Richard J. Povinelli

Accurate time series forecasting often requires higher temporal resolution than that provided by available data, such as when daily forecasts are needed from monthly data. Existing temporal disaggregation techniques, which typically handle only single, uniformly sampled time series, have limited applicability in real-world, multi-source scenarios. This paper introduces the Iterative Shifting Disaggregation (ISD) algorithm, designed to process and disaggregate time series derived from sensor-sourced low-frequency measurements, transforming multiple, nonuniformly sampled sensor data streams into a single, coherent high-frequency signal. ISD operates in an iterative, two-phase process: a prediction phase that uses multiple linear regression to generate high-frequency series from low-frequency data and correlated variables, followed by an update phase that redistributes low-frequency observations across high-frequency periods. This process repeats, refining estimates with each iteration cycle. The ISD algorithm’s key contribution is its ability to disaggregate multiple, nonuniformly spaced time series with overlapping intervals into a single daily representation. In two case studies using natural gas data, ISD successfully disaggregates billing cycle and grouped residential customer data into daily time series, achieving a 1.4–4.3% WMAPE improvement for billing cycle data and a 4.6–10.4% improvement for residential data over existing methods.

Local natural gas distribution companies (LDCs) require accurate demand forecasts across various time periods, geographic regions, and customer class hierarchies. Achieving coherent forecasts across these hierarchies is challenging but crucial for optimal decision making, resource allocation, and operational efficiency. This work introduces a method that structures the gas distribution system into cross-temporal hierarchies to produce accurate and coherent forecasts. We apply our method to a case study involving three operational regions, forecasting at different geographical levels and analyzing both hourly and daily frequencies. Trained on five years of data and tested on one year, our model achieves a 10% reduction in hourly mean absolute scaled error and a 3% reduction in daily mean absolute scaled error.

Aim: Approximately six million people suffer cardiac arrests worldwide per year with very low survival rates (<1%). Thus, the aim of this study is to estimate the probability of a poor outcome after cardiac arrest. Accurate outcome predictions avoid removing care too soon for patients with potentially good outcomes or continuing care for patients with likely poor outcomes. Method: The method is based on dynamical systems embedding theorems that show that a reconstructed phase space (RPS) topologically equivalent to an underlying system can be constructed from measured signals. Here the underlying system is the human brain after a cardiac arrest, and the signals are the EEG channels. We model the RPS with a Gaussian mixture model (GMM) and ensemble the output of the RPS-GMM with clinical data via XGBoost. Results: As team Blue and Gold in the Predicting Neurological Recovery from Coma After Cardiac Arrest: The George B. Moody PhysioNet Challenge 2023, our RPS-GMM-XGBoost method obtained a test set competition score of 0.426 and rank of 24/36.

After mating, female mosquitoes need animal blood to develop their eggs. In the process of acquiring blood, they may acquire pathogens, which may cause different diseases in humans such as malaria, zika, dengue, and chikungunya. Therefore, knowing the parity status of mosquitoes is useful in control and evaluation of infectious diseases transmitted by mosquitoes, where parous mosquitoes are assumed to be potentially infectious. Ovary dissections, which are currently used to determine the parity status of mosquitoes, are very tedious and limited to few experts. An alternative to ovary dissections is near-infrared spectroscopy (NIRS), which can estimate the age in days and the infectious state of laboratory and semi-field reared mosquitoes with accuracies between 80 and 99%. No study has tested the accuracy of NIRS for estimating the parity status of wild mosquitoes. In this study, we train an artificial neural network (ANN) models on NIR spectra to estimate the parity status of wild mosquitoes. We use four different datasets: An. arabiensis collected from Minepa, Tanzania (Minepa-ARA); An. gambiae s.s collected from Muleba, Tanzania (Muleba-GA); An. gambiae s.s collected from Burkina Faso (Burkina-GA); and An.gambiae s.s from Muleba and Burkina Faso combined (Muleba-Burkina-GA). We train ANN models on datasets with spectra preprocessed according to previous protocols. We then use autoencoders to reduce the spectra feature dimensions from 1851 to 10 and re-train the ANN models. Before the autoencoder was applied, ANN models estimated parity status of mosquitoes in Minepa-ARA, Muleba-GA, Burkina-GA and Muleba-Burkina-GA with out-of-sample accuracies of 81.9±2.8 (N = 274), 68.7±4.8 (N = 43), 80.3±2.0 (N = 48), and 75.7±2.5 (N = 91), respectively. With the autoencoder, ANN models tested on out-of-sample data achieved 97.1±2.2% (N = 274), 89.8 ± 1.7% (N = 43), 93.3±1.2% (N = 48), and 92.7±1.8% (N = 91) accuracies for Minepa-ARA, Muleba-GA, Burkina-GA, and Muleba-Burkina-GA, respectively. These results show that a combination of an autoencoder and an ANN trained on NIR spectra to estimate the parity status of wild mosquitoes yields models that can be used as an alternative tool to estimate parity status of wild mosquitoes, especially since NIRS is a high-throughput, reagent-free, and simple-to-use technique compared to ovary dissections.

This article addresses the need to divide a population of water utility customers into groups based on their similarities and differences, using only the measured flow data collected by water meters. After clustering, the groups represent customers with similar consumption behavior patterns and provide insight into “normal” and “unusual” customer behavior patterns for individually metered water utility customers within North America. The contributions of this work not only represent a novel work, but also solve a practical problem for the utility industry. This article introduces a method of agglomerative clustering using information theoretic distance measures on Gaussian mixture models within a reconstructed phase space, designed to accommodate a utility's limited human, financial, computational, and environmental resources. The proposed weighted variation of information distance measure for comparing Gaussian mixture models emphasizes those behaviors whose statistical distributions are more compact over those behaviors with large variation and contributes a novel addition to existing comparison options. We conduct several experiments with both synthetic and real data to show the reasonableness of the clustering results and their consistency.

Background: After mating, female mosquitoes need animal blood to develop their eggs. In the process of acquiring blood, they may acquire pathogens, which may cause different diseases to humans such as malaria, zika, dengue, and chikungunya. Therefore, knowing the parity status of mosquitoes is useful in control and evaluation of infectious diseases transmitted by mosquitoes, where parous mosquitoes are assumed to be potentially infectious. Ovary dissections, which currently are used to determine the parity status of mosquitoes, are very tedious and limited to very few experts. An alternative to ovary dissections is near-infrared spectroscopy (NIRS), which can estimate the age in days and the infectious state of laboratory and semi-field reared mosquitoes with accuracies between 80 and 99%. No study has tested the accuracy of NIRS for estimating the parity status of wild mosquitoes.Methods and results: In this study, we train artificial neural network (ANN) models on NIR spectra to estimate the parity status of wild mosquitoes. We use four different datasets:An. arabiensiscollected from Minepa, Tanzania (Minepa-ARA);An. gambiaecollected from Muleba, Tanzania (Muleba-GA);An. gambiaecollected from Burkina Faso (Burkina-GA); andAn.gambiaefrom Muleba and Burkina Faso combined (Muleba-Burkina-GA). We train ANN models on datasets with spectra preprocessed according to previous protocols. We then use autoencoders to reduce the spectra feature dimensions from 1851 to 10 and re-train ANN models. Before the autoencoder was applied, ANN models estimated parity status of mosquitoes in Minepa-ARA, Muleba-GA, Burkina-GA and Muleba-Burkina-GA with out-of-sample accuracies of 81.9 ± 2.8% (N=927), 68.7 ± 4.8% (N=140), 80.3 ± 2.0% (N=158), and 75.7 ± 2.5% (N=298), respectively. With the autoencoder, ANN models tested on out-of-sample data achieved 97.1 ± 2.2%, (N=927), 89.8 ± 1.7% (N=140), 93.3 ± 1.2% (N=158), and 92.7 ± 1.8% (N=298) accuracies for Minepa-ARA, Muleba-GA, Burkina-GA, and Muleba-Burkina-GA, respectively.Conclusion: These results show that a combination of an autoencoder and an ANN trained on NIR spectra to estimate parity status of wild mosquitoes yields models that can be used as an alternative tool to estimate parity status of wild mosquitoes, especially since NIRS is a high-throughput, reagent-free, and simple-to-use technique compared to ovary dissections.

Near infrared spectroscopy (NIRS) is currently complementing techniques to age-grade mosquitoes. NIRS classifies lab-reared and semi-field raised mosquitoes into < or ≥ 7 days old with an average accuracy of 80%, achieved by training a regression model using partial least squares (PLS) and interpreted as a binary classifier. We explore whether using an artificial neural network (ANN) analysis instead of PLS regression improves the current accuracy of NIRS models for age-grading malaria transmitting mosquitoes. We also explore if directly training a binary classifier instead of training a regression model and interpreting it as a binary classifier improves the accuracy. A total of 786 and 870 NIR spectra collected from laboratory reared An. gambiae and An. arabiensis, respectively, were used and pre-processed according to previously published protocols. The ANN regression model scored root mean squared error (RMSE) of 1.6 ± 0.2 for An. gambiae and 2.8 ± 0.2 for An. arabiensis; whereas the PLS regression model scored RMSE of 3.7 ± 0.2 for An. gambiae, and 4.5 ± 0.1 for An. arabiensis. When we interpreted regression models as binary classifiers, the accuracy of the ANN regression model was 93.7 ± 1.0% for An. gambiae, and 90.2 ± 1.7% for An. arabiensis; while PLS regression model scored the accuracy of 83.9 ± 2.3% for An. gambiae, and 80.3 ± 2.1% for An. arabiensis. We also find that a directly trained binary classifier yields higher age estimation accuracy than a regression model interpreted as a binary classifier. A directly trained ANN binary classifier scored an accuracy of 99.4 ± 1.0 for An. gambiae and 99.0 ± 0.6% for An. arabiensis; while a directly trained PLS binary classifier scored 93.6 ± 1.2% for An. gambiae and 88.7 ± 1.1% for An. arabiensis. We further tested the reproducibility of these results on different independent mosquito datasets. ANNs scored higher estimation accuracies than when the same age models are trained using PLS. Regardless of the model architecture, directly trained binary classifiers scored higher accuracies on classifying age of mosquitoes than regression models translated as binary classifiers. We recommend training models to estimate age of An. arabiensis and An. gambiae using ANN model architectures (especially for datasets with at least 70 mosquitoes per age group) and direct training of binary classifier instead of training a regression model and interpreting it as a binary classifier.

Background Near infrared spectroscopy (NIRS) is currently complementing techniques to age-grade mosquitoes. NIRS classifies lab-reared and semi-field raised mosquitoes into < or ≥ 7 days old with an average accuracy of 80%, achieved by training a regression model using partial least squares (PLS) and interpreted as a binary classifier. Methods and findings We explore whether using an artificial neural network (ANN) analysis instead of PLS regression improves the current accuracy of NIRS models for age-grading malaria transmitting mosquitoes. We also explore if directly training a binary classifier instead of training a regression model and interpreting it as a binary classifier improves the accuracy. A total of 786 and 870 NIR spectra collected from laboratory reared An. gambiae and An. arabiensis, respectively, were used and pre-processed according to previously published protocols. Based on ten-fold Monte Carlo cross-validation, an ANN regression model scored root mean squared error (RMSE) of 1.6 ± 0.2 for An. gambiae and 2.8 ± 0.2 for An. arabiensis; whereas the PLS regression model scored RMSE of 3.7 ± 0.2 for An. gambiae, and 4.5 ± 0.1 for An. arabiensis. When we interpreted regression models as binary classifiers, the accuracy of the ANN regression model was 93.7 ± 1.0 % for An. gambiae, and 90.2 ± 1.7 % for An. arabiensis; while PLS regression model scored the accuracy of 83.9 ± 2.3% for An. gambiae, and 80.3 ± 2.1% for An. arabiensis. We also find that a directly trained binary classifier yields higher age estimation accuracy than a regression model interpreted as a binary classifier. A directly trained ANN binary classifier scored an accuracy of 99.4 ± 1.0 for An. gambiae, and 99.0 ± 0.6% for An. arabiensis; while a directly trained PLS binary classifier scored 93.6 ± 1.2% for An. gambiae, and 88.7 ± 1.1% for An. arabiensis. Conclusion Training both regression and binary classification age models using ANNs yields models with higher estimation accuracies than when the same age models are trained using PLS. Regardless of the model architecture, directly trained binary classifiers score higher accuracy on classifying age of mosquitoes than a regression model translated as binary classifier. Therefore, we recommend training models to estimate age of An. gambiae and An. arabiensis using ANN model architectures and direct training of binary classifier instead of training a regression model and interpret it as a binary classifier.