Search

Accurate prior information of passenger flow demand on high‐speed railway is of great significance for the operation and the management of transportation systems. Various factors in modern social life have caused uncertainty at demand. Recently, individuals are increasingly depending on the online search results when choosing among different transportation modes, services, and destinations, which provide important basic information for forecasting the travel demand. This study employs Baidu search index to assist in capturing volatility of high‐speed railway passenger demands, offering insights into the travel inclinations and travelers' actions. Furthermore, we have given more in‐depth attention and analysis to their residual term accounting for the random nature caused by various factors. To this end, a sophisticated deep analysis mechanism based on data decomposition has been devised to extract and analyze the valuable information concealed within the residuals, so as to enhance the comprehension of the variability inherent in the high‐speed railway passenger flow. Meanwhile, an error correction strategy is implemented for all residual terms to improve further their forecasting accuracy. Experimental results from two real‐world datasets demonstrate the effectiveness and robustness of the developed hybrid approach across several popular evaluation indicators. Therefore, this study can function as a reliable instrument, provide sensible data‐driven guidance for resource allocation and make scientific decisions in the railway industry. [ABSTRACT FROM AUTHOR]

Background: It is difficult to detect the outbreak of emergency infectious disease based on the exiting surveillance system. Here we investigate the utility of the Baidu Search Index, an indicator of how large of a keyword is in Baidu's search volume, in the early warning and predicting the epidemic trend of COVID-19. Methods: The daily number of cases and the Baidu Search Index of 8 keywords (weighted by population) from December 1, 2019 to March 15, 2020 were collected and analyzed with times series and Spearman correlation with different time lag. To predict the daily number of COVID-19 cases using the Baidu Search Index, Zero-inflated negative binomial regression was used in phase 1 and negative binomial regression model was used in phase 2 and phase 3 based on the characteristic of independent variable. Results: The Baidu Search Index of all keywords in Wuhan was significantly higher than Hubei (excluded Wuhan) and China (excluded Hubei). Before the causative pathogen was identified, the search volume of "Influenza" and "Pneumonia" in Wuhan increased with the number of new onset cases, their correlation coefficient was 0.69 and 0.59, respectively. After the pathogen was public but before COVID-19 was classified as a notifiable disease, the search volume of "SARS", "Pneumonia", "Coronavirus" in all study areas increased with the number of new onset cases with the correlation coefficient was 0.69 ~ 0.89, while "Influenza" changed to negative correlated (rs: -0.56 ~ -0.64). After COVID-19 was closely monitored, the Baidu Search Index of "COVID-19", "Pneumonia", "Coronavirus", "SARS" and "Mask" could predict the epidemic trend with 15 days, 5 days and 6 days lead time, respectively in Wuhan, Hubei (excluded Wuhan) and China (excluded Hubei). The predicted number of cases would increase 1.84 and 4.81 folds, respectively than the actual number of cases in Wuhan and Hubei (excluded Wuhan) from 21 January to 9 February. Conclusion: The Baidu Search Index could be used in the early warning and predicting the epidemic trend of COVID-19, but the search keywords changed in different period. Considering the time lag from onset to diagnosis, especially in the areas with medical resources shortage, internet search data can be a highly effective supplement of the existing surveillance system. [ABSTRACT FROM AUTHOR]

Abstract Background It is difficult to detect the outbreak of emergency infectious disease based on the exiting surveillance system. Here we investigate the utility of the Baidu Search Index, an indicator of how large of a keyword is in Baidu’s search volume, in the early warning and predicting the epidemic trend of COVID-19. Methods The daily number of cases and the Baidu Search Index of 8 keywords (weighted by population) from December 1, 2019 to March 15, 2020 were collected and analyzed with times series and Spearman correlation with different time lag. To predict the daily number of COVID-19 cases using the Baidu Search Index, Zero-inflated negative binomial regression was used in phase 1 and negative binomial regression model was used in phase 2 and phase 3 based on the characteristic of independent variable. Results The Baidu Search Index of all keywords in Wuhan was significantly higher than Hubei (excluded Wuhan) and China (excluded Hubei). Before the causative pathogen was identified, the search volume of “Influenza” and “Pneumonia” in Wuhan increased with the number of new onset cases, their correlation coefficient was 0.69 and 0.59, respectively. After the pathogen was public but before COVID-19 was classified as a notifiable disease, the search volume of “SARS”, “Pneumonia”, “Coronavirus” in all study areas increased with the number of new onset cases with the correlation coefficient was 0.69 ~ 0.89, while “Influenza” changed to negative correlated (rs: -0.56 ~ -0.64). After COVID-19 was closely monitored, the Baidu Search Index of “COVID-19”, “Pneumonia”, “Coronavirus”, “SARS” and “Mask” could predict the epidemic trend with 15 days, 5 days and 6 days lead time, respectively in Wuhan, Hubei (excluded Wuhan) and China (excluded Hubei). The predicted number of cases would increase 1.84 and 4.81 folds, respectively than the actual number of cases in Wuhan and Hubei (excluded Wuhan) from 21 January to 9 February. Conclusion The Baidu Search Index could be used in the early warning and predicting the epidemic trend of COVID-19, but the search keywords changed in different period. Considering the time lag from onset to diagnosis, especially in the areas with medical resources shortage, internet search data can be a highly effective supplement of the existing surveillance system.

Severe outdoor air pollution in China has long constituted a threat to public health and environmental governance efficiency. Surveillance of online search behaviours that reflect real-time public risk perception and behavioural responses on environmental issues is essential in improving environmental governance. However, owing to the absence of an appropriate analytical framework, quantitative evidence of analysing online data is scarce in the extant research. To address this research gap, we analysed Baidu search indexes (BSI) keywords relevant to the public risk perception and behavioural responses on outdoor air pollution by a surveillance framework consisting of two parallel steps (prediction and monitoring). The proposed framework may improve environmental risk governance efficiency and benefit environmental risk governance policy decision-making through considering real-time public risk perceptions and behavioural response information. Using online search behaviour data, this study adds to the methodology for improving the efficacy of environmental social risk governance. [ABSTRACT FROM AUTHOR]

Introduction: With the advancement of society and technology, the importance of professional competency has been increasingly emphasized. This study aims to provide an in-depth analysis of the Baidu search index for "professional competency" from July 3, 2017, to October 6, 2023, across 34 provincial-level administrative regions in China. Methods: The study begins with a literature review to establish a theoretical framework. Subsequently, it delves into the fundamental concepts of complex networks, their parameters, and the principles of visible network graph algorithms. The application of system clustering and community detection algorithms is explored within this framework. Results: The empirical analysis starts with a provincial-level examination, involving the construction of visible network graphs, the exploration of scale-free networks, and discussions on regional system clustering. Following this, a nationwide analysis is conducted, focusing on the construction of visible network graphs to visualize the awareness of professional competency and studying community structures. Discussion: The findings provide insights into the awareness of professional competency across different regions. These insights are valuable for informing policy formulation and corporate decision-making, contributing to a deeper understanding of professional competency awareness and its implications. [ABSTRACT FROM AUTHOR]

BackgroundChina exited strict Zero-COVID policy with a surge in Omicron variant infections in December 2022. Given China’s pandemic policy and population immunity, employing Baidu Index (BDI) to analyze the evolving disease landscape and estimate the nationwide pneumonia hospitalizations in the post Zero COVID period, validated by hospital data, holds informative potential for future outbreaks.MethodsRetrospective observational analyses were conducted at the conclusion of the Zero-COVID policy, integrating internet search data alongside offline records. Methodologies employed were multidimensional, encompassing lagged Spearman correlation analysis, growth rate assessments, independent sample T-tests, Granger causality examinations, and Bayesian structural time series (BSTS) models for comprehensive data scrutiny.ResultsVarious diseases exhibited a notable upsurge in the BDI after the policy change, consistent with the broader trajectory of the COVID-19 pandemic. Robust connections emerged between COVID-19 and diverse health conditions, predominantly impacting the respiratory, circulatory, ophthalmological, and neurological domains. Notably, 34 diseases displayed a relatively high correlation (r > 0.5) with COVID-19. Among these, 12 exhibited a growth rate exceeding 50% post-policy transition, with myocarditis escalating by 1,708% and pneumonia by 1,332%. In these 34 diseases, causal relationships have been confirmed for 23 of them, while 28 garnered validation from hospital-based evidence. Notably, 19 diseases obtained concurrent validation from both Granger causality and hospital-based data. Finally, the BSTS models approximated approximately 4,332,655 inpatients diagnosed with pneumonia nationwide during the 2 months subsequent to the policy relaxation.ConclusionThis investigation elucidated substantial associations between COVID-19 and respiratory, circulatory, ophthalmological, and neurological disorders. The outcomes from comprehensive multi-dimensional cross-over studies notably augmented the robustness of our comprehension of COVID-19’s disease spectrum, advocating for the prospective utility of internet-derived data. Our research highlights the potential of Internet behavior in predicting pandemic-related syndromes, emphasizing its importance for public health strategies, resource allocation, and preparedness for future outbreaks.

Abstract Background Seasonal influenza and other respiratory tract infections are serious public health problems that need to be further addressed and investigated. Internet search data are recognized as a valuable source for forecasting influenza or other respiratory tract infection epidemics. However, the selection of internet search data and the application of forecasting methods are important for improving forecasting accuracy. The aim of the present study was to forecast influenza epidemics based on the long short-term memory neural network (LSTM) method, Baidu search index data, and the influenza-like-illness (ILI) rate. Methods The official weekly ILI% data for northern and southern mainland China were obtained from the Chinese Influenza Center from 2018 to 2021. Based on the Baidu Index, search indices related to influenza infection over the corresponding time period were obtained. Pearson correlation analysis was performed to explore the association between influenza-related search queries and the ILI% of southern and northern mainland China. The LSTM model was used to forecast the influenza epidemic within the same week and at lags of 1–4 weeks. The model performance was assessed by evaluation metrics, including the mean square error (MSE), root mean square error (RMSE) and mean absolute error (MAE). Results In total, 24 search queries in northern mainland China and 7 search queries in southern mainland China were found to be correlated and were used to construct the LSTM model, which included the same week and a lag of 1–4 weeks. The LSTM model showed that ILI% + mask with one lag week and ILI% + influenza name were good prediction modules, with reduced RMSE predictions of 16.75% and 4.20%, respectively, compared with the estimated ILI% for northern and southern mainland China. Conclusions The results illuminate the feasibility of using an internet search index as a complementary data source for influenza forecasting and the efficiency of using the LSTM model to forecast influenza epidemics.

Background: Seasonal influenza and other respiratory tract infections are serious public health problems that need to be further addressed and investigated. Internet search data are recognized as a valuable source for forecasting influenza or other respiratory tract infection epidemics. However, the selection of internet search data and the application of forecasting methods are important for improving forecasting accuracy. The aim of the present study was to forecast influenza epidemics based on the long short-term memory neural network (LSTM) method, Baidu search index data, and the influenza-like-illness (ILI) rate. Methods: The official weekly ILI% data for northern and southern mainland China were obtained from the Chinese Influenza Center from 2018 to 2021. Based on the Baidu Index, search indices related to influenza infection over the corresponding time period were obtained. Pearson correlation analysis was performed to explore the association between influenza-related search queries and the ILI% of southern and northern mainland China. The LSTM model was used to forecast the influenza epidemic within the same week and at lags of 1–4 weeks. The model performance was assessed by evaluation metrics, including the mean square error (MSE), root mean square error (RMSE) and mean absolute error (MAE). Results: In total, 24 search queries in northern mainland China and 7 search queries in southern mainland China were found to be correlated and were used to construct the LSTM model, which included the same week and a lag of 1–4 weeks. The LSTM model showed that ILI% + mask with one lag week and ILI% + influenza name were good prediction modules, with reduced RMSE predictions of 16.75% and 4.20%, respectively, compared with the estimated ILI% for northern and southern mainland China. Conclusions: The results illuminate the feasibility of using an internet search index as a complementary data source for influenza forecasting and the efficiency of using the LSTM model to forecast influenza epidemics. [ABSTRACT FROM AUTHOR]

Accurate freight volume forecasts play a crucial role in supporting regional economies and transportation systems. Nevertheless, the majority of previous studies rely on historical data, which poses challenges in promptly capturing the influence of market changes and emergency situations. To address this challenge, a novel prediction model based on the Baidu search index (BSI) and emergency is proposed, which comprises four distinct stages. In the first stage, we gathered the Baidu search index data and the COVID-19 index to furnish comprehensive, interactive, and timely information for the primary freight volume dataset. In the second stage, seasonal-trend decomposition procedures based on loess (STL) and empirical mode decomposition (EMD) are used to reduce data complexity. Additionally, we utilized boxplot to identify outliers in the remainder component. In the third stage, we employed seasonal naive prediction to forecast the seasonal component, while utilizing a genetic algorithm optimized backpropagation neural network (GABP) to predict other sub-sequences. To improve the prediction accuracy, we applied BPNN for error correction. Finally, to evaluate the performance of the proposed model, China's freight volume is selected as the subject of the study. The results demonstrate that the model incorporating BSI and COVID-19 information can effectively establish a dynamic forecasting system for freight volume. Furthermore, the data preprocessing techniques employed in this study successfully reduce data complexity and enhance the accuracy of prediction. In comparative experiments, the proposed model has been demonstrated to outperform all the contrasted models in terms of forecasting accuracy. [ABSTRACT FROM AUTHOR]

Introduction: With the advancement of society and technology, the importance of professional competency has been increasingly emphasized. This study aims to provide an in-depth analysis of the Baidu search index for “professional competency” from July 3, 2017, to October 6, 2023, across 34 provincial-level administrative regions in China.Methods: The study begins with a literature review to establish a theoretical framework. Subsequently, it delves into the fundamental concepts of complex networks, their parameters, and the principles of visible network graph algorithms. The application of system clustering and community detection algorithms is explored within this framework.Results: The empirical analysis starts with a provincial-level examination, involving the construction of visible network graphs, the exploration of scale-free networks, and discussions on regional system clustering. Following this, a nationwide analysis is conducted, focusing on the construction of visible network graphs to visualize the awareness of professional competency and studying community structures.Discussion: The findings provide insights into the awareness of professional competency across different regions. These insights are valuable for informing policy formulation and corporate decision-making, contributing to a deeper understanding of professional competency awareness and its implications.

It has been three years since the initial outbreak of COVID-19 in Wuhan, China, which incurred huge damage both physically and psychologically on human's normal life. As a prevention measure, the lockdown was first adopted by Wuhan, then by a long list of Chinese cities and many other major cities around the world. Lockdown is the most restrictive social distancing strategy, turning out effective in mitigating the spreading of COVID-19 on the community level, which, however, cuts off all social interactions and isolates healthy people from each other. The isolated nature of the lockdown could induce severe mental health issues, forming one major source of depression and domestic violence. Given the potential side effect, a comprehensive investigation based on reliable data sources is needed to evaluate the real psychological impact of COVID-19 lockdown and its evolution over time, particularly in the time when the Omicron variant, known for its low death risk, dominates the pandemic. Based on the Baidu Searching Index data collected for Wuhan and Shanghai, two major cities in China that suffered from long-lasting (over two months) lockdowns in 2020 and 2022, respectively, it is found that the major psychological issue during the lockdown period is not induced by the spreading of COVID-19, but by the execution of lockdown. With the deepening of knowledge about COVID-19 and the decrease in the death risk, the psychological impact of lockdown keeps increasing, while the impact of virus spreading becomes less important and even irrelevant to depression and domestic violence issues. The findings reveal that from the psychological perspective, the negative effect of lockdown already overweighs the positive one, which is especially true for the Omicron variant provided its almost ignorable death risk. Therefore, it is necessary to re-evaluate the yield and cost of lockdown for those countries where the COVID-19 pandemic has not yet come to an end. [ABSTRACT FROM AUTHOR]

Carbon price is not a "free-flowing" product of market transactions, but an economic symbol with political attributes that has a bearing on the endogenous security of China's national economic development. This study takes into account the key factors affecting carbon price, such as international carbon quota trading price, changes in macroeconomic environment, basic energy price and public concern about climate change. Based on the high-frequency, low-frequency, trend term decomposition and integrated noise reduction processing of carbon price data by DWT, CEEMDAN and SSA, the Transformers prediction model is constructed. Further, based on the combination of scenarios and the current status of power enterprises' revenues, a real-world carbon market "tolerance wall value" and a cut-off point are determined. The results show that (1) the predictions of the models using combined modal data are all better than those using unimodal and raw data. (2) The "wall value" of carbon price is 117.6 yuan/ton, 142.3 yuan/ton, and 182.7 yuan/ton under the low carbon follow (LCF) basic as usual (BAU), and economic follow (EF) scenarios, respectively, and all of them are reached by 2029. (3) Measured with realistic data, China's electric power enterprises have paid more than the value of profits (4 % of profits) with a low tolerance level to take part in carbon trading in the market. And under the EF and BAU scenarios, with high tolerance as a parameter, the trading price in China's carbon market will reach the wall value level (119.5yuan/ton) in June 2025 and October 2026, respectively. This research contributes to a more thorough understanding and forecasting of the function and future trajectory of China's carbon market and carbon pricing system. • Use Baidu search index to quantify climate change concerns. • Decomposition and reconstruction of multi-frequency carbon price data. • Determine the future value of the relevant elements based on scenario assumptions. • Calculate the upper limit of trading tolerance in the carbon market. • Integrated prediction based on heterogeneous structure information. [ABSTRACT FROM AUTHOR]

IntroductionScrub typhus, caused by Orientia tsutsugamushi, is a neglected tropical disease. The southern part of China is considered an important epidemic and conserved area of scrub typhus. Although a surveillance system has been established, the surveillance of scrub typhus is typically delayed or incomplete and cannot predict trends in morbidity. Internet search data intuitively expose the public's attention to certain diseases when used in the public health area, thus reflecting the prevalence of the diseases.MethodsIn this study, based on the Internet search big data and historical scrub typhus incidence data in Yunnan Province of China, the autoregressive integrated moving average (ARIMA) model and ARIMA with external variables (ARIMAX) model were constructed and compared to predict the scrub typhus incidence.ResultsThe results showed that the ARIMAX model produced a better outcome than the ARIMA model evaluated by various indexes and comparisons with the actual data.ConclusionsThe study demonstrates that Internet search big data can enhance the traditional surveillance system in monitoring and predicting the prevalence of scrub typhus and provides a potential tool for monitoring epidemic trends of scrub typhus and early warning of its outbreaks.

In the context of building a system of environmental pluralism and co-governance, it is of great signifi cance to systematically study the impact of public environmental concern on the environmental gover nance of enterprises and local governments, and provide theoretical support and policy suggestions f or guiding the public to actively participate in environmental governance. The study constructs a gam e model between the central government, local governments, enterprises and the public, analyzes the impact of public environmental concern on corporate environmental protection behavior and local go vernment environmental governance behavior, and uses the Baidu Smog Search Index from 2011 to 2 019 to analyze the impact of public environmental concern on enterprise environmental protection tec hnology investment and government environmental regulation intensity. The results show that public environmental concern has a positive impact on environmental governance through two channels: on the one hand, public environmental concern promotes local governments to increase the intensity of environmental regulations and investment in environmental protection infrastructure construction by reducing the degree of information asymmetry between the central government and local governmen ts and enterprises, curbing the collusion between local governments and enterprises, and ultimately e ncouraging enterprises to invest in environmental protection governance; On the other hand, the publ ic's environmental attention increases the market's investment demand for green products and enviro nmental protection governance enterprises, promotes the expansion of environmental governance ent erprises, and ultimately brings about the improvement of environmental quality. Theoretical and empi rical results show that public environmental concern has effectively played a binding role in informal e nvironmental regulations, encouraging polluting enterprises to transform environmental protection w hile effectively supervising local governments to strictly implement environmental regulations. The ce ntral government should pay attention to the public's environmental dissatisfaction; Establish and imp lement an environmental labeling system to meet the public's demand for green goods; Optimize the way of public environmental participation, and transform public environmental concern into public en vironmental participation. [ABSTRACT FROM AUTHOR]

Under the background of constructing a multiparty environment governance system, it is of great importance to systematically study the impact of public environmental attention on the environmental governance of enterprises and local governments. It could provide theoretical support and policy suggestions for guiding public participation in environmental governance. This paper constructed a game model among the central government, local governments, enterprises and the public to analyze the impact of public environmental attention on the game equilibrium of enterprises and local governments and empirically studied the impact of public environmental attention on enterprises' environmental investment and environmental regulation intensity using Baidu Haze Search Index from 2011 to 2019. We found that public environmental attention positively influenced environmental governance through two channels: firstly, public participation promoted local governments to improve environmental regulation intensity and environmental infrastructure construction and encouraged enterprises to invest in environmental governance by alleviating information asymmetry among game players; secondly, public participation increased the demand for green products and investments for environmental protection enterprises, which would bring about the improvement of environmental quality. The results showed that public environmental attention played an important role in environmental governance as an informal environmental regulation. The central government should pay attention to public environmental dissatisfaction, promote green labelling system to meet public demand for green products, and optimize the channels of public environmental participation. Besides, the government should implement differentiated guiding policies and incentive mechanisms based on the heterogeneous response of environmental governance to public environmental attention in regions with differentiated income levels, thus forming a more effective environment governance system with benign interaction between all the players. [ABSTRACT FROM AUTHOR]

Background: China exited strict Zero-COVID policy with a surge in Omicron variant infections in December 2022. Given China's pandemic policy and population immunity, employing Baidu Index (BDI) to analyze the evolving disease landscape and estimate the nationwide pneumonia hospitalizations in the post Zero COVID period, validated by hospital data, holds informative potential for future outbreaks., Methods: Retrospective observational analyses were conducted at the conclusion of the Zero-COVID policy, integrating internet search data alongside offline records. Methodologies employed were multidimensional, encompassing lagged Spearman correlation analysis, growth rate assessments, independent sample T-tests, Granger causality examinations, and Bayesian structural time series (BSTS) models for comprehensive data scrutiny., Results: Various diseases exhibited a notable upsurge in the BDI after the policy change, consistent with the broader trajectory of the COVID-19 pandemic. Robust connections emerged between COVID-19 and diverse health conditions, predominantly impacting the respiratory, circulatory, ophthalmological, and neurological domains. Notably, 34 diseases displayed a relatively high correlation (r > 0.5) with COVID-19. Among these, 12 exhibited a growth rate exceeding 50% post-policy transition, with myocarditis escalating by 1,708% and pneumonia by 1,332%. In these 34 diseases, causal relationships have been confirmed for 23 of them, while 28 garnered validation from hospital-based evidence. Notably, 19 diseases obtained concurrent validation from both Granger causality and hospital-based data. Finally, the BSTS models approximated approximately 4,332,655 inpatients diagnosed with pneumonia nationwide during the 2 months subsequent to the policy relaxation., Conclusion: This investigation elucidated substantial associations between COVID-19 and respiratory, circulatory, ophthalmological, and neurological disorders. The outcomes from comprehensive multi-dimensional cross-over studies notably augmented the robustness of our comprehension of COVID-19's disease spectrum, advocating for the prospective utility of internet-derived data. Our research highlights the potential of Internet behavior in predicting pandemic-related syndromes, emphasizing its importance for public health strategies, resource allocation, and preparedness for future outbreaks., Competing Interests: LX, ML, and YH were employed by Hybribio Medical Laboratory Group Ltd. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2024 Yu, Tan, Tang, Zhao, Ji, Lin, He, Gu, Liang, Wang, Chen, Yang, Xie, Wang, Liu, He, Chen, Wang, Wu, Zhao, Liu, Wang, Hao, Cen, Yao, Zhang, Liu, Lye, Hao, Wong and Cen.)

This paper analyzes the impacts of investor attention on the returns and volatility of commodity futures in China. Using online search volumes as a proxy for investor attention, we find that investor attention exhibits a positive contemporaneous relationship with returns and volatility. In addition, the online search variables are significant predictors of returns and volatility in the commodity futures markets. Moreover, as compared with personal computer searches, mobile searches have a more pronounced predictive effect on returns and volatility. Taken together, we suggest that investor attention can explain the concurrent price movement and variation in the commodity futures markets in China. [ABSTRACT FROM AUTHOR]

The existing index system for volatility forecasting only focuses on asset return series or historical volatility, and the prediction model cannot effectively describe the highly complex and nonlinear characteristics of the stock market. In this study, we construct an investor attention factor through a Baidu search index of antecedent keywords, and then combine other trading information such as the trading volume, trend indicator, quote change rate, etc., as input indicators, and finally employ the deep learning model via temporal convolutional networks (TCN) to forecast the volatility under high-frequency financial data. We found that the prediction accuracy of the TCN model with investor attention is better than those of the TCN model without investor attention, the traditional econometric model as the generalized autoregressive conditional heteroscedasticity (GARCH), the heterogeneous autoregressive model of realized volatility (HAR-RV), autoregressive fractionally integrated moving average (ARFIMA) models, and the long short-term memory (LSTM) model with investor attention. Compared with the traditional econometric models, the multi-step prediction results for the TCN model remain robust. Our findings provide a more accurate and robust method for volatility forecasting for big data and enrich the index system of volatility forecasting.

Background: Internet-derived data and the autoregressive integrated moving average (ARIMA) and ARIMA with explanatory variable (ARIMAX) models are extensively used for infectious disease surveillance. However, the effectiveness of the Baidu search index (BSI) in predicting the incidence of scarlet fever remains uncertain., Objective: Our objective was to investigate whether a low-cost BSI monitoring system could potentially function as a valuable complement to traditional scarlet fever surveillance in China., Methods: ARIMA and ARIMAX models were developed to predict the incidence of scarlet fever in China using data from the National Health Commission of the People's Republic of China between January 2011 and August 2022. The procedures included establishing a keyword database, keyword selection and filtering through Spearman rank correlation and cross-correlation analyses, construction of the scarlet fever comprehensive search index (CSI), modeling with the training sets, predicting with the testing sets, and comparing the prediction performances., Results: The average monthly incidence of scarlet fever was 4462.17 (SD 3011.75) cases, and annual incidence exhibited an upward trend until 2019. The keyword database contained 52 keywords, but only 6 highly relevant ones were selected for modeling. A high Spearman rank correlation was observed between the scarlet fever reported cases and the scarlet fever CSI (r s =0.881). We developed the ARIMA(4,0,0)(0,1,2) (12) model, and the ARIMA(4,0,0)(0,1,2) (12) + CSI (Lag=0) and ARIMAX(1,0,2)(2,0,0) (12) models were combined with the BSI. The 3 models had a good fit and passed the residuals Ljung-Box test. The ARIMA(4,0,0)(0,1,2) (12) , ARIMA(4,0,0)(0,1,2) (12) + CSI (Lag=0), and ARIMAX(1,0,2)(2,0,0) (12) models demonstrated favorable predictive capabilities, with mean absolute errors of 1692.16 (95% CI 584.88-2799.44), 1067.89 (95% CI 402.02-1733.76), and 639.75 (95% CI 188.12-1091.38), respectively; root mean squared errors of 2036.92 (95% CI 929.64-3144.20), 1224.92 (95% CI 559.04-1890.79), and 830.80 (95% CI 379.17-1282.43), respectively; and mean absolute percentage errors of 4.33% (95% CI 0.54%-8.13%), 3.36% (95% CI -0.24% to 6.96%), and 2.16% (95% CI -0.69% to 5.00%), respectively. The ARIMAX models outperformed the ARIMA models and had better prediction performances with smaller values., Conclusions: This study demonstrated that the BSI can be used for the early warning and prediction of scarlet fever, serving as a valuable supplement to traditional surveillance systems., (©Tingyan Luo, Jie Zhou, Jing Yang, Yulan Xie, Yiru Wei, Huanzhuo Mai, Dongjia Lu, Yuecong Yang, Ping Cui, Li Ye, Hao Liang, Jiegang Huang. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 30.10.2023.)

Background: Sexually transmitted diseases (STDs) are a serious issue worldwide. With the popularity of the internet, online health information-seeking behavior (OHISB) has been widely adopted to improve health and prevent disease., Objective: This study aimed to investigate the short-term and long-term effects of different types of OHISBs on STDs, including syphilis, gonorrhea, and AIDS due to HIV, based on the Baidu index., Methods: Multisource big data were collected, including case numbers of STDs, search queries based on the Baidu index, provincial total population, male-female ratio, the proportion of the population older than 65 years, gross regional domestic product (GRDP), and health institution number data in 2011-2018 in mainland China. We categorized OHISBs into 4 types: concept, symptoms, treatment, and prevention. Before and after controlling for socioeconomic and medical conditions, we applied multiple linear regression to analyze associations between the Baidu search index (BSI) and Baidu search rate (BSR) and STD case numbers. In addition, we compared the effects of 4 types of OHISBs and performed time lag cross-correlation analyses to investigate the long-term effect of OHISB., Results: The distributions of both STD case numbers and OHISBs presented variability. For case number, syphilis, and gonorrhea, cases were mainly distributed in southeastern and northwestern areas of China, while HIV/AIDS cases were mostly distributed in southwestern areas. For the search query, the eastern region had the highest BSI and BSR, while the western region had the lowest ones. For 4 types of OHISB for 3 diseases, the BSI was positively related to the case number, while the BSR was significantly negatively related to the case number (P<.05). Different categories of OHISB have different effects on STD case numbers. Searches for prevention tended to have a larger impact, while searches for treatment tended to have a smaller impact. Besides, due to the time lag effect, those impacts would increase over time., Conclusions: Our study validated the significant associations between 4 types of OHISBs and STD case numbers, and the impact of OHISBs on STDs became stronger over time. It may provide insights into how to use internet big data to better achieve disease surveillance and prevention goals., (©Xuan Li, Kun Tang. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 12.05.2023.)

Introduction: Scrub typhus, caused by Orientia tsutsugamushi , is a neglected tropical disease. The southern part of China is considered an important epidemic and conserved area of scrub typhus. Although a surveillance system has been established, the surveillance of scrub typhus is typically delayed or incomplete and cannot predict trends in morbidity. Internet search data intuitively expose the public's attention to certain diseases when used in the public health area, thus reflecting the prevalence of the diseases., Methods: In this study, based on the Internet search big data and historical scrub typhus incidence data in Yunnan Province of China, the autoregressive integrated moving average (ARIMA) model and ARIMA with external variables (ARIMAX) model were constructed and compared to predict the scrub typhus incidence., Results: The results showed that the ARIMAX model produced a better outcome than the ARIMA model evaluated by various indexes and comparisons with the actual data., Conclusions: The study demonstrates that Internet search big data can enhance the traditional surveillance system in monitoring and predicting the prevalence of scrub typhus and provides a potential tool for monitoring epidemic trends of scrub typhus and early warning of its outbreaks., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2022 Wang, Zhang, Lu, Lv, Wang, Zhu, Ai, Mao, Tan and Qi.)

This paper investigates the sleeplessness in Chinese cities during the coronavirus disease 2019 (COVID-19) pandemic. We provide first evidence of a link from daily COVID-19 cases resulting in sleep loss in a panel of Chinese cities. We use Wuhan, which was the first city to be completely locked down, as basis to present the result that sleeplessness has become a considerably serious issue owing to the COVID-19 pandemic. In using the intervention policy of various cities as exogenous shocks, we find that lockdown policies significantly increase the sleeplessness level of Chinese cities. In addition, the severity of COVID-19 pandemic significantly exacerbates the negative effect of lockdown policies on sleep quality in the city. Overall, this study indicates that policy makers should pay more attention to public mental health when citizens recover from COIVD-19 by investigating the unintended consequences of COVID-19 on sleeplessness level of cities., (© 2022 Elsevier Inc. All rights reserved.)

With the boom of big data, the Internet contains more and more personal behavior information, but it is difficult to extract effectively. A model involving multivariate processing capability must be constructed to deal with these time series with complex characteristics. In this paper, a novel hybrid model embedding Baidu Search Index is therefore proposed to implement multi-step ahead subway passenger flow forecasting. Firstly, we collect data from informative Baidu Search Index, reduce dimensionality, and screen out the powerful predictors by statistical analysis. Secondly, we extract matching common modes at similar time scales between the subway passenger flow and screened Baidu Search Index via multivariate mode decomposition being optimized by multi-objective algorithm. Furthermore, to eliminate pseudo statistical causality, we select the optimal combination of modal components between subway passenger flow and its corresponding Baidu Search Index at each time scale by an innovative multi–modal analysis strategy. Thirdly, we reconstruct the forecasting values of each selected optimal combination as the final results. The empirical results of Beijing, Shanghai and Guangzhou show that the proposed model can significantly outperform six benchmark models in both the level and directional accuracy. So introducing Baidu Search Index creates a sound opportunity to enhance the subway passenger flow forecasting ability. [ABSTRACT FROM AUTHOR]

With the approval of the vaccine in mainland China, concerns over its safety and efficacy emerged. Since the Chinese vaccine has been promoted by the Chinese government for months and got emergency approval from the World Health Organization. The Chinese vaccination program is yet to be identified from the perspective of local populations. The COVID-19 vaccine-related keywords for the period from January 2019 to April 2021 were examined and queried from the Baidu search index. The searching popularity, searching trend, demographic distributions and users' demand were analyzed. The first vaccine enquiry emerged on 25th January 2020, and 17 vaccination keywords were retrieved and with a total BSI value of 13,708,853. The average monthly searching trend growth is 21.05% ( p  < 0.05) and was led by people aged 20-29 (39.22%) years old. Over 54.93% of the demand term search were pandemic relevant, and the summed vaccine demand ratio was 44.79%. With the rising search population in COVID-19 vaccination, education programs and materials should be designed for teens and people above the 40 s. Also, vaccine-related birth safety should be alerted and further investigated., Competing Interests: Declaration of conflicting interests: The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article., (© The Author(s) 2022.)

• The effect of investor attention in online P2P lending is tested. • Investor attention decreases the P2P platform average projects' interest rate by 15%. • The investor attention effect is stronger in the low and middle attention level groups. • The result is robust after controlling for possible reverse causality. We focus on 224 big Peer-to-Peer(P2P) lending platforms from 2015 to 2019 to test whether investor attention affects the return in an online lending environment. As investors mainly obtain the platform information through Internet, we use Baidu search index as a proxy of investor attention. Empirical results show that investor attention will decrease the platform average projects' interest rate by 15%, and this effect is considerably stronger for middle attention level platforms. The result is robust after controlling for the endogeneity caused by possible reverse causality. [ABSTRACT FROM AUTHOR]

The existing index system for volatility forecasting only focuses on asset return series or historical volatility, and the prediction model cannot effectively describe the highly complex and nonlinear characteristics of the stock market. In this study, we construct an investor attention factor through a Baidu search index of antecedent keywords, and then combine other trading information such as the trading volume, trend indicator, quote change rate, etc., as input indicators, and finally employ the deep learning model via temporal convolutional networks (TCN) to forecast the volatility under high-frequency financial data. We found that the prediction accuracy of the TCN model with investor attention is better than those of the TCN model without investor attention, the traditional econometric model as the generalized autoregressive conditional heteroscedasticity (GARCH), the heterogeneous autoregressive model of realized volatility (HAR-RV), autoregressive fractionally integrated moving average (ARFIMA) models, and the long short-term memory (LSTM) model with investor attention. Compared with the traditional econometric models, the multi-step prediction results for the TCN model remain robust. Our findings provide a more accurate and robust method for volatility forecasting for big data and enrich the index system of volatility forecasting. [ABSTRACT FROM AUTHOR]

Chinese residents are becoming more and more concerned about the living environment especially under the situation of environmental degradation caused by the unbalanced and inadequate economic development. The widespread of internet use provide a new way for public to express the dissatisfaction on environmental pollution. Although the public is the main body of society, the public concern over environmental issues are rarely studied. In this paper, the impact of public concern over haze on haze pollution is quantitatively examined by the utilization of econometric model. Specifically, the Baidu search index (BSI) is utilized as indicators for public concern. Using the panel data consisting of 13 cities in Beijing-Tianjin-Hebei region from the period from January 2014 to December 2019, estimation results showed a significant improvement effect of public concern on haze pollution. In general, the public concern can improve the air quality in a short turn. However, this improvement effect varies with different economic development levels. These findings can help policy makers to better understand the role of public in social governance and improve the air quality in China with the inclusion of public participation. Unlabelled Image • Impact of public concern over haze on haze pollution are studied. • Big data of Baidu Search is utilized to indicate public concern. • Public concern can improve the haze pollution significantly in a short term. • Sustainable mitigation measures can be constructed with the inclusion of public participation. [ABSTRACT FROM AUTHOR]