Analysis
Revision as of 17:20, 16 October 2017 by Matilda.tan.2016 (talk | contribs)
Overview | Data Preparation | Analysis and Results |
Contents
Preliminary Analysis in JMP
Due to the possibility of describing sickness in slang or terms which we are not as familiar with, the JMP text explorer was used to identify such terms through its in-built text analysis function.Terms such as those circled in red in the figure below were considered when compiling the list of symptoms to filter the posts with, in order to get more accurate results.
Also notable were its mention of a music festival and a truck accident, which we would later learn more about in our following analysis through visualisation of the data in Tableau.
Data Analysis in Tableau
Analysis of travel patterns
Comparison with content from posts
Hourly microblog feed dashboard
Strategy for target narrowing
Creating a holistic view
Proliferation of disease dashboard
Insights from Analysis
Suspicious events: Possible human to human transmission
Suspicious events: Possible wind and water transmission
Results and conclusion
Origin and channel of transmission