About the big data
In this context we don’t need to use any algorithm for phishing or any complex data analysis. The data can be restricted spatially (to a country and
particular sets of information) so these are the parts of big data that can be linked to derive achievable outcomes. Possible methods:
Twitter sentiment analysis
Google search analysis
Twitter sentiment analysis
By using social media (Twitter, Facebook, etc.) - we can get a better profile for what has happened in the examined area. By scanning tweets for epidemic terminology, we can follow a contagion’s velocity.
Search features: Spatial characteristics - infected area
Special Terminology: epidemiological words, words what are indicating the disease
Terminology for each language or dialect
Disadvantages – Sociodemographic characteristics
44.2 percent of the population have never attended school - Literacy Rates of Population 10 years and above: 48.6% not literate
Households by main source of information: 71.1% radio; 0.6% social media
13% have access to internet
Used method
Data is pulled by RapidMiner from the Twitter, "AYLIEN Text Analysis" identify the sentiment, the result is dropped to Tableau.