Investigating the Spatial-Temporal Variation of Pre-Trip Searching in an Urban Agglomeration
Round 1
Reviewer 1 Report
I would suggest this manuscript is sent to journal which aims to developing big data analysis method instead of Sustainability journal. The reasons are explained by comparing the paragraphs underneath:
(1)
https://www.mdpi.com/journal/sustainability/sections/geography_sustainability Geography and Sustainability
The Section “Geography and Sustainability” aims to serve as the focal point for developing, coordinating and implementing interdisciplinary research and education to promote sustainable development through an integrated geography perspective. The Section encourages wider analysis and innovative thinking about global and regional sustainability by bridging and synthesising physical geography and human geography.
(2)
(line 12-17) The study brings several original results. First, tourist volume and SVI from source cities show distance attenuation. Second, SVI is a precursor to 13 changes in tourist volume. The precursory time rises with the increase of traffic time spatially. Third, we validated a VAR model and improved its accuracy by constructing it based on the spatial-temporal differentiation of search features. This is crucial for tourism departments to carry out short term precision marketing and provide personalized services.
(3)
(line 360-368) Cyberspace interacts with physical worlds on a spatial and material level [57]. On the internet, time and distance do not vanish but are recreated and reflected in geo-cyber space [58]. However, few attempts have been made to deduce the information search flow’s time-space features. Previous research has utilized this indicator to forecast destinations. However, this method ignores the tourism market’s internal spatial-temporal variation [50]. Using cellular signaling monitoring and search engine data from big data platforms, we investigated the temporal and spatial differentiation of visitors’ TIS. A time and accuracy-optimized approach was proposed for forecasting tourist volume from many sources.
Comment:
Number (2) and (3) above are not interrelated to (1)
Author Response
Point 1: I would suggest this manuscript is sent to journal which aims to developing big data analysis method instead of Sustainability journal. The reasons are explained by comparing the paragraphs underneath:
Response 1: We thank the reviewer for this suggestion. This study aims to investigate the relationship between online search behaviour and tourist arrivals through big data methods, which can help improve the prediction accuracy of daily tourist arrivals, thus providing theoretical support and management insights for destination managers to improve destination management and promote sustainable regional development. Therefore, the author believes that this study is in line with the theme of "Geography and Sustainability" and can be submitted to Sustainability journal.
Reviewer 2 Report
One reference (ref: 44) in red font. Please check and make sure to check all references are correctly cited.
Some figures are not aligned in the paragraph (e.g. Figure 4, and Figure 6) and most figures with no sources, and the data are adapted from other materials.
Please show some examples of the search engine in the literature and empirical result to give more overview of the result.
Empriral result and conclusion need to be improved in referring back to the figure that used in the paper.
English structuring of sentences needs to be improved.
Author Response
Point 1: One reference (ref: 44) in red font. Please check and make sure to check all references are correctly cited.
Response 1: Thanks for the reviewer's comments. We will further check and correct all literature citations.
Point 2: Some figures are not aligned in the paragraph (e.g. Figure 4, and Figure 6) and most figures with no sources, and the data are adapted from other materials.
Response 2: We thank the reviewer for pointing this out and we will add a description of the data source and make further corrections to the formatting of the article
Point3: Please show some examples of the search engine in the literature and empirical result to give more overview of the result.
Response 3: We thank the reviewer for pointing out this issue. We will strengthen the literature citation of the empirical results.
The prediction power of VAR and vector error correction (VEC) models, which reflects the causal relationship between the forecasting object and the influencing factor, is better than the traditional time-series forecasting model[48]. The prediction accuracy of VAR model is related to multiple parameters, including lag order, model selection criteria, covariance matrix of residuals, etc. It is necessary to select and adjust these parameters according to specific situations in order to improve the prediction ability of the model.
For example, Hu et al. adopted feature extraction technology, combined econometrics, integrated learning and hybrid methods to build models and optimized the performance of tourism prediction with field effect tubes[50].
First, tourist volume and SVI from source cities demonstrate attenuation due to distance. There is a direct correlation between tourist volume and SVI daily statistics. That is to say, SVI is predictive of changes in visitor numbers. According to Buhalis, search engines are the first choice for families to collect information when making travel plans, and SIVs can be used as an important indicator for destination prediction of tourist behavior[66]. Among the source area, tourists from major source cities are the main predictive of SVI.
Second, when traffic time increases spatially, the precursory time for tourists’ online searches increases. The SVI results on the lead time for changes in tourist flow can be interpreted as the time required for tourists to gather knowledge prior to traveling in order to acquire various search behavior characteristics based on spatial cognition. For example, tourist search query data about "hotels and flights" can significantly improve the accuracy of local tourist numbers forecasts[29]. Additionally, it demonstrates that travelers traveling across short distances spend less time preparing for information than long distances.
Point 4: Empirical result and conclusion need to be improved in referring back to the figure that used in the paper.
Response 4: We thank the reviewer for pointing out this issue. We will further refine the empirical findings and conclusions.
Cyberspace interacts with physical worlds on a spatial and material level[64]. On the internet, time and distance do not vanish but are recreated and reflected in geo-cyberspace[65]. This study examines online information search behavior by examining the spatial-temporal interactions between the pre-trip and post-arrival stages. The spatial-temporal link between and multivariate models is investigated. In comparison to a previous study on the characteristics of China’s domestic pre-tour searching behavior[48,54], this paper quantifies the length of tourist flow lags as distance increases for different cities ranging from 1 to 8. By investigating the spatial-temporal interactions between the pre-trip stage and the after-arrival stage, this paper uncovers the temporal and spatial deviations of pre-trip searching and validates their potential application in tourism forecasting, making an original contribution to the study of online information search behavior.
The contributions of this study are manifold. Predictability and interpretability have been the focus of this study. Firstly, we have attempted to integrate time and space into a tourism demand forecasting model. Few attempts have been made to deduce the infor-mation search flow’s time-space features. Previous research has utilized this indicator to forecast destinations. However, this method ignores the tourism market’s internal spa-tial-temporal variation [57]. Using cellular signaling monitoring and search engine data from big data platforms, we investigated the temporal and spatial differentiation of visitors’ TIS. A time and accuracy-optimized approach was proposed for forecasting tourist volume from many sources. Compared with the existing research results, the model constructed in this paper has wider usability.
Secondly, our research has proposed a new method for predicting daily visitor num-bers at tourist attractions based on accurate data from mobile roaming data and search engines. At the same time, we have integrated a variety of research methods, making the research results more general and scalable, and allowing for more accurate tourism fore-casting models. We used the Zhongshan Scenic Area in Nanjing as a case study to exam-ine daily tourist flow and SVI data in 40 Yangtze River Delta source cities over 640 days. The Granger causality test, the VAR model, and the impulse response function were used to determine the geographical distribution disparities between nearby source cities, web search data, and the spatiotemporal connection between tourist flow and SVI. Finally, us-ing these correlations, we used ARMA and VAR models to forecast tourist arrivals in a scenic location. We finish by saying the following:
First, tourist volume and SVI from source cities demonstrate attenuation due to distance. There is a direct correlation between tourist volume and SVI daily statistics. That is to say, SVI is predictive of changes in visitor numbers. According to Buhalis, search engines are the first choice for families to collect information when making travel plans, and SIVs can be used as an important indicator for destination prediction of tourist behavior[66]. Among the source area, tourists from major source cities are the main predictive of SVI.
Second, when traffic time increases spatially, the precursory time for tourists’ online searches increases. The SVI results on the lead time for changes in tourist flow can be interpreted as the time required for tourists to gather knowledge prior to traveling in order to acquire various search behavior characteristics based on spatial cognition. For example, tourist search query data about "hotels and flights" can significantly improve the accuracy of local tourist numbers forecasts[29]. Additionally, it demonstrates that travelers traveling across short distances spend less time preparing for information than long distances.
Finally, the prediction model developed using the spatial-temporal differentiation of search features exhibits high predictability. The forecasting model for scenic tourist flow that incorporates SVI is more accurate than the classic time series forecasting model that does not incorporate SVI. It is capable of estimating the inter-daily tourist volume, which is critical for monitoring and managing the tourist flow in the scenic site.
Thus, when marketing and promoting beautiful sites, enterprises can choose the optimal moment to spread their message based on tourists’ past search behavior patterns in various regions. Travel agencies and destination management organizations (DMOs) should close the gap between travel intention and actual arrival to alter their advertising efforts. These insights can be used to improve the accuracy of tourism advertisements and hence their efficacy. In terms of future research implications, we can select additional prediction models based on the particular spatial-temporal correlations between tourist volume and SVI at the city and daily scales, hence optimizing forecast accuracy rather than at the country and monthly scales. Additionally, a prospective tourist from a variety of cities may be interested in a variety of factors during their pre-trip research, such as sceneries or service facilities. Spatial and temporal differentiation is critical, a vital task that requires further investigation.
Point 5: Comments on the Quality of English Language: English structuring of sentences needs to be improved.
Thanks to the reviewer's comments on the language quality of the paper, we will further improve the English sentence structure
Reviewer 3 Report
Well Explained research paper, on 'Investigating the Spatial-Temporal Variation of Pre-Trip Searching in An Urban Agglomeration' with proper methodology and appropriate citations.
The objectives, methodology, and results are well aligned. Still in methodology, LPR could have been explained a little more. Explanations of ARMA and VAR are justified.
In Empirical results, the data collection period and a good comparison of search engines - Baidu and Google trends mentioned. The collection of daily tourist data in collaboration with the Telecom Operator shows the in-depth study of the research. Interesting results are drawn in the study area that tourists generally search for tourist information 2 to 8 days before arriving at a particular destination.
Conclusions are thoroughly supported by the results.
Researchers must study in the future, the longer-distance tourist flow and online search behavior.
Author Response
Piont：
Well Explained research paper, on 'Investigating the Spatial-Temporal Variation of Pre-Trip Searching in An Urban Agglomeration' with proper methodology and appropriate citations.
The objectives, methodology, and results are well aligned. Still in methodology, LPR could have been explained a little more. Explanations of ARMA and VAR are justified.
In Empirical results, the data collection period and a good comparison of search engines - Baidu and Google trends mentioned. The collection of daily tourist data in collaboration with the Telecom Operator shows the in-depth study of the research. Interesting results are drawn in the study area that tourists generally search for tourist information 2 to 8 days before arriving at a particular destination.
Conclusions are thoroughly supported by the results.
Researchers must study in the future, the longer-distance tourist flow and online search behavior.
Response: We thank reviewer very much for the recognition of our study and suggestions. We will further explain the meaning of LPR.
To investigate the influence of SVI on the spatial distribution characteristics of tourist flow in the precursory days, we examined the lag of impulse response peak (LRP) in the function graphs of various cities, which refers to the time delay between the point of application of an impulse input signal and the peak of the output response signal., and is an important parameter used to characterize the behavior of a system and design control algorithms. This study use LRP to indicate the most significant days for the number of tourist lags behind the SVI.
Reviewer 4 Report
This paper takes the spatio-temporal interaction before and after travel as the research object to explore the network information search behavior. The big data obtained through mobile roaming and search engines provide accurate data at both daytime and city scales, enabling this paper to study the relationship between daily tourist arrivals and pre-trip searches in 40 cities within the Yangtze River Delta urban agglomeration. The research shows that the distance between passenger flow and source city decreases with the index index of search volume. The index of search volume is the precursor of the change of the number of visitors; The precursor time increases spatially with the increase of traffic time.
However, the biggest problem of this study is what theoretical contribution it provides to related disciplines. The author needs to further elaborate on the theoretical contribution of the paper.
Author Response
Point:
This paper takes the spatio-temporal interaction before and after travel as the research object to explore the network information search behavior. The big data obtained through mobile roaming and search engines provide accurate data at both daytime and city scales, enabling this paper to study the relationship between daily tourist arrivals and pre-trip searches in 40 cities within the Yangtze River Delta urban agglomeration. The research shows that the distance between passenger flow and source city decreases with the index index of search volume. The index of search volume is the precursor of the change of the number of visitors; The precursor time increases spatially with the increase of traffic time.
However, the biggest problem of this study is what theoretical contribution it provides to related disciplines. The author needs to further elaborate on the theoretical contribution of the paper.
Response: We thank reviewer for the recognition of our study and the suggestions. We will further clarify the theoretical contribution of this study.
The contributions of this study are manifold. Firstly, we have attempted to integrate time and space into a tourism demand forecasting model. Few attempts have been made to deduce the information search flow’s time-space features. Previous research has utilized this indicator to forecast destinations. However, this method ignores the tourism market’s internal spatial-temporal variation [51]. Using cellular signaling monitoring and search engine data from big data platforms, we investigated the temporal and spatial differentiation of visitors’ TIS. A time and accuracy-optimized approach was proposed for forecasting tourist volume from many sources.
Secondly, our research has proposed a new method for predicting daily visitor numbers at tourist attractions based on accurate data from mobile roaming data and search engines. At the same time, we have integrated a variety of research methods, making the research results more general and scalable, and allowing for more accurate tourism forecasting models. We used the Zhongshan Scenic Area in Nanjing as a case study to examine daily tourist flow and SVI data in 40 Yangtze River Delta source cities over 640 days. The Granger causality test, the VAR model, and the impulse response function were used to determine the geographical distribution disparities between nearby source cities, web search data, and the spatiotemporal connection between tourist flow and SVI. Finally, using these correlations, we used ARMA and VAR models to forecast tourist arrivals in a scenic location.
Reviewer 5 Report
This study examined the spatial-temporal interactions between pre-trip and after-arrival online information search behavior from 40 cities in the Yangtze River Delta urban agglomeration.
The manuscript is well-written and presents respectable findings, however before its approval, it must be improved in the following ways:
In the abstract section, the author should include the quantitative results.
The introduction could be expanded, and more related research sources should be cited.
The author should define all abbreviations before using them even if they are well known.
The author should give a brief overview of the VAR model, Granger causality test, and ARMA prediction model.
The author should mention the parameters that can affect the prediction accuracy of a VAR model.
The Granger causality test can increase the precision of a VAR model but can also result in false positives. What do you do to fix it?
The author should mention the name of the software/tools used for spatial analysis and visualization of maps.
The quality of the English Language is good, There is a typo errors in many places, and the author should correct them.
Author Response
Point 1: In the abstract section, the author should include the quantitative results.
Response 1: Thanks to the reviewer for pointing this issue out. We will add the quantitative results in the abstract section.
First, tourists generally search for tourist information 2–8 days before arriving at destinations, while tourist volume and SVI from source cities show distance attenuation.
Point 2: The introduction could be expanded, and more related research sources should be cited.
Response2: Thanks to the reviewer for pointing this out. We will further enrich the introduction by citing more relevant studies.
Tourism demand forecasting is of great significance in the tourism industry, as it could help tourism practitioners and tourism managers to better formulate tourism plans, resource allocation and marketing strategies to promote sustainable development of the destination[10,11]. At the same time, forecasting tourism demand can also help tourism enterprises and government departments to better respond to changes in the tourism mar-ket and improve the quality and efficiency of tourism services[12], thereby promoting the sustainable development of tourism.
Traditional tourism demand forecasting methods include time series analysis[13], re-gression analysis[14] and neural networks[15]. In recent years, with the development of big data technology[12] and artificial intelligence technology[16], tourism demand fore-casting has also gradually developed towards data-driven and intelligent direction.
Point 3: The author should define all abbreviations before using them even if they are well known.
Response 3: Thanks to the reviewer for pointing out this issue. We will make all definitions abbreviated.
Furthermore, as a statistical method for testing causality between time series, this study used Granger causality tests to investigate the relationship between tourist volume and SVI in each city and demonstrate the causal relationship in the temporal dimension. Finally, the impulse response function was utilized to examine the dynamic influence of SVI on the tourism flow in each city by creating the Vector Autoregression Model (VAR) with each city’s tourist arrivals and SVI data. VAR is a multivariate time series analysis model, which is used to describe the linkage relationship between multiple variables.
To investigate the influence of SVI on the spatial distribution characteristics of tourist flow in the precursory days, we examined the lag of impulse response peak (LRP) in the function graphs of various cities, which refers to the time delay between the point of application of an impulse input signal and the peak of the output response signal, and is an important parameter used to characterize the behavior of a system and design control algorithms. This study use LRP to indicate the most significant days for the number of tourist lags behind the SVI.
We used historical data from the Yangtze River Delta’s tourist flow time series to anticipate tourist flow using the Autoregressive Moving Average Model (ARMA), which represents time series data as a linear combination of autoregressive terms and moving average terms, taking into account both the historical data of the time series itself and the influence of random error terms.
Point 4: The author should give a brief overview of the VAR model, Granger causality test, and ARMA prediction model.
Response 4: Thanks to the reviewer for pointing out this issue. We will present the VAR model, the Grange causality test and the ARMA prediction model.
Furthermore, as a statistical method for testing causality between time series, this study used Granger causality tests to investigate the relationship between tourist volume and SVI in each city and demonstrate the causal relationship in the temporal dimension. Fi-nally, the impulse response function was utilized to examine the dynamic influence of SVI on the tourism flow in each city by creating the Vector Autoregression Model (VAR) with each city’s tourist arrivals and SVI data. VAR is a multivariate time series analysis model, which is used to describe the linkage relationship between multiple variables.
Then, we used historical data from the Yangtze River Delta’s tourist flow time series to an-ticipate tourist flow using the Autoregressive Moving Average Model (ARMA), which rep-resents time series data as a linear combination of autoregressive terms and moving aver-age terms, taking into account both the historical data of the time series itself and the in-fluence of random error terms.
Point 5: The author should mention the parameters that can affect the prediction accuracy of a VAR model.
Response 5: Thanks to the reviewer for pointing out this issue. We will explain the parameters that affect the predictive accuracy of the VAR model.
The prediction accuracy of VAR models is related to several parameters, including lag order, model selection criteria, residual covariance matrix, etc. In order to improve the predictive ability of the model, it is necessary to select and adjust these parameters according to specific circumstances.
Point 6: The Granger causality test can increase the precision of a VAR model but can also result in false positives. What do you do to fix it?
Response 6: Thanks to the reviewer for pointing this out.
Granger causality test (GCT) is a method used to test causal relationships between time series and can be used to improve the prediction accuracy of VAR models. However, the Granger causality test also has a problem of false positives, where the test results may erroneously assume the existence of causal relationships. To address this problem, this study adopted the following methods: 1. To solve the problem of false positives, we used various comparison correction methods such as Bonferroni correction and Holm's correction. These methods can correct test results and reduce the risk of false positives. 2. We used a more stringent significance level to lower the probability of false positives.
Point 7: The author should mention the name of the software/tools used for spatial analysis and visualization of maps.
Response 7: Thanks to the reviewer for pointing this out. The names of the software/tools used will be explained.
The above methods primarily use Arcgis and Stata for data analysis.
Point 8: Comments on the Quality of English Language: The quality of the English Language is good, There is a typo errors in many places, and the author should correct them.
Response 8: Thanks to the reviewer for raising this issue. We will further correct any language errors in the article.
Round 2
Reviewer 2 Report
Improvement have been made according to the comments.
Author Response
We would like to thank Reviewer 2 for the positive feedback and affirmation of our work. We are glad that our research has been well-received. We would like to express our sincere gratitude to the editorial team and the reviewers for their recognition of our efforts and for giving us this opportunity to further enhance our work.
