search

Digital-Water-Publics / Thames21-Socio-Ecological-Dashboard

Rivers charity Thames21 and Oxford University are co-creating a dashboard for river-related social media sentiment in the Thames basin
0 stars 2 forks source link

Write non-technical description of methodology #13

Closed hepeb0t closed 2 years ago

hepeb0t commented 2 years ago

Note key facts and figures (geographical area, number of rivers, description and size of data set, algorithms) in a non-technical language on the About page

natesheehan commented 2 years ago

add to about page

natesheehan commented 2 years ago

The project covers the Thames River Basin management catchment which contains 541 water bodies in twenty management catchments. image

Datasets are aggregated at the water body scale and contain sentiment polarity, emotional sentiment detection and common phrases. The total number of tweets collected is over 4 million tweets with a temporal resolution of January 1st 2008 to the present day. Our method utilised an augmented version of the NRC emotional lexicon and a polarized context cluster algorithm in order to determine a given waterbodies sentiment score. The NRC emotional lexicon is a dictionary of English words and their associations with eight basic emotions (anger, fear, anticipation, trust, surprise, sadness, joy, and disgust) and two sentiments (negative and positive). The annotations were manually done by crowdsourcing. The polarised context cluster algorithm gives better results than a simple lookup dictionary approach and functions by clustering groups of words in a tweet (normally around 4) and checks whether valence shifters effect the overall sentiment in the word cluster.

natesheehan commented 2 years ago

going to close this