mutual Information (LSMI). LSMI leverages dependency measures to ensure representation invariance and local smoothness. While prior works have used information theory dependency measures like mutual information, these are known to be computationally expensive due to the density estimation step. In contrast, our proposed DRN with LSMI formulation does not require the density estimation step and can be used as an alternative to approximate mutual information. Experiments on the CT based COVID-19 Detection and COVID-19 Severity Detection Challenges of the 2nd COV19D competition [24] demonstrate the effectiveness of our method compared to the baseline method of such competition. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Machine learning approach for estimating the human-related VOC emissions in a university classroom.

Indoor air quality becomes increasingly important, partly because the COVID-19 pandemic increases the time people spend indoors. Research into the prediction of indoor volatile organic compounds (VOCs) is traditionally confined to building materials and furniture. Relatively little research focuses on estimation of human-related VOCs, which have been shown to contribute significantly to indoor air quality, especially in densely-occupied environments. This study applies a machine learning approach to accurately estimate the human-related VOC emissions in a university classroom. The time-resolved concentrations of two typical human-related (ozone-related) VOCs in the classroom over a five-day period were analyzed, i.e., 6-methyl-5-hepten-2-one (6-MHO), 4-oxopentanal (4-OPA). By comparing the results for 6-MHO concentration predicted via five machine learning approaches including the random forest regression (RFR), adaptive boosting (Adaboost), gradient boosting regression tree (GBRT), extreme gradient boosting (XGboost), and least squares support vector machine (LSSVM), we find that the LSSVM approach achieves the best performance, by using multi-feature parameters (number of occupants, ozone concentration, temperature, relative humidity) as the input. The LSSVM approach is then used to predict the 4-OPA concentration, with mean absolute percentage error (MAPE) less than 5%, indicating high accuracy. By combining the LSSVM with a kernel density estimation (KDE) method, we further establish an interval prediction model, which can provide uncertainty information and viable option for decision-makers. The machine learning approach in this study can easily incorporate the impact of various factors on VOC emission behaviors, making it especially suitable for concentration prediction and exposure assessment in realistic indoor settings.

Local Naïve Bayes for Predicting Evolution of COVID-19 Patients on Self Organizing Maps

The most recent Clinical Decision Support Systems use the potential of Machine Learning techniques to target clinical problems, avoiding the use of explicit rules. In this paper, a model to monitor and predict the risk of unfavourable evolution (UE) during hospitalization of COVID-19 patients is proposed. It combines Self Organizing Maps and local Naïve Bayes (NB) classifiers because of interpretation purposes. We used the results of six blood tests (leukocytes, D-dimer, among others) provided by a Spanish hospital group. The probabilistic approach allows us to get the daily risk of UE for each patient in an interpretable way. Several variants of the NB classifiers family have been explored, mainly weighting and likelihood estimation (parametric and nonparametric). Despite the over-simplified assumptions of the NB classifiers, they provided good predictive results in terms of sensitivity and specificity. The model with nonparametric likelihood estimation provided the best risk prediction over time even when designed with a limited number of samples. Specifically, the median value and interquartil range for the risk prediction were quite reliable even 10 days before the event day for patients hospitalized longer than 7 days. The risk median values also agree with the gold-standard for patients with a hospital stay shorter than 7 days, though the interquartil range can be too wide (probably because of the variability in the inpatient days - sometimes, just 2 days). Though a deepest analysis considering more patients and features would be convenient, our results show the potential of the proposed approach, both from a technical and clinical viewpoint. © 2022 IEEE.

Autoencoder-based Unsupervised Anomaly Detection for Covid-19 Screening on Chest X-Ray Images

The COVID-19 outbreak is a major global catastrophe of our time and the largest hurdle since World War II. According to WHO, as of July 2022, there are more than 571 million confirmed cases of COVID-19 and over six million deaths. The issue of identifying unexpected inputs based on trained examples of normal data is known as anomaly detection. In the case of diagnosing covid-19, Chest X-ray disorders that are hardly apparent are extremely challenging to identify. Although various well-known supervised classification methods are being applied for that purpose, however in the real scenario, healthy patients' data is tremendously available but contaminated samples are scarce. The process of gathering samples from ill patients is troublesome and takes a lengthy time. To address the issue of data imbalance in anomaly detection, this research demonstrates an unsupervised learning technique using a convolutional autoencoder in which the training phase does not include any infected sample. Being trained only with the healthy data, The patterns of the healthy samples are preserved in latent vector space and can differentiate ill samples by observing substantial divergence from the distribution of healthy data. Higher reconstruction error and lower KDE (Kernel Density Estimation) indicate affected data. By contrasting the reconstruction error and KDE of healthy data with anomalous data, the suggested technique is feasible for identifying anomalous samples. © 2022 IEEE.

Dilated Convolution to Capture Scale Invariant Context in Crowd Density Estimation

Crowd Density Estimation (CDE) can be used ensure safety of crowds by preventing stampedes or reducing spread of disease which was made urgent with the rise of Covid-19. CDE a challenging problem due to problems such as occlusion and massive scale variations. This research looks to create, evaluate and compare different approaches to crowd counting focusing on the ability for dilated convolution to extract scale-invariant contextual information. In this work we build and train three different model architectures: a Convolutional Neural Network (CNN) without dilation, a CNN with dilation to capture context and a CNN with an Atrous Spatial Pyramid Pooling (ASPP) layer to capture scale-invariant contextual features. We train each architecture multiple times to ensure statistical significance and evaluate them using the Mean Squared Error (MSE), Mean Average Error (MAE) and Grid Average Mean Absolute Error (GAME) on the Shang-haiTech and UCF CC 50 datasets. Comparing the results between approaches we find that applying dilated convolution to more sparse crowd images with little scale variations does not make a significant difference but, on highly congested crowd images, dilated convolutions are more resilient to occlusion and perform better. Furthermore, we find that adding an ASPP layer improves performance in the case when there are significant differences in the scale of objects within the crowds. The code for this research is available at https://github.com/ThishenP/crowd-density. © 2022, EasyChair. All rights reserved.

A Generalized Entropy Approach to Portfolio Selection under a Hidden Markov Model

This paper develops a dynamic portfolio selection model incorporating economic uncertainty for business cycles. It is assumed that the financial market at each point in time is defined by a hidden Markov model, which is characterized by the overall equity market returns and volatility. The risk associated with investment decisions is measured by the exponential Rényi entropy criterion, which summarizes the uncertainty in portfolio returns. Assuming asset returns are projected by a regime-switching regression model on the two market risk factors, we develop an entropy-based dynamic portfolio selection model constrained with the wealth surplus being greater than or equal to the shortfall over a target and the probability of shortfall being less than or equal to a specified level. In the empirical analysis, we use the select sector ETFs to test the asset pricing model and examine the portfolio performance. Weekly financial data from 31 December 1998 to 30 December 2018 is employed for the estimation of the hidden Markov model including the asset return parameters, while the out-of-sample period from 3 January 2019 to 30 April 2022 is used for portfolio performance testing. It is found that, under both the empirical Sharpe and return to entropy ratios, the dynamic portfolio under the proposed strategy is much improved in contrast with mean variance models.

A Clustering Mechanism to Identify Close Contact for the Ship Passenger Health

COVID-19 is spreading globally, and this spread is continuous. Ships have become the leading platform for virus transmission as a means of transportation. The small space of ships makes the possibility of virus outbreaks highly increased. The current way to effectively interrupt the spread of the virus is to track close contacts and physically isolate them. Therefore, the identification of close contacts becomes critical. This paper proposes a close contact identification algorithm applicable to the ship environment. The user ID is creatively proposed as the initialized location point cluster in this algorithm. And the KDE is introduced into the clustering process of the algorithm, and the center of the cluster is calculated by using the KDE of the location points as weights. The threshold value is used as the criterion for merging the clusters. Finally, the correct cluster result is obtained. This algorithm can provide technical support for ship companies to sustainably manage ships in the post-epidemic era, thus serving the purpose of maximizing the protection of ship passengers' health. © 2022 IEEE.

Tracing out the effect of transportation infrastructure on NO2 concentration levels with Kernel Density Estimation by investigating successive COVID-19-induced lockdowns.

This study aims to investigate the effect of transportation infrastructure on the decrease of NO2 air pollution during three COVID-19-induced lockdowns in a vast region of France. For this purpose, using Sentinel-5P satellite data, the relative change in tropospheric NO2 air pollution during the three lockdowns was calculated. The estimation of regional infrastructure intensity was performed using Kernel Density Estimation, being the predictor variable. By performing hotspot-coldspot analysis on the relative change in NO2 air pollution, significant spatial clusters of decreased air pollution during the three lockdowns were identified. Based on the clusters, a novel spatial index, the Clustering Index (CI) was developed using its Coldspot Clustering Index (CCI) variant as a predicted variable in the regression model between infrastructure intensity and NO2 air pollution decline. The analysis revealed that during the three lockdowns there was a strong and statistically significant relationship between the transportation infrastructure and the decline index, CCI (r = 0.899, R2 = 0.808). The results showed that the largest decrease in NO2 air pollution was recorded during the first lockdown, and in this case, there was the strongest inverse correlation with transportation infrastructure (r = -0.904, R2 = 0.818). Economic and population predictors also explained with good fit the decrease in NO2 air pollution during the first lockdown: GDP (R2 = 0.511), employees (R2 = 0.513), population density (R2 = 0.837). It is concluded that not only economic-population variables determined the reduction of near-surface air pollution but also the transportation infrastructure. Further studies are recommended to investigate other pollutant gases as predicted variables.

Advanced Pedestrian State Sensing Method for Automated Patrol Vehicle Based on Multi-Sensor Fusion.

At present, the COVID-19 pandemic still presents with outbreaks occasionally, and pedestrians in public areas are at risk of being infected by the viruses. In order to reduce the risk of cross-infection, an advanced pedestrian state sensing method for automated patrol vehicles based on multi-sensor fusion is proposed to sense pedestrian state. Firstly, the pedestrian data output by the Euclidean clustering algorithm and the YOLO V4 network are obtained, and a decision-level fusion method is adopted to improve the accuracy of pedestrian detection. Then, combined with the pedestrian detection results, we calculate the crowd density distribution based on multi-layer fusion and estimate the crowd density in the scenario according to the density distribution. In addition, once the crowd aggregates, the body temperature of the aggregated crowd is detected by a thermal infrared camera. Finally, based on the proposed method, an experiment with an automated patrol vehicle is designed to verify the accuracy and feasibility. The experimental results have shown that the mean accuracy of pedestrian detection is increased by 17.1% compared with using a single sensor. The area of crowd aggregation is divided, and the mean error of the crowd density estimation is 3.74%. The maximum error between the body temperature detection results and thermometer measurement results is less than 0.8°, and the abnormal temperature targets can be determined in the scenario, which can provide an efficient advanced pedestrian state sensing technique for the prevention and control area of an epidemic.

Geographic Information Systems and spatial distribution of COVID 19 cases in Mexico

This study analyzes the evolution of the spatial distribution in areas with a high density of infections. The information is organized and linked to a geographic database considering the political and administrative divisions by state and municipalities. Afterward, delivery metrics and spatial statistics were applied to detect distribution patterns. Since November 2020, a trend has been identified in the concentration of cases towards the central zone of Mexico. The study recognizes the decision-making of the government through the application and strict monitoring of restrictive measures like social distancing and the use of masks;a priority in regions with the most significant risk of spread. The enforcement of Geographic Information Systems for the monitoring, follow-up, prevention, and control of the pandemic makes it possible to identify and report the areas with the severest risk of contagion of the virus.

Multi Layered Deep Neural Network for Feature Extraction in Cross Domain Crowd Counting

Automated crowd density monitoring is an emerging area of research. It is a vital technology that assists during recent disease outbreaks in preserving social distancing, crowd management and other widespread applications in public security and traffic control. Modern methods to count people in crowded scenes mainly rely on Convolutional Neural Network (CNN) based models. But the model's ability to adapt for different domains which is referred to as cross domain crowd counting is a challenging task. To remedy this difficulty, many researchers used Spatial Fully Convolutional Network (SFCN) based crowd counting models with synthetic crowd scene data. They covered many image domains with few-shot learning to reduce the domain adaptation gap between source and target image domains. In this paper, we propose a new multi-layered model architecture instead of SFCN single-layered model architecture. The proposed model extracts more meaningful features in image scenes along with large scale variations to increase the accuracy in cross domain crowd counting. Furthermore, with extensive experiments using four real-world datasets and analysis, we show that the proposed multi-layered architecture performs well with synthetic image data and few-shot learning in reducing domain shifts. © 2022 IEEE.

Exploiting probability density function of deep convolutional autoencoders' latent space for reliable COVID-19 detection on CT scans.

We present a probabilistic method for classifying chest computed tomography (CT) scans into COVID-19 and non-COVID-19. To this end, we design and train, in an unsupervised manner, a deep convolutional autoencoder (DCAE) on a selected training data set, which is composed only of COVID-19 CT scans. Once the model is trained, the encoder can generate the compact hidden representation (the hidden feature vectors) of the training data set. Afterwards, we exploit the obtained hidden representation to build up the target probability density function (PDF) of the training data set by means of kernel density estimation (KDE). Subsequently, in the test phase, we feed a test CT into the trained encoder to produce the corresponding hidden feature vector, and then, we utilise the target PDF to compute the corresponding PDF value of the test image. Finally, this obtained value is compared to a threshold to assign the COVID-19 label or non-COVID-19 to the test image. We numerically check our approach's performance (i.e. test accuracy and training times) by comparing it with those of some state-of-the-art methods.

Spatial and Temporal Analysis of COVID-19 in the Elderly Living in Residential Care Homes in Portugal.

BACKGROUND: The goal of this study is to identify geographic areas for priority actions in order to control COVID-19 among the elderly living in Residential Care Homes (RCH). We also describe the evolution of COVID-19 in RHC throughout the 278 municipalities of continental Portugal between March and December 2020. METHODS: A spatial population analysis of positive COVID-19 cases reported by the Portuguese National Health Service (NHS) among the elderly living in RCH. The data are for COVID-19 testing, symptomatic status, comorbidities, and income level by municipalities. COVID-19 measures at the municipality level are the proportion of positive cases of elderly living in RCH, positive cases per elderly living in RCH, symptomatic to asymptomatic ratio, and the share of comorbidities cases. Spatial analysis used the Kernel density estimation (KDE), space-time statistic Scan, and geographic weighted regression (GWR) to detect and analyze clusters of infected elderly. RESULTS: Between 3 March and 31 December 2020, the high-risk primary cluster was located in the regions of Braganca, Guarda, Vila Real, and Viseu, in the Northwest of Portugal (relative risk = 3.67), between 30 September and 13 December 2020. The priority geographic areas for attention and intervention for elderly living in care homes are the regions in the Northeast of Portugal, and around the large cities, Lisbon and Porto, which had high risk clusters. The relative risk of infection was spatially not stationary and generally positively affected by both comorbidities and low-income. CONCLUSION: The regions with a population with high comorbidities and low income are a priority for action in order to control COVID-19 in the elderly living in RCH. The results suggest improving both income and health levels in the southwest of Portugal, in the environs of large cities, such as Lisbon and Porto, and in the northwest of Portugal to mitigate the spread of COVID-19.

Density Map-based vehicle counting in remote sensing images with limited resolution

Observing traffic flow is of great significance to contemporary urban management. Overhead images, as represented by remote sensing images, provide a major source of information about traffic flow. However, the spatial resolutions of most common high-resolution remote sensing images are often limited to 0.5 m and even below, which makes it unrealistic to count vehicles by means of widely used object detection methods. Therefore, to explore the potential of remote sensing data for studying global urban development and management, this paper introduces a density map-based vehicle counting method for remote sensing imagery with limited resolution. Density map-based models regard the vehicle counting task as estimating the density of vehicle targets in terms of pixel values. We propose an improved CNN-based network, called Congested Scene Recognition Network Minus (CSRNet—), that generates a density map of vehicles from the input remote sensing imagery. A new dataset, RSVC2021, which was generated from the public DOTA and ITCVD datasets, is also introduced for network training and testing. A benchmark on the RSVC2021 dataset is accordingly established and CSRNet— is selected as the baseline model for subsequent experiments. A set of GF-2 time series images with a resolution of 1 m taken before, during and after the COVID-19 epidemic lockdown covering Wuhan city are applied for real-world application testing. The testing results on both the RSVC2021 dataset and real satellite images confirm that, in terms of both the counting values and the visualized density maps, the proposed method achieves good performance and exhibits considerable application potential in this task. The generating codes of RSVC2021 dataset will be publicly available at https://github.com/YinongGuo/RSVC2021-Dataset.

On semiparametric inference for periodically modulated density functions

We consider semiparametric inference for seasonally modulated density functions. Asymptotic results for kernel based estimators and simultaneous confidence bands are derived. The method is illustrated by an analysis of COVID-19 data from six European countries. [ FROM AUTHOR] Copyright of Communications in Statistics: Theory & Methods is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full . (Copyright applies to all s.)

DEEP PEDESTRIAN DENSITY ESTIMATION FOR SMART CITY MONITORING

Recently, requirement of city monitoring and maintenance using ICT techniques increases with the help of transportation system. In addition, the spread of COVID-19 has increased the demand for managing pedestrian traffic volume. To contribute to these trends, in this paper, we propose a new pedestrian radar map system in order to estimate pedestrian density on streets and sidewalks. Our system uses e-bikes to collect 360-degree images and visualize pedestrian positions as a radar map. In evaluations, we confirm the accuracies of the radar maps and pedestrian density by using KITTI dataset and by carrying out a field experiment. © 2021 IEEE

CLUSTERING ON THE TORUS BY CONFORMAL PREDICTION

Motivated by the analysis of torsion (dihedral) angles in the backbone of proteins, we investigate clustering of bivariate angular data on the torus [-π,π) × [-π,π). We show that naive adaptations of clustering methods, designed for vector-valued data, to the torus are not satisfactory and propose a novel clustering approach based on the conformal prediction framework. We construct several prediction sets for toroidal data with guaranteed finitesample validity, based on a kernel density estimate and bivariate von Mises mixture models. From a prediction set built from a Gaussian approximation of the bivariate von Mises mixture, we propose a data-driven choice for the number of clusters and present algorithms for an automated cluster identification and cluster membership assignment. The proposed prediction sets and clustering approaches are applied to the torsion angles extracted from three strains of coronavirus spike glycoproteins (including SARS-CoV-2, contagious in humans). The analysis reveals a potential difference in the clusters of the SARS-CoV-2 torsion angles, compared to the clusters found in torsion angles from two different strains of coronavirus, contagious in animals. © Institute of Mathematical Statistics, 2021.