Given the importance of web search volume for reflecting tourists'preferences for certain tourism services and destinations,incorporating these data into forecasting models can significantly improve forecasting pe...Given the importance of web search volume for reflecting tourists'preferences for certain tourism services and destinations,incorporating these data into forecasting models can significantly improve forecasting performance.This study enriches the literature on tourism demand forecasting and tourists'search behavior through segmented Baidu search volume data.First,this study divides Baidu search volume data based on volume sources and periods.Then,by analyzing the most relevant keywords in tourism demand in different segments,this study captures the dynamic characteristics of tourist search behavior.Finally,this study adopts a series of econometric and machine learning models to further improve the performance of tourism demand and forecasting.The findings indicate that tourists’search behavior has changed significantly with the prevalence and popularization of 4G technology and suggest that search volume improves forecasting performance,especially search volume on mobile terminals,from 2014M1–2019M12.展开更多
Coronavirus disease 2019(COVID-19)is an emerging infectious disease,and it is important to detect early and monitor the disease trend for policymakers to make informed decisions.We explored the predictive utility of B...Coronavirus disease 2019(COVID-19)is an emerging infectious disease,and it is important to detect early and monitor the disease trend for policymakers to make informed decisions.We explored the predictive utility of Baidu Search Index and Baidu Information Index for early warning of COVID-19 and identified search keywords for further monitoring of epidemic trends in Guangxi.A time-series analysis and Spearman correlation between the daily number of cases and both the Baidu Search Index and Baidu Information Index were performed for seven keywords related to COVID-19 from January 8 to March 9,2020.The time series showed that the temporal distributions of the search terms“coronavirus,”“pneumonia”and“mask”in the Baidu Search Index were consistent and had 2 to 3 days'lead time to the reported cases;the correlation coefficients were higher than 0.81.The Baidu Search Index volume in 14 prefectures of Guangxi was closely related with the number of reported cases;it was not associated with the local GDP.The Baidu Information Index search terms“coronavirus”and“pneumonia”were used as frequently as 192,405.0 and 110,488.6 per million population,respectively,and they were also significantly associated with the number of reported cases(rs>0.6),but they fluctuated more than for the Baidu Search Index and had 0 to 14 days'lag time to the reported cases.The Baidu Search Index with search terms“coronavirus,”“pneumonia”and“mask”can be used for early warning and monitoring of the epidemic trend of COVID-19 in Guangxi,with 2 to 3 days'lead time.展开更多
Influenza is a kind of infectious disease, which spreads quickly and widely. The outbreak of influenza has brought huge losses to society. In this paper, four major categories of flu keywords, “prevention phase”, “...Influenza is a kind of infectious disease, which spreads quickly and widely. The outbreak of influenza has brought huge losses to society. In this paper, four major categories of flu keywords, “prevention phase”, “symptom phase”, “treatment phase”, and “commonly-used phrase” were set. Python web crawler was used to obtain relevant influenza data from the National Influenza Center’s influenza surveillance weekly report and Baidu Index. The establishment of support vector regression (SVR), least absolute shrinkage and selection operator (LASSO), convolutional neural networks (CNN) prediction models through machine learning, took into account the seasonal characteristics of the influenza, also established the time series model (ARMA). The results show that, it is feasible to predict influenza based on web search data. Machine learning shows a certain forecast effect in the prediction of influenza based on web search data. In the future, it will have certain reference value in influenza prediction. The ARMA(3,0) model predicts better results and has greater generalization. Finally, the lack of research in this paper and future research directions are given.展开更多
Surveillance is an essential work on infectious diseases prevention and control.When the pandemic occurred,the inadequacy of traditional surveillance was exposed,but it also provided a valuable opportunity to explore ...Surveillance is an essential work on infectious diseases prevention and control.When the pandemic occurred,the inadequacy of traditional surveillance was exposed,but it also provided a valuable opportunity to explore new surveillance methods.This study aimed to estimate the transmission dynamics and epidemic curve of severe acute respiratory syndrome coronavirus 2(SARS-Co V-2)Omicron BF.7 in Beijing under the emergent situation using Baidu index and influenza-like illness(ILI)surveillance.A novel hybrid model(multiattention bidirectional gated recurrent unit(MABG)-susceptible-exposed-infected-removed(SEIR))was developed,which leveraged a deep learning algorithm(MABG)to scrutinize the past records of ILI occurrences and the Baidu index of diverse symptoms such as fever,pyrexia,cough,sore throat,anti-fever medicine,and runny nose.By considering the current Baidu index and the correlation between ILI cases and coronavirus disease 2019(COVID-19)cases,a transmission dynamics model(SEIR)was formulated to estimate the transmission dynamics and epidemic curve of SARS-Co V-2.During the COVID-19 pandemic,when conventional surveillance measures have been suspended temporarily,cases of ILI can serve as a useful indicator for estimating the epidemiological trends of COVID-19.In the specific case of Beijing,it has been ascertained that cumulative infection attack rate surpass 80.25%(95%confidence interval(95%CI):77.51%-82.99%)since December 17,2022,with the apex of the outbreak projected to transpire on December 12.The culmination of existing patients is expected to occur three days subsequent to this peak.Effective reproduction number(Rt)represents the average number of secondary infections generated from a single infected individual at a specific point in time during an epidemic,remained below 1 since December 17,2022.The traditional disease surveillance systems should be complemented with information from modern surveillance data such as online data sources with advanced technical support.Modern surveillance channels should be used primarily in emerging infectious and disease outbreaks.Syndrome surveillance on COVID-19 should be established to following on the epidemic,clinical severity,and medical resource demand.展开更多
基金partly supported by the National Natural Science Foundation of China under Grant No.72101197by the Fundamental Research Funds for the Central Universities under Grant No.SK2021007.
文摘Given the importance of web search volume for reflecting tourists'preferences for certain tourism services and destinations,incorporating these data into forecasting models can significantly improve forecasting performance.This study enriches the literature on tourism demand forecasting and tourists'search behavior through segmented Baidu search volume data.First,this study divides Baidu search volume data based on volume sources and periods.Then,by analyzing the most relevant keywords in tourism demand in different segments,this study captures the dynamic characteristics of tourist search behavior.Finally,this study adopts a series of econometric and machine learning models to further improve the performance of tourism demand and forecasting.The findings indicate that tourists’search behavior has changed significantly with the prevalence and popularization of 4G technology and suggest that search volume improves forecasting performance,especially search volume on mobile terminals,from 2014M1–2019M12.
基金supported by the Health and Emergency Skills Training Center of Guangxi(HESTCG202104)National Natural Science Foundation of China(11971479)Guangxi Bagui Honor Scholarship and Chinese State Key Laboratory of Infectious Disease Prevention and Control.
文摘Coronavirus disease 2019(COVID-19)is an emerging infectious disease,and it is important to detect early and monitor the disease trend for policymakers to make informed decisions.We explored the predictive utility of Baidu Search Index and Baidu Information Index for early warning of COVID-19 and identified search keywords for further monitoring of epidemic trends in Guangxi.A time-series analysis and Spearman correlation between the daily number of cases and both the Baidu Search Index and Baidu Information Index were performed for seven keywords related to COVID-19 from January 8 to March 9,2020.The time series showed that the temporal distributions of the search terms“coronavirus,”“pneumonia”and“mask”in the Baidu Search Index were consistent and had 2 to 3 days'lead time to the reported cases;the correlation coefficients were higher than 0.81.The Baidu Search Index volume in 14 prefectures of Guangxi was closely related with the number of reported cases;it was not associated with the local GDP.The Baidu Information Index search terms“coronavirus”and“pneumonia”were used as frequently as 192,405.0 and 110,488.6 per million population,respectively,and they were also significantly associated with the number of reported cases(rs>0.6),but they fluctuated more than for the Baidu Search Index and had 0 to 14 days'lag time to the reported cases.The Baidu Search Index with search terms“coronavirus,”“pneumonia”and“mask”can be used for early warning and monitoring of the epidemic trend of COVID-19 in Guangxi,with 2 to 3 days'lead time.
文摘Influenza is a kind of infectious disease, which spreads quickly and widely. The outbreak of influenza has brought huge losses to society. In this paper, four major categories of flu keywords, “prevention phase”, “symptom phase”, “treatment phase”, and “commonly-used phrase” were set. Python web crawler was used to obtain relevant influenza data from the National Influenza Center’s influenza surveillance weekly report and Baidu Index. The establishment of support vector regression (SVR), least absolute shrinkage and selection operator (LASSO), convolutional neural networks (CNN) prediction models through machine learning, took into account the seasonal characteristics of the influenza, also established the time series model (ARMA). The results show that, it is feasible to predict influenza based on web search data. Machine learning shows a certain forecast effect in the prediction of influenza based on web search data. In the future, it will have certain reference value in influenza prediction. The ARMA(3,0) model predicts better results and has greater generalization. Finally, the lack of research in this paper and future research directions are given.
基金supported by grants from the Chinese Academy of Medical Sciences(CAMS)Innovation Fund for Medical Sciences(2021I2M-1-044)。
文摘Surveillance is an essential work on infectious diseases prevention and control.When the pandemic occurred,the inadequacy of traditional surveillance was exposed,but it also provided a valuable opportunity to explore new surveillance methods.This study aimed to estimate the transmission dynamics and epidemic curve of severe acute respiratory syndrome coronavirus 2(SARS-Co V-2)Omicron BF.7 in Beijing under the emergent situation using Baidu index and influenza-like illness(ILI)surveillance.A novel hybrid model(multiattention bidirectional gated recurrent unit(MABG)-susceptible-exposed-infected-removed(SEIR))was developed,which leveraged a deep learning algorithm(MABG)to scrutinize the past records of ILI occurrences and the Baidu index of diverse symptoms such as fever,pyrexia,cough,sore throat,anti-fever medicine,and runny nose.By considering the current Baidu index and the correlation between ILI cases and coronavirus disease 2019(COVID-19)cases,a transmission dynamics model(SEIR)was formulated to estimate the transmission dynamics and epidemic curve of SARS-Co V-2.During the COVID-19 pandemic,when conventional surveillance measures have been suspended temporarily,cases of ILI can serve as a useful indicator for estimating the epidemiological trends of COVID-19.In the specific case of Beijing,it has been ascertained that cumulative infection attack rate surpass 80.25%(95%confidence interval(95%CI):77.51%-82.99%)since December 17,2022,with the apex of the outbreak projected to transpire on December 12.The culmination of existing patients is expected to occur three days subsequent to this peak.Effective reproduction number(Rt)represents the average number of secondary infections generated from a single infected individual at a specific point in time during an epidemic,remained below 1 since December 17,2022.The traditional disease surveillance systems should be complemented with information from modern surveillance data such as online data sources with advanced technical support.Modern surveillance channels should be used primarily in emerging infectious and disease outbreaks.Syndrome surveillance on COVID-19 should be established to following on the epidemic,clinical severity,and medical resource demand.
