Mobile and Location Based Services

In einem einwöchigen Seminar unter der Leitung von Prof. Anto Aasa tauchte ich tief in die faszinierende Welt der "Mobile & Location Based Services" ein. Das Seminar bot eine umfangreiche Vorlesung über mobile und standortbezogene Dienste. Meine Hauptaufgabe nach dem Seminar bestand darin, einen Aufsatz zum Thema "Datenschutz in Deutschland (Europa) und die Entwicklung von LBS" zu verfassen.

Für die praktische Komponente habe ich zwei Projekte durchgeführt, für die ich intaraktive Visualisierungen in Python implementiert habe:

1. Analyse der ausländischen Besucher in Estland: Auf der Grundlage passiver mobiler Positionsbestimmung entwarf ich eine
interaktive Karte mit Plotly Scattermapbox, die einen Einblick in die Bewegungen und Besuchsmuster ausländischer Besucher in
Estland bietet. Der Clou an diesem Projekt war, dass die Länder aus Gründen des Datenschutzes mit Namen von Wildtieren
dargestellt wurden.

2. Analyse von Touristenreisen auf der Grundlage von GPS-Daten: Anhand von Daten aus der Android-App "MobilityLog" analysierte
ich die Bewegungsmuster eines Touristen in Kalifornien. Die daraus resultierende Karte zeigte die Reiseroute und wies auf
Übernachtungen, Essensstopps und mehr hin. Die Verarbeitung war aufgrund von Unstimmigkeiten im Datensatz nicht ganz
einfach, aber die Überwindung dieser Probleme machte die Ergebnisse umso lohnender.

Meine gesamte analytische Arbeit und die daraus resultierenden Visualisierungen werden auf dieser Seite vorgestellt. Schauen Sie sich die interaktiven Karten an und vertiefen Sie sich in die Muster, die Dichte und die einzigartigen Nutzererfahrungen aus beiden Projekten.

Link to my code on github: https://github.com/Leonieen/MLBS

Python

Keywords:

Data Analyticss

Plotly

Data visualization

Estonia MLBS

Video suchen...

Estonia_UniqueUser

00:53

Video abspielen

Estonia_Time

01:06

Video abspielen

Estonia_MapFilters

00:24

Video abspielen

Municipalities with the most activities and those with the most unique users

The table lists the 5 municipalities with the most activities and those with the most unique users.

Density of calling activities

The heat map also illustrates the density of calling activities. In the future, a few improvements should be made here, due to the enormous density in Tallinn, which makes a gradation very unclear. This and the following visualisations were also implemented with Plotly in Python.

Map Filters

Interacative Map with Filters: The map was created with a Plotly Scattermapbox and gets updated by a Dash callback. Via the filters certain spatial patterns can be explored. It seems that the least visitors are from Wolf and the month with most visits is July. Most Visitors are calling from Tallinn or Tartu. The density in the large cities is also particularly visible in the animation, also created with a Scattermapbox, as well as an increased density on the border with Russia.

Animated Map

Interacative Map with Filters: The map was created with a Plotly Scattermapbox and gets updated by a Dash callback. Via the filters certain spatial patterns can be explored. It seems that the least visitors are from Wolf and the month with most visits is July. Most Visitors are calling from Tallinn or Tartu. The density in the large cities is also particularly visible in the animation, also created with a Scattermapbox, as well as an increased density on the border with Russia.

Calling Activities

The figure shows the calling density of the municipalities. A drop-down menu can also be used to distinguish between the individual tourist countries in order to determine any differences.
Overall, most calling activities take place on the northern Baltic Sea coast. A similar picture emerges from the analysis of the foxes. In the case of bears, there is also increased activity on the border with Russia and in the case of wolves in the municipality with the number 392 on Hiiumaa.

Hexbin map calling activities

Since the previously created heat map was not entirely satisfactory, a Hexbin map was also implemented to visualise a better gradation. For this purpose, the range of activities based on quantiles was lowered in order to still illustrate the high density in Tallinn, but also to show a gradual gradation of the other regions. This was additionally animated to include temporal changes, such as increased activity in July.

Hexbin map calling activities animated

Since the previously created heat map was not entirely satisfactory, a Hexbin map was also implemented to visualise a better gradation. For this purpose, the range of activities based on quantiles was lowered in order to still illustrate the high density in Tallinn, but also to show a gradual gradation of the other regions. This was additionally animated to include temporal changes, such as increased activity in July.

Time analysis

To get an even better impression of the calling activities of the individual countries in the given time frame, they were listed individually in bar charts. Particularly noticeable here are Wednesdays in the wolves, where there seem to have been a particularly large number of calls. Beforehand, all time data were split into the corresponding classifications in Python. A particularly large number of callers were present and made calls in July and August. The temporal patterns of the wolves and their frequency, however, do not quite coincide with the spatial patterns.

Time analysis

To get an even better impression of the calling activities of the individual countries in the given time frame, they were listed individually in bar charts. Particularly noticeable here are Wednesdays in the wolves, where there seem to have been a particularly large number of calls. Beforehand, all time data were split into the corresponding classifications in Python. A particularly large number of callers were present and made calls in July and August. The temporal patterns of the wolves and their frequency, however, do not quite coincide with the spatial patterns.

Time analysis

To get an even better impression of the calling activities of the individual countries in the given time frame, they were listed individually in bar charts. Particularly noticeable here are Wednesdays in the wolves, where there seem to have been a particularly large number of calls. Beforehand, all time data were split into the corresponding classifications in Python. A particularly large number of callers were present and made calls in July and August. The temporal patterns of the wolves and their frequency, however, do not quite coincide with the spatial patterns.

Time analysis

To get an even better impression of the calling activities of the individual countries in the given time frame, they were listed individually in bar charts. Particularly noticeable here are Wednesdays in the wolves, where there seem to have been a particularly large number of calls. Beforehand, all time data were split into the corresponding classifications in Python. A particularly large number of callers were present and made calls in July and August. The temporal patterns of the wolves and their frequency, however, do not quite coincide with the spatial patterns.

Time analysis

To get an even better impression of the calling activities of the individual countries in the given time frame, they were listed individually in bar charts. Particularly noticeable here are Wednesdays in the wolves, where there seem to have been a particularly large number of calls. Beforehand, all time data were split into the corresponding classifications in Python. A particularly large number of callers were present and made calls in July and August. The temporal patterns of the wolves and their frequency, however, do not quite coincide with the spatial patterns.

Time analysis

To get an even better impression of the calling activities of the individual countries in the given time frame, they were listed individually in bar charts. Particularly noticeable here are Wednesdays in the wolves, where there seem to have been a particularly large number of calls. Beforehand, all time data were split into the corresponding classifications in Python. A particularly large number of callers were present and made calls in July and August. The temporal patterns of the wolves and their frequency, however, do not quite coincide with the spatial patterns.

Calling activities of unique users

The same implementation goes for the exploration of unique users. The user can select a specific user id and in the first map the calling places are plotted and connected with lines. In the animation the markers are not linked with lines, but the temporal, chronological sequence is clearly visible.

Calling activities of unique users

The same implementation goes for the exploration of unique users. The user can select a specific user id and in the first map the calling places are plotted and connected with lines. In the animation the markers are not linked with lines, but the temporal, chronological sequence is clearly visible.

Calling Activities

The figure shows the calling density of the municipalities. A drop-down menu can also be used to distinguish between the individual tourist countries in order to determine any differences.
Overall, most calling activities take place on the northern Baltic Sea coast. A similar picture emerges from the analysis of the foxes. In the case of bears, there is also increased activity on the border with Russia and in the case of wolves in the municipality with the number 392 on Hiiumaa.

Calling Activities

The figure shows the calling density of the municipalities. A drop-down menu can also be used to distinguish between the individual tourist countries in order to determine any differences.
Overall, most calling activities take place on the northern Baltic Sea coast. A similar picture emerges from the analysis of the foxes. In the case of bears, there is also increased activity on the border with Russia and in the case of wolves in the municipality with the number 392 on Hiiumaa.

Map with the atennas

Interacative Map with Filters: The map was created with a Plotly Scattermapbox and gets updated by a Dash callback. Via the filters certain spatial patterns can be explored. It seems that the least visitors are from Wolf and the month with most visits is July. Most Visitors are calling from Tallinn or Tartu. The density in the large cities is also particularly visible in the animation, also created with a Scattermapbox, as well as an increased density on the border with Russia.

Calling Activities

The figure shows the calling density of the municipalities. A drop-down menu can also be used to distinguish between the individual tourist countries in order to determine any differences.
Overall, most calling activities take place on the northern Baltic Sea coast. A similar picture emerges from the analysis of the foxes. In the case of bears, there is also increased activity on the border with Russia and in the case of wolves in the municipality with the number 392 on Hiiumaa.

GPS MLBS

Video suchen...

GPS_ThematicMap

01:09

Video abspielen

GPS_Days_Route

00:52

Video abspielen

GPS_Mileage

00:07

Video abspielen

3D visualization of the route

To get a feeling for the spatial nature of the route, I plotted the GPS points in a 3D scatterplot and added the floor plan of California for spatial location. According to my internet research, California is on average 884m above sea level, which is why I also placed the data from the geojson there in the scatterplot.

I also tried to identify temporal patterns for this data set. This turned out to be rather not so purposeful. In the data set of the previous project, there were only points and time data, if an activity was also committed, here continuous data.
These visualizations were nevertheless included to see whether there are certain temporal trends or whether, for example, lunch breaks or overnight stays also become visible.
However, this visualization method is not without problems for the analysis of this question. For example, all data in the dataset were previously thinned where there were duplicate time entries, these could significantly affect the results.
Similarly, the frequency of GPS data is probably not the best measure, but rather the distances traveled between time points.

3D visualization of the route

To get a feeling for the spatial nature of the route, I plotted the GPS points in a 3D scatterplot and added the floor plan of California for spatial location. According to my internet research, California is on average 884m above sea level, which is why I also placed the data from the geojson there in the scatterplot.

3D visualization of the route

To get a feeling for the spatial nature of the route, I plotted the GPS points in a 3D scatterplot and added the floor plan of California for spatial location. According to my internet research, California is on average 884m above sea level, which is why I also placed the data from the geojson there in the scatterplot.

Daily mileage

The daily mileage of the trips has been calculated with geopy.distance from the geodesic package. For this a list of tuples of the Latitude and Longitude columns has been created. The distance between two GPS points has been calculated with geodesic(point1, point2).miles Then in a for-loop the total distance traveled in a day has been calculated

Daily Routes

Unfortunately the dataset showed some difficulties for the processing in Python. First, there were incorrectly displayed values in the timestamps, which is why the dataset could not be read in at first. After correcting this data, the commas for example in the coordinates were replaced by dots, and the separators of the dataset, originally tabs, were replaced by semicolons for better readability in Python. Here, however, difficulties arose with the time columns. Here the dates were displayed in different formats within one column. Some of the dates were in the format year-month-day or all may values were in the format year-day-month, which had to be corrected and adjusted with appropriate queries and loops in Python. However, this always led to problematic classifications. In the map shown below, the individual days can be selected and the corresponding route section is displayed, but here there can be inconsistencies due to the date problems already mentioned.

Thematic Map

Thematic Map: In the interactive map, the route was first displayed and, due to the large amount of data, all time points that occur more than once were removed to improve performance. It was noticeable that my route looks slightly different from the examples provided. Based on the time points and location properties, overnight stops, restaurants and shopping stops could be identified. In addition, markers were added to special sights and places.

Daily mileage

The daily mileage of the trips has been calculated with geopy.distance from the geodesic package. For this a list of tuples of the Latitude and Longitude columns has been created. The distance between two GPS points has been calculated with geodesic(point1, point2).miles Then in a for-loop the total distance traveled in a day has been calculated