Difference between revisions of "Jarvis Methodology"

From Analytics Practicum
Jump to navigation Jump to search
 
Line 10: Line 10:
  
 
| style="background:none;" width="1%" |  
 
| style="background:none;" width="1%" |  
| style="padding:0.2em; font-size:100%; background-color:#f1cf0e;  border-bottom:3px solid #476a37; border-top:3px solid #476a37;border-left:3px solid #476a37; border-right:3px solid #476a37;text-align:center; color:#F5F5F5" width="10%" |  
+
| style="padding:0.2em; font-size:100%; background-color:#f1cf0e;  border-bottom:3px solid #476a37; border-top:3px solid #476a37;text-align:center; color:#F5F5F5" width="10%" |  
 
[[JAR v.IS Project Proposal| <font color="#b1260e" size=2 face="Century Gothic"><b>PROJECT PROPOSAL</b></font>]]
 
[[JAR v.IS Project Proposal| <font color="#b1260e" size=2 face="Century Gothic"><b>PROJECT PROPOSAL</b></font>]]
  

Latest revision as of 10:52, 17 January 2017

Click here to return to AY16/17 T2 Group List

Jarvis.png

HOME

 

PROJECT PROPOSAL

 

PROJECT FINDINGS

 

PROJECT MANAGEMENT

 

DOCUMENTATION

 

ABOUT US

Background Data Methodology
Methodology

We have conducted our Exploratory Data Analysis with the process outlined below:

  • At the page level
    • Growth of page likes by country + geographical map
    • Data visualization of Facebook insights data by geography, date and gender

  • At the post level
    • Word cloud
    • Topic cluster analysis - to verify how well the current categories match the content
    • ANOVA test to test significance of mean likes on posts of the type links vs video

  • At the video level
    • Video duration
    • Video views

Based on the first iteration EDA, we have concluded to focus our further analysis on video data and posts with a focus on video data.