scholarly journals Fusing a Bayesian case velocity model with random forest for predicting COVID-19 in the U.S.

2020 ◽  
Author(s):  
Gregory L Watson ◽  
Di Xiong ◽  
Lu Zhang ◽  
Joseph A Zoller ◽  
John Shamshoian ◽  
...  
Keyword(s):  
New York ◽  
Random Forest ◽  
Compartmental Model ◽  
Mixed Model ◽  
Population Level ◽  
Velocity Model ◽  
Random Forest Algorithm ◽  
Nonlinear Mixed Model ◽  
Uncertainty Estimates ◽  
The U.S

Predictions of COVID-19 case growth and mortality are critical to the decisions of political leaders, businesses, and individuals grappling with the pandemic. This predictive task is challenging due to the novelty of the virus, limited data, and dynamic political and societal responses. We embed a Bayesian nonlinear mixed model and a random forest algorithm within an epidemiological compartmental model for empirically grounded COVID-19 predictions. The Bayesian case model fits a location-specific curve to the velocity (first derivative) of the transformed cumulative case count, borrowing strength across geographic locations and incorporating prior information to obtain a posterior distribution for case trajectory. The compartmental model uses this distribution and predicts deaths using a random forest algorithm trained on COVID-19 data and population-level characteristics, yielding daily projections and interval estimates for infections and deaths in U.S. states. We evaluate forecasting accuracy on a two-week holdout set, finding that the model predicts COVID-19 cases and deaths well, with a mean absolute scaled error of 0.40 for cases and 0.32 for deaths throughout the two-week evaluation period. The substantial variation in predicted trajectories and associated uncertainty between states is illustrated by comparing three unique locations: New York, Ohio, and Mississippi. The sophistication and accuracy of this COVID-19 model offer reliable predictions and uncertainty estimates for the current trajectory of the pandemic in the U.S. and provide a platform for future predictions as shifting political and societal responses alter its course.

scholarly journals Pandemic velocity: Forecasting COVID-19 in the US with a machine learning & Bayesian time series compartmental model

2021 ◽  
Vol 17 (3) ◽  
pp. e1008837
Author(s):  
Gregory L. Watson ◽  
Di Xiong ◽  
Lu Zhang ◽  
Joseph A. Zoller ◽  
John Shamshoian ◽  
...  
Keyword(s):  
New York ◽  
Time Series ◽  
Random Forest ◽  
Compartmental Model ◽  
Predictive Accuracy ◽  
Population Level ◽  
Random Forest Algorithm ◽  
Case Count ◽  
Uncertainty Estimates ◽  
Bayesian Time Series

Predictions of COVID-19 case growth and mortality are critical to the decisions of political leaders, businesses, and individuals grappling with the pandemic. This predictive task is challenging due to the novelty of the virus, limited data, and dynamic political and societal responses. We embed a Bayesian time series model and a random forest algorithm within an epidemiological compartmental model for empirically grounded COVID-19 predictions. The Bayesian case model fits a location-specific curve to the velocity (first derivative) of the log transformed cumulative case count, borrowing strength across geographic locations and incorporating prior information to obtain a posterior distribution for case trajectories. The compartmental model uses this distribution and predicts deaths using a random forest algorithm trained on COVID-19 data and population-level characteristics, yielding daily projections and interval estimates for cases and deaths in U.S. states. We evaluated the model by training it on progressively longer periods of the pandemic and computing its predictive accuracy over 21-day forecasts. The substantial variation in predicted trajectories and associated uncertainty between states is illustrated by comparing three unique locations: New York, Colorado, and West Virginia. The sophistication and accuracy of this COVID-19 model offer reliable predictions and uncertainty estimates for the current trajectory of the pandemic in the U.S. and provide a platform for future predictions as shifting political and societal responses alter its course.

scholarly journals Fusing a Bayesian Case Velocity Model with Random Forest for Predicting COVID-19 in the U.S.

2020 ◽  
Author(s):  
Gregory L. Watson ◽  
Di Xiong ◽  
Lu Zhang ◽  
Joseph A. Zoller ◽  
John Shamshoian ◽  
...  
Keyword(s):  
Random Forest ◽  
Velocity Model ◽  
The U.S

The Role of Voice Quality in Mandarin Sarcastic Speech: An Acoustic and Electroglottographic Study

2020 ◽  
Vol 63 (8) ◽  
pp. 2578-2588
Author(s):  
Shanpeng Li ◽  
Wentao Gu ◽  
Lei Liu ◽  
Ping Tang
Keyword(s):  
Random Forest ◽  
Gender Difference ◽  
Mixed Model ◽  
Linear Mixed Model ◽  
Voice Quality ◽  
Quality Parameters ◽  
Prosodic Features ◽  
Random Forest Algorithm ◽  
Classification Analysis

Purpose Sarcasm is a specialized speech act in daily vocal communication usually characterized by unique prosodic features, but the role of voice quality in expressing sarcasm has not been explored much. The goal of this study is to explore the voice quality features of Mandarin sarcastic speech in comparison to sincere speech. Method Fifteen male and 15 female native speakers of Mandarin uttered 31 target sentences with both sincere and sarcastic attitudes. Nine voice quality parameters extracted from the acoustic and electroglottographic signals were analyzed using a linear mixed model, and a classification analysis using a random forest algorithm was conducted to identify the relative contribution of these parameters to the differentiation between sincere and sarcastic utterances. Results In comparison to sincere speech, sarcastic speech had a creakier voice, which was characterized by a lower fundamental frequency, a greater degree of vocal fold adduction (i.e., higher contact quotient), lesser noise (i.e., higher harmonics-to-noise ratio), and more multiple pulsing (i.e., higher subharmonic-to-harmonic ratio). The interaction effect revealed a gender difference in the use of creakier voice to express sarcasm in Mandarin. The classification analysis using the random forest algorithm showed that the nine voice quality parameters resulted in 84.0% and 83.7% identification rates for sarcastic and sincere utterances, respectively. Conclusions The results of this preliminary study support the role of voice quality in expressing sarcasm in Mandarin speech. Using a set of voice quality parameters, sarcastic and sincere utterances can be effectively identified. Furthermore, there is a gender difference in the use of creakier voice in expressing Mandarin sarcastic speech. Supplemental Material https://doi.org/10.23641/asha.12743780

METHOD SUGGESTING CITY WALKING ROUTES FOR PEDESTRIANS USING AN EXAMPLE OF SAINT-PETERSBURG

2019 ◽  
pp. 71-76
Author(s):  
A.E. Semenov
Keyword(s):  
Random Forest ◽  
Develop Model ◽  
Random Forest Algorithm ◽  
Saint Petersburg ◽  
Pedestrian Navigation ◽  
Factors Influencing ◽  
The City

The method of pedestrian navigation in the cities illustrated by the example of Saint-Petersburg was investigated. The factors influencing people when they choose a route for their walk were determined. Based on acquired factors corresponding data was collected and used to develop model determining attractiveness of a street in the city using Random Forest algorithm. The results obtained shows that routes provided by the method are 14% more attractive and just 6% longer compared with the shortest ones.

Contributions of Joseph Hardcastle to Accounting Theory

2008 ◽  
Vol 35 (2) ◽  
pp. 145-179 ◽  
Author(s):  
George C. Romeo ◽  
James J. McKinney
Keyword(s):  
New York ◽  
Turn Of The Century ◽  
Strong Impact ◽  
New York University ◽  
Honorary Degree ◽  
Cpa Exam ◽  
Accounting Theory ◽  
The U.S

Joseph Hardcastle was one of the foremost authorities on subjects connected with the mathematics of finance and other topics in accounting in the late 19th and early 20th centuries. As a teacher, author, and leader in the profession, he figured prominently in the elevation of accountancy. Hardcastle is relatively unknown in the literature except for having the distinction of scoring the highest grades on the first CPA exam in New York in 1896. However, he was well respected during his time as one of the premier theorists in accounting and was awarded an honorary degree of Master of Letters by New York University. Because of his prolific writings, his teaching of future accountants, and his interactions with members of the Institute of Accounts, he had a strong impact on the “science of accounts,” the dominant accounting theory in the U.S. at the turn of the century.

Sign in / Sign up

Export Citation Format

Share Document