Micro-Lecture Twitter Text Mining

This small lecture snippet was held as part of the BMBF-project ICAML and on invitation as well for the lecture Geodata Mining at the Chair of Cartography of TU Munich on July, 2nd.

Here is a walk-through of all the steps involved in basic text mining from Twitter

The following slide deck can serve as a table-of-contents if you want to learn how to

• acquire social media data exemplified with the Twitter API
• how to work with very large JSON collections using the excellent j qquery language and associated tools
• how to apply the excellent FastText text embedding engine to such collections of text

It is up to you to apply this knowledge for something good!

If you got interested, check out our research on what you can do with text embeddings in spatial computing (Häberle et al., 2019)

Bibliography

1. Häberle, M., Werner, M., & Zhu, X. (2019). Geo-Spatial Text-Mining from Twitter—A Feature Space Analysis with a view towards Building Classification in Urban Regions. European Journal of Remote Sensing, 52(sup2), 2–11. https://doi.org/10.1080/22797254.2019.1586451 [BibTeX]

   @article{geospatial_textmining_featurespace,
     author = {Häberle, Matthias and Werner, Martin and Zhu, Xiaoxiang},
     title = {Geo-Spatial Text-Mining from Twitter—A Feature Space Analysis with a view towards Building Classification in Urban Regions},
     journal = {European Journal of Remote Sensing},
     volume = {52},
     number = {sup2},
     pages = {2-11},
     year = {2019},
     publisher = {Taylor & Francis},
     doi = {10.1080/22797254.2019.1586451}
   }