Visualisation for Social Science Datacsbob/talk/tofts11visualising.pdfDriver for visualisation Data...

Visualisation for Social Science DataDr Chris Tofts

concinnitās ltd

Visiting Prof, Computer Science, Swansea University

+This talk

Introduction and motivating examples

Hans Rosling – Health data

Bob Laremee – Turtle movement

Hans ‘on the cheap’

SMAT provision extended example

Primary Issues

Questions

+Great starting point

+A good starting example –Nightingale’s Rose

+Napolean’s retreat

+Self-Organised Poverty Map

+Re-Mapped

+Sparklines

+Social graph view

+Hypergeometric in 3D

+Beaches in Brazil – google fusion maps

+Chernoff faces

+Driver for visualisation

Data - Increasing generating very large data sets – usually automatically

Information - within which are some interesting patterns

Understanding – presenting the patterns so a human can make use of them.

The aim of good visualisation is to take the storm of data, and

change it into information and consequently increase

knowledge.

+Common features of these examples

High >2 dimensionality

Lots of data

Interactive – some

Relationship data

+Fundamental questions

What is the output device

Application

How much manual intervention is required?

Is the data live?

Comprehension

Influence

Publication

+Why do we still use paper?

15’’ computer screen at 1280 approimately 100 dpi

Cheap modern printer 600 dpi

Good one 1200 dpi

So paper is 36 to 144 times better, but

No interactivity

Hard to access raw resolution without DIRECT generation of pdf.

+Other problems...

Handing over data - trust

Limited time

Different language

+Impact of visual data

+Extended example on interactions with Social support

Multiple different views of the data supporting different questions.

+Visualising SMAT support

Basic data is an interaction at a particular time

Raw information as text

Manual translation – keep format short

+First visualisation

+Showing interaction capability

+Coping with lots of organisations

+Convert to ‘raw’ pdf

+2d Organisation view

Family CFS

Agency Outreach SS SDP MidWife PSALT Pharmacist

Cyrenians Carers Referrer Pharmacy Housing SwanseaPrison SW

APO CMHT WGADA CDAT GP Probation OccupationalHealth

ChildAndFamily HospitalDetox Mother Client College CJIT GrpPhyscotherapist

ComputingCourse GIT SANDS HealthVisitor ProgressToWork IndPhyscotherapist Custody

CAB CDIT SureStart Dentist SANDPIT ShawTrustProject SCBU

Obstetrician PrincesTrust Fairbridge HO BrigendPrison Pharmacy(Other) RPG

+With a particular client

Family CFS

Agency Outreach SS SDP MidWife PSALT Pharmacist

Cyrenians Carers Referrer Pharmacy Housing SwanseaPrison SW

APO CMHT WGADA CDAT GP Probation OccupationalHealth

ChildAndFamily HospitalDetox Mother Client College CJIT GrpPhyscotherapist

ComputingCourse GIT SANDS HealthVisitor ProgressToWork IndPhyscotherapist Custody

CAB CDIT SureStart Dentist SANDPIT ShawTrustProject SCBU

Obstetrician PrincesTrust Fairbridge HO BrigendPrison Pharmacy(Other) RPG

+Combining timeline

+3D animation

+Online Systems

Google charts

IBM many eyes

Google maps

+Fundamental problem

Input syntax’s are all different!

+Pajek view of graph

*Network network name*Vertices N1 "lab" coordX coordY value shape x_fact factX y_fact factY ic color [activ_int]2 "lab" coordX coordY value shape x_fact factX y_fact factY ic color [activ_int]…N "lab" coordX coordY value shape x_fact factX y_fact factY ic color [activ_int]*Arcs :1 "relation name"init_vertex term_vertex value w width c color [activ_int]…*Arcs :2 "relation name"init_vertex term_vertex value w width c color [activ_int]……*Arcs :R "relation name"init_vertex term_vertex value w width c color [activ_int]…*Edges :1 "relation name"init_vertex term_vertex value w width c color [activ_int]…*Edges :2 "relation name"init_vertex term_vertex value w width c color [activ_int]……*Edges :R "relation name"init_vertex term_vertex value w width c color [activ_int]

+Graphviz view

digraph finite_state_machine {node [shape = doublecircle]; LR_0 LR_3 LR_4 LR_8;node [shape = circle];rankdir=LR;LR_0 -> LR_2 [ label = "SS(B)" ];LR_0 -> LR_1 [ label = "SS(S)" ];LR_1 -> LR_3 [ label = "S($end)" ];LR_2 -> LR_6 [ label = "SS(b)" ];LR_2 -> LR_5 [ label = "SS(a)" ];LR_2 -> LR_4 [ label = "S(A)" ];LR_5 -> LR_7 [ label = "S(b)" ];LR_5 -> LR_5 [ label = "S(a)" ];LR_6 -> LR_6 [ label = "S(b)" ];LR_6 -> LR_5 [ label = "S(a)" ];LR_7 -> LR_8 [ label = "S(b)" ];LR_7 -> LR_5 [ label = "S(a)" ];LR_8 -> LR_6 [ label = "S(b)" ];LR_8 -> LR_5 [ label = "S(a)" ];

+H3 view

0 and(Gr3sub1) 1 image1 and(Gr2sub20) 1 image2 and(Gr1sub114) 1 image3 COMP434 1 main3 COMP436 1 main3 COMP433 1 main3 COMP435 1 main3 COMP432 1 main2 and(Gr1sub35) 1 image3 COMP567 1 main3 COMP570 1 main3 COMP573 1 main

+Udraw(graph) – used to be DaVinci

+JSON view 1 of N

var json = { "children": [ { "children":

[ { "children": [], "data": { "description": "$jit namespace and $jit()

for append vis to Global\n \n Example:\n \n new $jit.Hypertree({\n 'injectInto': 'myvisContainer'\n });",

"$angularWidth": 7490, "days": 111, "$color": "#FCD9A1", "size": 7490 }, "id": "Source/Coordinates/Complex.js", "name": "Complex.js" }, { "children": [], "data": { "description": "Fixed polar interpolation problem when theta = pi", "$angularWidth": 6390, "days": 2, "$color": "#B0AAF6", "size": 6390 }, "id": "Source/Coordinates/Polar.js", "name": "Polar.js" } ],

+My view

14/11/7 Client-ass->WGCADA21/11/7 WGCADA-l->Client21/4/08 Refer-l->WGCADA28/4/08 Client-DNA->WGCADA12/5/08 WGCADA-l->Client20/5/08 Client-DNA->WGCADA22/1/09 TSU-l->WGCADA29/1/09 Client-m->WGCADA29/1/09 TSU-m->WGCADA6/2/09 Client-ass->WGCADA24/2/09 Client-m->WGCADA25/2/09 WGCADA-t->Client03/3/09 WGCADA-m->Client04/3/09 WGCADA-t->Client

Not going to do!

+Coping with multiple formats

Simple PERL programs

Why PERL....

Stable version freely available

Its what it was designed to do!

Libraries for direct PDF generation

Libraries for lots and lots of stuff...

Even graphics to do builds

+Another way forward

Your browser is now equivalent to your machine (well almost)

Firefox/Chrome/Safari have stable versions of these features

Combination of CSS, Javascript and HTML exceedingly powerful

JIT has many great examples and is growing

Mapping via google maps

Use project students in CS, but you will have to hand over some data!

+Unlikely to ‘beat’ hand drawn!

+Online resources

http://www.amazon.com/Visual-Display-Quantitative-Information-2nd/dp/0961392142/ref=sr_1_1?s=books&ie=UTF8&qid=1301313153&sr=1-1,

Tufte’s bulletin board http://www.edwardtufte.com/bboard/q-and-a?topic_id=1

Periodic table of charts http://www.visual-literacy.org/periodic_table/periodic_table.html

Google charts http://code.google.com/apis/chart/

IBM charts http://www-958.ibm.com/software/data/cognos/manyeyes/page/Visualization_Options.html

javascript data visualisation http://thejit.org/

H3 system http://graphics.stanford.edu/papers/h3/

service visualisation http://www.edpuss.co.uk/SerViz/sv.htm

Source for Hans Rosling http://www.youtube.com/watch?v=jbkSRLYSojo&feature=player_embedded#at=183

Animation examples http://www.edpuss.co.uk/Animation

Tamarra Muntzer’s software page http://www.cs.ubc.ca/~tmm/software.html

+More resources

Icons http://www.gerdarntz.org/isotype

Google map icons collection http://code.google.com/p/google-maps-icons/

Visualisation for Social Science Datacsbob/talk/tofts11visualising.pdfDriver for visualisation Data...

Documents

Data visualisation - Big data

Data Visualisation in Sharepoint

Data Behaviours Analytics Model for Big Data Visualisation · Data-type visualisation targets the particular type of data, such as text data visualisation (Afzal et al., 2012), video

Visualisation & Data Storytelling

Data Visualisation with QGIS

Interactive Treemaps for Visualisation of Call Centre Datacsbob/talk/qpc/roberts16submission135.pdf · Eurographics Conference on Visualization (EuroVis) 2016 K-L. Ma, G. Santucci,

Visualisation of data instruments

Visualisation and data integration

Interactive Data Visualisation Alessa Gross · 2015-10-13 · Interactive Data Visualisation Applying an interactive data visualisation to Irish road accident data to study usability

Data visualisation using Javascript

Open Data and Visualisation

Beautiful Data Visualisation & D3

Visualisation of metabolic data

Data Visualisation: Observations

Data Visualisation

Agile data visualisation

Data Visualisation - An Introduction

Poster Data-visualisation EN

Data Visualisation for Libraries

Data Visualisation Summit - The Innovation Enterpriseie.theinnovationenterprise.com/eb/DataVizLondonBrochure.pdf · Data visualisation greatly enhances not only data comprehension,