new icn messageflickr-free-ic3d pan white
Campus TU Dresden Tag & Emoji Clustering | by Sieboldianus
Back to photostream

Campus TU Dresden Tag & Emoji Clustering

With the new clustering

code in Python it is easy to adapt the Tag Maps technique to other data and information.


This map is based on photos with tags and emojis from both Flickr (2007-2018) and Instagram (2010-2018).


There're about 2400 Flickr photos (from 300 users with 15,000 tags) and 10,000 Instagram photos (from 5000 users, with 50k tags and 6500 emojis) - not much for an area of this size.


The key to equal mapping of emojis and tags here was to process both sets of data separately and normalize the resulting weights to the same range. This way, emojis and tags have an equal chance to appear on the map.


I think it is quite interesting how both sets of information supplement each other.


Map tiles by Carto, under CC BY 3.0. Data by OpenStreetMap, under ODbL.

1 fave
Uploaded on February 11, 2018