OF TERROR BIJOY RAVEENDRAN INFORMATION amp INTERFACE DESIGN JULY 26 2013 NATIONAL INSTITUTE OF DESIGN 30 years of every terrorist incident from 1970 2011 200K records Retrieved from the Global Terrorism DB ID: 322073
Download Presentation The PPT/PDF document "THE SHADOW" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
THE SHADOWOF TERROR
BIJOY RAVEENDRANINFORMATION & INTERFACE DESIGN
JULY 26 2013 | NATIONAL INSTITUTE OF DESIGNSlide2
30 years of every terrorist incident from 1970 – 2011 (200K records)
Retrieved from the Global Terrorism DB
Date of attack, place of attack, group name, target, casualities, weapons used, damages, motive,
descriprion
Supplementary information from Wikipedia
DATASET
DATASETSlide3
‘Exploratory’ more than ‘explanatory’
Data is in the form of events – so have a timeline and plot it on a map
No story at the moment – so emphasize on the numbers and threat level
Gather supplementary information from Wikipedia
CONSIDERATIONS
CONSIDERATIONSSlide4
INITIAL ITERATIONS
ITERATIONSlide5
Markers on map can be represented as explosions (or the color of fire)
Use opacity to show overlapping layers or use ‘no-overlap’ (ruled out later)Make the map borderless , add more negative space around content, so emphasis goes to subject matter
Tree maps a better option over bar charts for summary data
CONSIDERATIONS (contd.)
CONSIDERATIONSSlide6
Scaled down dataset (30 years to 7 years)
Removed unused data columns
Removed ‘invalids’
0 or unknown month/day/yearUnknown city/state/country
0
casualities/wounded and 0/unknown damages
SANITIZING DATA
SANITIZING DATASlide7
Used Google’s ‘Geo’ API to geo-tag locations
Pivot tables based on ‘terrorist organizations’, ‘country of attack’, and ‘mode of attack’ columns.Formula to calculate severity of attack based on following data:
Number of people killedNumber of people wounded
Estimated damage to property
ENRICHING DATA
ENRICHING DATASlide8
Severity = ( dead_index
* 10 + wounded_index
* 7.5 + damage_index * 2.5 )
ENRICHING DATA (contd.)
Dead_Index
0-40 : 140-80 : 2
80-120 : 3120-160 : 4160 – 200 :
5
Wound_Index
0-75 :
1
75-150 :
2
150-225 :
3
225-300 :
4
>
300:
5
Damage_Index
<1million :
1
1m – 1b :
2
>1b :
3
ENRICHING DATASlide9
Excel for data/business logicTileMill
for map creationMapBox for map hosting
Google Docs for geo-codingGoogle Visualization API to generate tree-mapsHTML/Javascript
for layout
For proposed future use :
MapBox.js/Wax for custom tooltips/legendFlash for intro screenTOOLS
TOOLSSlide10
SEE VISUALIZATION NOW!
Goto visualizationSlide11
‘Invalids’ may be incidents of high severity
TileMill rejects non geo-coded locations – skews the numbers
Geo-coding API is not entirely accurate – combination of city/country not yielding better results. May have to manually encode.
SOURCES OF ERROR
ERRORSSlide12
All deaths are considered the same – shouldn’t assassinations be considered to be of higher priority?
Large data shown as “############”
If ‘others’ occurs in any of the top 10 visualizations, it has been ignored
SOURCES OF ERROR
ERRORSSlide13
Incorporate complete data set
Complete other sections
Include opening sequence to set up the story
Adding animation to map markers will help emphasize more on numbers
FUTURE PLANS
FUTURE PLANSSlide14
FUTURE PLANS
FUTURE PLANS
Link tooltip to related news article (probably for high severity incidents)
Remove top 5 countries from tree-map. Have a
choroplethic
view on country wise impact instead.Include – ‘Your country has XX% chance of being attacked’Use different color marker to represent zero casuality
attacksMatch-upsSlide15
THANK YOU
JULY 26 2013 | NATIONAL INSTITUTE OF DESIGN