Network analysis based on institutions
The following example will explain how to create a network of different institutions by using a world map and how to feed different parameter into the VOSviewer to recieve the desired map.
1. Create a concept: What should be displayed by the map?
A concept is needed to be able to generate the required data. The following questions can be helpful for this and are answered exemplary:
What should be displayed?
- Visualization of the (fictious) international relationships of RWTH Aachen by using a worldmap.
Which basis is required?
- Name of the institutions, their coordiantes and the amount of publications in cooperation with RWTH Aachen
- Basis of the worldmap
- Network file
Which balls should be connected?
- This example may illustrate the (fictious) cooperations of RWTH Aachen by using the VOSviewer, therefore the different institutions will be visually connected to RWTH. Of course, the different institutions can also be connected with one another by using the same principle. In the following example, institution A and B will be linked to express a relationship (ID 1 and 2).
How are the balls weighted?
- The balls are weighted by the number ob publications in cooperation with RWTH.
How should which balls be coloured?
- Groups can be assigned to different clusters and adjusted in their colour.
- Cluster 1: RWTH Aachen / blue
- Cluster 2: Institutions with more than 1000 publications / violet
- Cluster 3: Institutions with less than 1000 publications / turquoise
- Cluster 21: coordinates world map / grey
2. Basis world map
The basis for creating a world map in VOSviewer is provided by TU Delft inside the "AIDA Booklet" (https://www.vosviewer.com/getting-started by selecting "AIDA booklet"). Case 21 shows a map with GPS points to visualize the borders of the different continents. There are 3 templates for different resolutions:
- simple resolution
- normal resolution
- high resolution
More information and the download link for the required data can also be found here: https://www.tudelft.nl/en/library/research-analytics/case-21-tu-delft-top-collaborators-2
Beside the file for the world map (file 1), a network file (file 2) is required to connect the different institutions with one another, which is implemented through lines between the balls on the map to show the relationship. File 1 gives every institution an ID, which starts by 0 and rises from that point on. These IDs are the baseline for file 2.
3. Create data basis
To visualize a network in VOSviewer, actual data in an adequate form is required - the so called data basis. It can be generated and downloaded with the help of databases like Web of Science or Scopus.
The following example uses fictious data to show the different options of VOSviewer and the impact of various parameters to the final presentation:
- The primary focus is on RWTH Aachen
- The aim is to show the extent of publications between RWTH Aachen and various cooperation partners
- To make use of all possibilities the worldmap provides, 10 fictious partner institutions will be spread all over the globe. Coordinates are required to assign the different institutions to spots on the map.
- To illustrate the impact of different amounts of publications to the chart, the institutions will have either a very high or a very low amount of publications:
Starting point |
Location |
Longitude |
Latitude |
Publications |
Fictious |
---|---|---|---|---|---|
RWTH Aachen | Aachen |
6.080031 |
50.7788 |
84807 |
1 |
Data basis for the fictious cooperation partners:
Fictious institution |
Location |
Longitude |
Latitude |
Publications |
Fictious |
---|---|---|---|---|---|
A |
New York |
-74.0060152 |
40.7127281 |
38695 |
15 |
B |
Kapstadt |
18.417396 |
-33.928992 |
30 |
1 |
C |
Rio de Janeiro |
-43.2093727 |
-22.9110137 |
1560 |
3850 |
D |
Rom |
12.4829321 |
41.8933203 |
15842 |
1 |
E |
Tokyo |
139.759455 |
35.682839 |
50 |
3 |
F |
Seoul |
126.9782914 |
37.5666791 |
8596 |
1 |
G |
Vancouver |
-123.113952 |
49.2608724 |
203 |
500 |
H |
Mumbai |
72.878176 |
19.0785451 |
5248 |
1 |
I |
Madrid |
-3.7035825 |
40.4167047 |
1 |
1 |
J |
Sydney |
151.2082848 |
-33.8698439 |
15986 |
1 |
4. Concept implementation: create files
Institutions, their coordinates, the dedicated clusters, weighting, etc. needs to be filled in individually above the coordinates that generate the world map. The different columns are named and identified as follows:
- id: starts from 0, rises, distinct identifier
- label: name or title, labels the ball
- x: longitude
- y: latitude
- cluster weight: all IDs with the same cluster (number) belong to a group, which can be differentiated by colours
- weight: weighting, determines the size of the ball
- score: an option, to add an additional weighting; the representation can be switched between "size" and "strength"
The resulting file 1 for this example looks as follows:
The network file, file 2, contains the following data:
0 |
1 |
0 | 2 |
0 | 3 |
0 | 4 |
0 | 5 |
0 | 6 |
0 | 7 |
0 | 8 |
0 | 9 |
0 | 10 |
1 | 2 |
File 1: map.txt; File 2: net.txt
5. Create map in VOSviewer
After opening the VOSviewer, the two files can be inserted and therefore combined on the left side of the program by using the option "open".
Afterwards the map can be generated by pressing "ok". The scale needs to be adjusted (see right side) and the RWTH Aachen as well as the institutions A to J can be found. The connections between RWTH and the institutions can be seen as well as the connection between A and B:
The option "strength" can be selected for weighting on the right side:
The resulting map changes as follows:
The different balls can now be coloured individually, for example in RWTH colours. It can be selected by using the option "Cluster Colors" and "Edit Colors" in the right side of the screen. For this example the selected weighting is "size" and coloured as follows:
- Cluster 1 in RWTH-blue: RGB 0-84-159 (RWTH)
- Cluster 2 in RWTH-purple: RGB 122-111-172 (more than 1000 publications)
- Cluster 3 in RWTH-turquoise: RGB 0-152-161 (less than 1000 publications)
- Cluster 21 in RWTH-grey: RGB 156-158-159 (borders)
The exact settings and scalings can be found on the right side.
The menu on the left provides the option to save, copy as screenshot or share the final world map.