Geotagged Sentiment Analysis of Tweets for Subjective Well Being Metrics

Sentiment of the past 24hr global tweets containing ‘global warming’. Despite its biases, sentiment analysis of social media can be one source of data for ‘subjective well being’ towards a topic – supporting UN Sustainable Development Goals (SDG).

Cities and governments are increasingly incorporating these indicators with traditional economic metrics. This display is using the #onemilliontweetmap based on Abdul Fatir’s code in github.

Harvard’s Center for Geographic Analysis and MIT have built the Twitter Sentiment Geographical Index (TSGI). This is “to study the effect of climate change on human well-being using social-media data”. The index contains 7.4Billion geotagged tweets and is open to the public.

The excellent recent paper ‘Three Horizons for Future Geoscience’ highlights the need for greater links between traditional geosciences and sustainability, society and well-being.

Using sentiment analysis trends in this area may perhaps be one way to assess changes over time and societal attitudes towards the geosciences in this complex system.

References:

https://onemilliontweetmap.com/?center=2.8113711933311403,29.179687500000004&zoom=1&search=Global%20warming&timeStep=0&timeSelector=0&hashtag1=&hashtag2=sad&sidebar=yes&hashtagBattle=0&timeRange=0&timeRange=25&heatmap=1&sun=0&cluster=1&rendering=sentiment

https://gis.harvard.edu/dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/3IL00Q

https://www.escubed.org/articles/10.3389/esss.2023.10079/full

#naturallanguageprocessing #sentimentanalysis #geography #gis #socialmedia #wellbeing #artificialintelligence #geosciences