=Paper=
{{Paper
|id=Vol-2855/main_short_3
|storemode=property
|title=An Interactive Dashboard for Traveler Mobility Analysis
|pdfUrl=https://ceur-ws.org/Vol-2855/main_short_3.pdf
|volume=Vol-2855
|authors=Lukas Vorwerk,Linus W. Dietz
|dblpUrl=https://dblp.org/rec/conf/wsdm/VorwerkD21
}}
==An Interactive Dashboard for Traveler Mobility Analysis==
https://ceur-ws.org/Vol-2855/main_short_3.pdf
ACM WSDM WebTour 2021, March 12th, 2021 Jerusalem, Israel 13
An Interactive Dashboard for Traveler Mobility Analysis
Lukas Vorwerk Linus W. Dietz
Department of Informatics Department of Informatics
Technical University of Munich Technical University of Munich
Garching, Germany Garching, Germany
lukas.vorwerk@tum.de linus.dietz@tum.de
ABSTRACT in global traveling behavior and point out relevant use cases and
Showcasing research in data mining is a challenging topic. Devel- applications.
oped models of everyday phenomena are often only understandable
by domain experts which might lead to low general adoption. We 2 TWITTER MOBILITY DATA SET
demonstrate an interactive dashboard to visualize the complex do- Mobility data about individuals is valuable, since it can be used for
main of international travel over time. Given that spatio-temporal many purposes and it is quite complicated to collect it on a larger
phenomena such as mobility can not effectively visualized using scale. Most data sets are limited to national boundaries, since tradi-
consumer software, we developed a web-based system, where users tional methods of collection, such as mobile phone communication
can explore how the behavior of international and domestic travel- records [4] or national statistics are tied to the respective administra-
ers changes over time and easily create their own analyses. tive regions. With the increasing adoption of online location-based
social networks (LBSNs), mobility traces of humans have become
KEYWORDS more widely available and this data has the advantage that it is
Tourist mobility, Data mining, Visualization, Dashboards typically not impeded by traveling to foreign countries. Twitter
is especially useful for researchers, since the service is inherently
1 INTRODUCTION AND RELATED WORK public and it provides official APIs to query user timelines.
Whenever a user tweets with the geolocation option enabled,
Human mobility analysis is a widely researched topic with many her approximate location is attached to the message. The downside
facets such as next location prediction [10], discovering recurring of LBSN mobility data is that it gives an incomplete picture of the
activity patterns [7], determining different traveler types [2], or user’s mobility — the location is only known when she decides to
deriving touristic travel regions [3]. An important problem of all tweet. This makes it necessary to assess the quality of the data with
research efforts concerning human mobility is the visualization of respect to the use case, eventually, discarding trips of insufficient
the results. To establish an understanding of the resulting mod- data quality.
els, visualizations are oftentimes very selectively included into the
manuscripts as plots. This, however, only shows a part of the find-
Table 1: Scope of the underlying Twitter mobility data set (as
ings and it is hard for practitioners in the tourism industry to utilize
of January 27, 2021)
the academic findings to improve their services. To improve upon
this, we have developed an interactive traveler mobility dashboard
that visualizes patterns of global mobility. The implementation as a Feature Value
web application offers users the flexibility to create their very own Tweets with geolocation 16,551,608
analyses. Users 302,747
The underlying data stems from a self-collected data set from International Travelers 128,539
Twitter. Due to that most content of Twitter is available for the Domestic Travelers 174,208
general public, it has been frequently used as a source for re- Trips 1,788,432
searching human mobility [5, 6]. We employed our earlier ap- International Trips 395,963
proaches to mine trips from geotagged tweets [1, 3] and aggre- Domestic Trips 1,392,469
gate the metrics into an interactive dashboard available at http: Countries visited 241
//mobility-dashboard.cm.in.tum.de. Observation period 2015 – 2021
Dashboards provide users with a customizable overview about
the state of a system without requiring them to work with the raw
data and using programming techniques. Thus, they are an ideal Since 2018, we collected timelines of Twitter users and using the
tool for the general public to explore complex topics such as stock Python tripmining library3 , we segment geolocated tweets of a
markets, misinformation on social media1 or global pandemics2 . user’s timeline into periods of being at home and on travel. If a user
In this position paper, we present our initial prototype for a tweets at a relatively constant pace, this data can be quite reliable,
traveler mobility dashboard based on mobility data derived from which is also one of the of the library’s features to assess using
social media. We present the motivations and design considerations various metrics.
to build a dashboard that can visualize analyses concerning trends Table 1 summarizes the statistics of our Twitter trips data set
used for the dashboard. As the data collection is ongoing, these
1 http://csmr.umich.edu/projects/iffy-quotient/ [8]
2 https://coronavirus.jhu.edu/map.html 3 https://github.com/LinusDietz/tripmining
Copyright © 2021 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
ACM WSDM WebTour 2021, March 12th, 2021 Jerusalem, Israel 14
Lukas Vorwerk and Linus W. Dietz
Figure 1: Screenshot of the Mobility Dashboard. See http://mobility-dashboard.cm.in.tum.de
numbers are constantly growing. While only parts of the data set specific chart areas. To the best of our knowledge, this platform is
are currently publicly available4 , we are able to provide mobility the first offering an interactive analysis of global travel trends.
data to interested researchers on a trip-level.
3 MOBILITY DASHBOARD
The dashboard consists of three interactive visualizations that invite
users to engage with the data and create mobility analyses on their
own. The central feature is a choropleth world map that reflects the
number of inbound or outbound trips on a country-level basis. The
exact number is shown when hovering over the respective area on
the map. By using a toggle switch, the user can decide whether to
display inbound or outbound trips. Upon clicking on the area of
a specific country, users are presented with a detailed analysis of
this country. They can now determine where the residents of this
country prefer to travel to and from which other countries travelers
visit this country in which quantities. Figure 1 shows the number
of outbound trips in our database undertaken since 2015.
Besides the map, a list containing the top ten countries for the
selected scenario is displayed. An additional visualization in the
form of a line chart displays the mean and median trip duration per
Figure 2: Inbound Trips to the United Kingdom in 2019.
year or month. To provide insights into the data the visualizations
are based on, the dashboard also includes the statistics shown in
Table 1 and a chart that indicates the number of trips per year or
month (see Figure 2). Furthermore, a time slider offers the possibility 4 LIMITATIONS AND USE CASES
to uncover temporal trends over the last decade. All of the described Given the intrinsic properties of the data, some analyses work better
dashboard elements are interconnected, i.e., if the user selects a than others using the dashboard. What works quite well is to get an
specific country or year, all visualizations are updated accordingly. indication of the popularity of a destination country as well as the
The dashboard is developed using Dash5 , a “Python framework popular travel seasons aggregated on a monthly basis. Certainly, the
for building web analytic applications”. The usage of Dash enables raw numbers have little meaning, since they only account for the
further interactions with the visualizations such as zooming in on mobility of Twitter users. The relative popularity of countries in the
4 https://github.com/LinusDietz/JITT2020-Mining-Trips-Replication [3] Western World, however, should be meaningful. On the contrary,
5 https://dash.plotly.com/introduction the numbers for countries where Twitter is not much adopted, e.g.,
Copyright © 2021 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
ACM WSDM WebTour 2021, March 12th, 2021 Jerusalem, Israel 15
An Interactive Dashboard for Traveler Mobility Analysis
due to state censorship, are certainly not representative. Further- in America. Finally, we plan to develop further analyses, e.g., to
more, the overall amount of data might be enormous (cf. Table 1), reveal which countries have often been traveled to together.
but when subdividing it into 7 years (currently about 73 months),
241 countries, and inbound and outbound trips, the amount of data REFERENCES
points can become quite low, especially for small countries. Finally, [1] Linus W. Dietz, Daniel Herzog, and Wolfgang Wörndl. 2018. Deriving Tourist
Mobility Patterns from Check-in Data. In WSDM Workshop on Learning from
it should be noted that the data collection is slower than its creation, User Interactions. Los Angeles, CA, USA.
and it takes time until enough recent trips are collected. For this [2] Linus W. Dietz, Rinita Roy, and Wolfgang Wörndl. 2018. Characterisation of
reason, the most recent data is the most incomplete, since trips that Traveller Types Using Check-in Data from Location-Based Social Networks. In
Information and Communication Technologies in Tourism, Juho Pesonen and Julia
are currently occurring can not yet be included in the data set. Neidhardt (Eds.). Springer, Cham, 15–26.
Therefore, the target audience of the dashboard is rather the [3] Linus W. Dietz, Avradip Sen, Rinita Roy, and Wolfgang Wörndl. 2020. Min-
general public than researchers. The number of possible analyses is ing Trips from Location-Based Social Networks for Clustering Travelers and
Destinations. Information Technology & Tourism 22, 1 (March 2020), 131–166.
intentionally limited at the moment, since we believe that increasing https://doi.org/10.1007/s40558-020-00170-6
complexity would hinder layman to interact with the dashboard to [4] Marta C. González, César A. Hidalgo, and Albert-László Barabási. 2008. Un-
derstanding individual human mobility patterns. Nature 453, 7196 (June 2008),
produce their own simple, but useful analyses. 779–782. https://doi.org/10.1038/nature06958
[5] Bartosz Hawelka, Izabela Sitko, Euro Beinat, Stanislav Sobolevsky, Pavlos Kaza-
5 CONCLUSIONS AND FUTURE WORK kopoulos, and Carlo Ratti. 2014. Geo-located Twitter as proxy for global mobility
patterns. Cartography and Geographic Information Science 41, 3 (Feb. 2014),
This paper described the interactive visualization of international 260–271. https://doi.org/10.1080/15230406.2014.890072
traveler mobility from Twitter. The underlying data has been col- [6] Raja Jurdak, Kun Zhao, Jiajun Liu, Maurice AbouJaoude, Mark Cameron, and
David Newth. 2015. Understanding Human Mobility from Twitter. PLOS ONE 10,
lected over the last year and shows the high value of LBSN data 7 (July 2015). https://doi.org/10.1371/journal.pone.0131469
for such analyses. With the presented dashboard, we provide a tool [7] Anastasios Noulas, Salvatore Scellato, Cecilia Mascolo, and Massimiliano Pontil.
that can be used to perform own analyses on the complex domain 2011. An empirical study of geographic user activity patterns in Foursquare. In
Fifth International Conference on Weblogs and Social Media (ICWSM ’11). AAAI,
of international travel. Currently, the dashboard allows to explore Palo Alto, CA, USA, 570–573.
the temporal variability of the number of inbound and outbound [8] Paul Resnick, Aviv Ovadya, and Garlin Gilchrist. 2019. Iffy quotient: A platform
health metric for misinformation. Technical Report. School of Information Center
trips of Twitter users. for Social Media Responsability University of Michigan.
In the future, we plan to improve the granularity of the analysis [9] Avradip Sen and Linus W. Dietz. 2019. Identifying Travel Regions Using Location-
into smaller regions, such as federal states or touristic regions [9]. Based Social Network Check-in Data. Frontiers in Big Data 2, 12 (June 2019).
https://doi.org/10.3389/fdata.2019.00012
Furthermore, it should be possible to set the timeframe to custom [10] Weimin Zheng, Xiaoting Huang, and Yuan Li. 2017. Understanding the tourist
dates, to be able to observe seasonal trends e.g., introduced by school mobility using GPS: Where is the next place? Tourism Management 59 (April
2017), 267–280. https://doi.org/10.1016/j.tourman.2016.08.009
breaks or holidays associated with traveling, such as Thanksgiving
Copyright © 2021 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).