Difference between revisions of "IS428 Team wiki: Group12(CJS) Proposal"

From Visual Analytics for Business Intelligence
Jump to navigation Jump to search
Line 153: Line 153:
 
|-
 
|-
 
|  
 
|  
[[File:CJSApproach.png|600px|center]]
+
[[File:CJSApproach_1.png|600px|center]]
 
  ||  
 
  ||  
 
* Draw box-plots for different factors to see the how important these factors are in contributing to the overall happiness score  
 
* Draw box-plots for different factors to see the how important these factors are in contributing to the overall happiness score  

Revision as of 03:59, 25 November 2018

CJSlogo.png
Back to Project Groups

HOME

 

PROPOSAL

 

POSTER

 

APPLICATION

 

RESEARCH PAPER


PROBLEM & MOTIVATION

Problem Since 2012, the United Nations Sustainable Development Solutions Network (SDSN) has been publishing a world happiness report that presents the happiness levels of 156 countries. 6 parameters that include GDP per capita, social support, healthy life expectancy, freedom to make life choices, generosity, and perception of corruption are used to come up with the happiness index for these different countries.

Our visualization, thus, explores the different trends between these 6 parameters and the final happiness index to see exactly which parameters has a higher weightage in the final score.

Motivation The United Nations SDSN reports the happiness ranking of the individual countries yearly. However, trends regarding these 6 parameters are not clearly presented in each of these reports. Importance of such trends can be recognized when we taking the example of Singapore. In 2018, Singapore’s was ranked 34th out of 155 countries in terms of happiness. Though this may seem somewhat high, when compared with their 26th ranking in 2017 will we realise that Singapore is becoming a less “happier” country. Hence, it is important that such trends be explored into to get a clearer understanding of what makes a happy country.

Secondly, the reports does not accurately define which of the 6 parameters play a bigger role in deciding the final happiness index of the various countries. A country ranked first in the world happiness index might not rank first in one of the 6 factors. Thus, it could be interesting to identify the relationship between the happiness index and the 6 different factors that contribute to it and identify what make a happier country as per the standards of UN. Perhaps, consistencies or inconsistencies across the yearly reports years could be observed.

OBJECTIVES
  1. Identify the overall trends of happiness ranking index of different countries over the years from 2015 to 2018.
  2. Identify the rankings of the each of 6 parameters that make up the happiness index of countries.
  3. Compare the 6 parameters between the selected countries and see how they fare in relation to their final happiness index rankings from the years 2015 to 2018.
  4. Compare different countries by geographic locations to find out the happiness pattern in different regions.
DATA DESCRIPTION
Dataset/ Source Data Attributes Rationale of Usage

World Happiness Report Data 2015-2018
https://s3.amazonaws.com/happiness-report/2018/WHR2018Chapter2OnlineData.xls
table2.1

  • County
  • Year
  • Life Ladder
  • Log GDP per capita
  • Social Support
  • Healthy life expectancy at birth
  • Freedom to make life choices
  • Generosity
  • Perceptions of corruption
  • Positive affect
  • Negative affect
  • Confidence in national government
  • Democratic Quality
  • Delivery Quality
  • Standard deviation of ladder by country-year
  • Stand deviation/ Mean of ladder by country-year
  • GINI index (World Bank estimate)
  • GINI index (World Bank estimate), average 2000-15
  • GINI of household income reported in Gallup, by wp5-year

See the overall performance of each individual country changes from year 2006 onwards

World Happiness Report Data 2017

https://s3.amazonaws.com/happiness-report/2018/WHR2018Chapter2OnlineData.xls
Figure2.2

  • Country
  • Happiness score
  • Whisker-high
  • Whisker-low
  • Dystopia + residual
  • GDP Per capita
  • Social support
  • Healthy life expectancy
  • Freedom to make life choices
  • Generosity
  • Corruption

Show data of one particular year(2017), its overall score and thus can get how the final happiness score is calculated

SURVEY OF RELATED WORK

Our visualization are inspired by the following works:

Reference Visualization Work Our Team's Take-away

World Map

CJSref1.jpg

Source: http://shiny.rstudio.com/gallery/superzip-example.html

  • It can be seen in this visualization that 2 different colours are used - one to represent "yes" and one to represent "no". A similar design could be used in our visualization where two different colors of different intensity can be used to show the happiness index of different countries. One colour would be used to categorise the more "happier" countries while the other colour would categorise the less "happier" countries. This would be complemented with different hue intensities to represent the magnitude of the various happiness scores.
  • Upon clicking each of the states, a tooltip is shown that gives a brief information overview of that particular area. This could be useful in our visualization as we provide more information on the happiness index than just the happiness scores. Perhaps, scores of the 6 factors can be shown in the tooltip of our visualization.

Interactive Ranking Bar Chart

CJSref2.jpg

Source: https://www.gapminder.org/tools/#$chart-type=barrank

  • This bar chart presents a slider that allows the users to view the changes in data in a very interactive manner.
  • The ranking are also featured on the side of the graph to help identify which countries are highly ranked from year to year. This could be useful in observing changes in happiness index over the years.
  • The different countries are colour-coded based on the regions they are in. For example, Asian countries are in pink while countries in yellow are in Europe.
  • Overall, the bar chart visualization makes it easy to distinguish the magnitudes of scores between the different countries and thus, is something that could help add value to the exploration of the happiness index.

Parallel Coordinates Plot

CJSrefParallel.jpg

Source: https://bl.ocks.org/jasondavies/1341281

  • As taught in class, this parallel coordinates visualization provides a more consolidated view as the user can observe how the 6 factors link together for each country. Furthermore, users can click to highlight each of the individual countries.
OUR APPROACH
Work Flow Description
CJSApproach 1.png
  • Draw box-plots for different factors to see the how important these factors are in contributing to the overall happiness score
  • Display Choropleth Maps based on the happiness ranking to observe different regions' happiness status
  • Display ranking of different countries based on individual parameters and see their position in the overall happiness score ranking
  • As the 6 factors are given in different scales, normalization of data ((x - min(x)) / (max(x) - min(x))) was conducted for the 6 factors. After normalization, they were presented in a radar chart and compared to see the strenghs and weaknesses between countries.
PROPOSED STORYBOARD
Visualization Description

Main Visualisation: Choropleth Map

CJStory1 1.jpg

Purpose: To show the geographical distribution of the happiness index and the 6 parameters for selected years using a choropleth map. Similar to the survey of related work, this helps the user identify trends between regions. For example, users could see if, for example, Asian countries are “happier” than African countries.

Features to be included:

  • 2 Different colours of varying intensity would be used to represent the varying values of each factors.
  • Filter for user to observe data by individual years and happiness index factors.
  • A tooltip which shows the happiness scores of the different countries will display upon hovering over the different regions on the map.

Main Visualisation: Bar chart

CJStory2.jpg

Purpose: To identify any correlation between the 6 sub-factors and the happiness index. Users can see, for example, what the happiness index rankings are for countries with just the highest GDP.

Features to be included:

  • The bar length would be used to visualise the magnitude of the selected parameters while the number to the side of the bar chart would refer to the happiness index rank of the country.
  • There would be a slider below the chart that would allow users to scroll through and see the bar chart visualisations for the different years.
  • A filter so that the user can select the various parameter to analyse.

Main Visualisation: Parallel Coordinates with Boxplot

cener

Purpose: To help identify trends between the 6 parameters among the different countries. In our visualization, each of the axes would represent the 6 different parameters that make up the happiness index, and each of the different lines will represent the different countries. Thus, the user can compare all the parameters instantly and identify relationships between them.

Features to be included:

  • Boxplot on each of the axis to enable the user to see the distribution for each of the parameter.
  • Brushing feature for users to select a range values in the chart.
  • Filter for people to select the year.
KEY CHALLENGES
Key Technical Challenges Proposed Solution

No centralized data. Data are all divided year by year.

  • Gather all data and join all sets together.
  • Clean and transform the data.
Unfamiliar with Visualization Tools
  • Research on how to use there tools properly.
  • Consultation with professor to get help.
Unfamiliar with R as a new programming language
  • Learn R individually.
  • Incorporate Transformed html page with Javascript & JQuery to give smooth and user-friendly interface.
VISUALIZATION SKETCH

Please check out our visualization sketch to better understand the process. Overall, all these graphs will be tied together on a web-page categorised by different tabs. First tab would feature the choropleth map while the second and third tabs would feature the bar chart and the box plot respectively.

CJSthinkingProcess.png

For section 1(left side of the above image), users would be greeted with a general view of the data from 2015 to 2018 through the geographical distribution visualisation. This section aims to fulfill objective 1. Users can hover over the map and see the 6 parameters that make up the happiness index of the country in the specific year. User need to select the year for Choropleth map.
Visualization presented:

  1. Line Graph
  2. Choropleth map

Section 2(top right of the image) presents a more drilled down view as trends specifically related to the 6 parameters that make up the final happiness index can be seen here after the user selects the parameter and year. Tentatively, user would only need to select any one of the 6 parameters and the chart will change dynamically according to the year. This section aims to fulfill objective 2.
Visualizations presented:

  1. Bar Chart

For the section 3(bottom right of the image), the different y-axes represent the different parameters and lines are plotted for all the different countries.This section aims to fulfill objective 3.
Visualizations presented:

  1. Parallel Coordinates
MILESTONES AND TASK DESCRIPTIONS

Milestones are indicated with the star.

CJSProjTimeline.png
REFERENCES
  1. https://s3.amazonaws.com/happiness-report/2018/WHR2018Chapter2OnlineData.xls
  2. https://s3.amazonaws.com/happiness-report/2018/WHR2018Chapter2OnlineData.xls
  3. http://shiny.rstudio.com/gallery/superzip-example.html
  4. https://www.gapminder.org/tools/#$chart-type=barrank
  5. https://bl.ocks.org/jasondavies/1341281
COMMENTS

Please provide us with any comments to improve our visualization better.