Twitter Multi Variate Analysis
Twitter feeds for 3 different streams, Ethereum, Bitcoin, Facebook for March and April 2018
Does, otherwise uncorrelated twitter channels, tend to discuss on similar topics as a result of world level event?
- Contingency table: Count of words (Top 10) / Log of Count of Words (Top 50)
- Week 9 – 18
- Week 9, 16, 18
- Week 13, 14, 17
- Week 10, 11, 12, 15
For log of count of words
Plotting word occurances vs week - Top 10
Plotting word occurances vs weeks - Top 10
Plotting word occurances vs week - Top 50
Plotting word occurances vs weeks - Top 50
Mapping word occurances vs week (Only means of each dataset)
For Count of words (Top 10)
For log of count of words
Dividing weeks into groups
For Count of words (Top 10)
For log of count of words (Top 50)