Data Visualization

INF-604: Data Analysis

Lecturer: Dr. Sothea HAS

Outline

Motivation
Bivariate Visualization
Multivariate Visualization
Time series data
Animated charts/graphs

🌐 https://clauswilke.com/dataviz/

Motivation

`Gapminder dataset` (1704, 5)

This dataset captures the world’s evolution from \(1952\) to \(2007\).
Now, take a look at the data from year \(2007\).

Code

from gapminder import gapminder
data2007 = gapminder[gapminder.year == 2007]  # filter to year 2007
data2007.iloc[:5,:].drop(columns=['year']).style.hide()

country	continent	lifeExp	pop	gdpPercap
Afghanistan	Asia	43.828000	31889923	974.580338
Albania	Europe	76.423000	3600523	5937.029526
Algeria	Africa	72.301000	33333216	6223.367465
Angola	Africa	42.731000	12420476	4797.231267
Argentina	Americas	75.320000	40301927	12779.379640

Motivation

`Gapminder dataset` (1704, 5)

This dataset captures the world’s evolution from \(1952\) to \(2007\).
Now, take a look at the data from year \(2007\) (summary).

Code

quan_vars = ["pop", "lifeExp", "gdpPercap"]
data2007[quan_vars].describe().transpose().drop(columns=["count", "25%", "75%"]).transpose()

	pop	lifeExp	gdpPercap
mean	4.402122e+07	67.007423	11680.071820
std	1.476214e+08	12.073021	12859.937337
min	1.995790e+05	39.613000	277.551859
50%	1.051753e+07	71.935500	6124.371108
max	1.318683e+09	82.603000	49357.190170

Motivation

`Gapminder dataset` (1704, 5)

This dataset captures the world’s evolution from \(1952\) to \(2007\).
Now, take a look at the data from year \(2007\) (visualization).

Code

from plotly.subplots import make_subplots
import plotly.graph_objects as go
fig = make_subplots(rows=1, cols=3, 
              subplot_titles=("Boxplot of pop", "Violinplot of lifeExp", "Histogram of GDP Per Capita"))
fig.add_trace(go.Box(y=data2007['pop'], name="pop"), col=1, row=1)
fig.add_trace(go.Violin(y=data2007['lifeExp'], name="lifeExp"), row=1, col=2)
fig.add_trace(go.Histogram(x=data2007['gdpPercap'],
              name="gdpPercap"), row=1, col=3)
fig.update_layout(height=280, width=1000)
fig.update_yaxes(type="log", row=1, col=1)
fig.update_xaxes(title="Population", row=1, col=1)
fig.update_xaxes(title="Life Expectancy", row=1, col=2)
fig.update_xaxes(title="GDP Per Capita", row=1, col=3)
fig.show()

Hans Rosling’s 200 Countries, 200 Years in 4 Minutes.

Bivariate Visualization

Quantitative vs quantitative: Scatterplot

Scatterplot shows trends/relation of quantitative pairs.
Let’s visualize relation: gdpPercap wih lifeExp & pop.

Code

import plotly.graph_objects as go
import plotly.express as px
data2007 = gapminder.query("year == 2007")
fig1 = px.scatter(data2007, x="gdpPercap", y="lifeExp", hover_name="country", opacity=0.7)
fig1.update_traces(marker=dict(size=10))
fig1.update_layout(height=350, width=500, title="The world GDP vs LifeExp in 2007")
fig1.show()

Code

data2007 = gapminder.query("year == 2007")
fig2 = px.scatter(data2007, x="gdpPercap", y="pop", hover_name="country", opacity=0.7)
fig2.update_traces(marker=dict(size=10))
fig2.update_layout(height=350, width=500, title="The world GDP vs Population in 2007")
fig2.show()

Bivariate Visualization

Quantitative vs quantitative: Scatterplot

Scatterplot shows trends/relation of the quantitative pair.
Let’s visualize relation: gdpPercap wih lifeExp & pop.

Code

fig1.update_layout(title="The world (log) GDP vs Population 2007 ")
fig1.update_xaxes(type="log")
fig1.show()

Code

fig2.update_layout(title="The world GDP vs (log) Population 2007 ")
fig2.update_yaxes(type="log")
fig2.show()

Bivariate Visualization

Quantitative vs quantitative: Scatterplot

GPD vs Life Expectancy:
- General trend: Countries with high GPD tend to be healthier.
- There are also a few countries with economy well above average yet health condition is still bad.

GPD vs Population:
- General trend: no clear trend!
- GDP per capita does not appear to be significantly influenced by a country’s population size.

Bivariate Visualization

Quantitative vs qualitative: Conditional

To see relation between Values within different Group, we can use:
- Conditional Boxplots: boxplots within different groups.
- Conditional Histogram/Density.

Code

sorted_data = data2007.sort_values(by='lifeExp')
fig = px.box(data2007, x="continent", y="lifeExp", hover_name="country", color="continent", category_orders={'continent': sorted_data['continent']})
fig.update_layout(title="Life Expectancy on each continent in 2007", height=300, width=450)
fig.show()

Key: The distinction of quantitative values between different groups indicates a connection between the pairs.
Example:
- Clear distinction of lifeExp accross different continent suggests that there is a relation between the two.
- continent is useful for predicting / explaining lifeExp.

Bivariate Visualization

Quantitative vs qualitative: Conditional

To see relation between Values within different Group, we can use:
- Conditional Boxplots: boxplots within different groups.
- Conditional Histogram/Density.

Code

import plotly.figure_factory as ff
group_labels = list(data2007.continent.unique())
hist_data = [data2007.lifeExp[data2007.continent == x] for x in group_labels]
colors = ["#f1ab17", "#f13c26", "#9be155", "#4ab8dc", "#d567f3"]
fig = ff.create_distplot(hist_data, group_labels, colors=colors,
                         bin_size=1.5, show_rug=False)
fig.update_layout(title="Life Expectancy on each continent in 2007", height=300, width=450)
fig.show()

Key: The distinction of quantitative values between different groups indicates a connection between the pairs.
Example:
- Clear distinction of lifeExp accross different continent suggests that there is a relation between the two.
- continent is useful for predicting / explaining lifeExp.

Bivariate Visualization

Quantitative vs qualitative: Conditional

How about GDP on each continent?

Code

sorted_data = data2007.sort_values(by='gdpPercap')
hist_data_gdp = [data2007.gdpPercap[data2007.continent == x] for x in group_labels]
colors = ["#f1ab17", "#f13c26", "#9be155", "#4ab8dc", "#d567f3"]
fig3 = px.box(data2007, x="continent", y="gdpPercap", hover_name="country", color="continent", category_orders={'continent': sorted_data['continent']})
fig3.update_layout(title="GDP per Capita on each continent in 2007", height=350, width=500)
fig3.show()

Code

sorted_data = data2007.sort_values(by='gdpPercap')
hist_data_gdp = [data2007.gdpPercap[data2007.continent == x] for x in group_labels]
colors = ["#f1ab17", "#f13c26", "#9be155", "#4ab8dc", "#d567f3"]
fig3 = px.violin(data2007, x="continent", y="gdpPercap", hover_name="country", color="continent", category_orders={'continent': sorted_data['continent']})
fig3.update_layout(title="GDP per Capita on each continent in 2007", height=350, width=500)
fig3.show()

Example:
- The separation of GDP per Capita between coninents is not as clear as Life Expectancy, yet one can still see the differences.
- continent is useful for predicting / explaining gdpPercap though not as strong/clear as with lifeExp.

Bivariate Visualization

Qualitative vs qualitative: mosaic plot

We don’t have many qualitative columns,
I do grouped GDP:
- If GDP \(\leq 33.33\%\) 👉 Developing
- elif GDP \(\leq 66.66\%\) 👉 Emerging
- else GDP \(\geq 66.66\%\) 👉 Developed.

Bivariate Visualization

Qualitative vs qualitative: mosaic plot

We don’t have many qualitative columns,
I do grouped GDP:
- If GDP \(\leq 33.33\%\) 👉 Developing
- elif GDP \(\leq 66.66\%\) 👉 Emerging
- else GDP \(\geq 66.66\%\) 👉 Developed.

Example:
- As GDP seems to be related to continent, it remains true with categorical GDP.
- In Asia, the three types of economic conditions are well balanced, whereas the majority of African countries are developing, followed by emerging economies.

Code

from statsmodels.graphics.mosaicplot import mosaic
import matplotlib.pyplot as plt
import pandas as pd
fig, ax = plt.subplots(figsize=(7, 5))
plt.rcParams.update({'font.size': 15})
def prop(key):
    if "Asia" in key:
        return {'color': '#51cb4b'}
    if "Africa" in key:
        return {'color': '#e35441'}
    if "Americas" in key:
        return {'color': '#41b4e3'}
    if "Europe" in key:
        return {'color': '#dda63e'}
    if "Oceania" in key:
        return {'color': '#b374df'}

data2007['gdp_category'] = pd.qcut(data2007['gdpPercap'], q=3, labels=['Developing', 'Emerging', 'Developed'])
mosaic(data2007.sort_values('continent'), ['continent','gdp_category'], 
    gap=0.01, properties = prop, 
    label_rotation=30, ax=ax)
plt.title("Mosaicplot of categorical GDP vs Continent")
plt.show()

Bivariate Visualization

Qualitative vs qualitative: grouped barplots

We don’t have many qualitative columns,
I do grouped GDP:
- If GDP \(\leq 33.33\%\) 👉 Developing
- elif GDP \(\leq 66.66\%\) 👉 Emerging
- else GDP \(\geq 66.66\%\) 👉 Developed.

Example:
- As GDP seems to be related to continent, it remains true with categorical GDP.
- In Asia, the three types of economic conditions are well balanced, whereas the majority of African countries are developing, followed by emerging economies.

Code

fig = px.histogram(
    data2007, x="continent", color="gdp_category")
fig.update_layout(width=510, height=470, 
    title='Stacked Barplot of Categorical GDP vs Continent')
fig.show()

Code

fig = px.histogram(
    data2007, x="continent", color="gdp_category", barmode='group')
fig.update_layout(width=510, height=470, title='Grouped Barplot of Categorical GDP vs Continent')
fig.show()

Multiple Information

Color: quantitative & qualitative

Color can represent:
- qualitative data (discrete color).
- quantitative (in form of gradient)

Code

import numpy as np
data2007[' '] = np.repeat('Data', data2007.shape[0])
fig = px.scatter(
    data2007, x="gdpPercap", y="lifeExp",
    hover_name="country", size_max=80, color=" ")
fig.update_layout(width=472, height=400, title='Life Expectancy vs GDP per Capita & Continent')
fig.update_xaxes(type="log")
fig.show()

Multiple Information

Color: quantitative & qualitative

Color can represent:
- qualitative data (discrete color).
- quantitative (in form of gradient)
Example:
- Color = continent, which is a categorical column.

Code

fig = px.scatter(
    data2007, x="gdpPercap", y="lifeExp", color="continent", hover_name="country", size_max=80)
fig.update_layout(width=500, height=400, title='Life Expectancy vs GDP per Capita & Continent')
fig.update_xaxes(type="log")
fig.show()

Multiple Information

Color: quantitative & qualitative

Color can represent:
- qualitative data (discrete color).
- quantitative (in form of gradient)
Example:
- Color = continent, which is a categorical column.
- Color = leftExp, which is a quantitative column.

Code

fig = px.scatter(
    data2007, x="gdpPercap", y="lifeExp", color="lifeExp", hover_name="country", size_max=80)
fig.update_layout(width=500, height=240, title='Life Expectancy vs GDP per Capita & Continent')
fig.update_xaxes(type="log")
fig.show()

Multiple Information

Shape/Symbol: qualitative

Shape for representing qualitative data.
Example:
- Symbol = gdp_category.
- Color = continent.

Combining numerous colors and symbols can complicate a graph.

Use them carefully and only when appropriate.

Code

fig = px.scatter(
    data2007, x="gdpPercap", y="lifeExp", color="continent", 
    hover_name="country", symbol='gdp_category', size_max=80)
fig.update_layout(width=500, height=350, title='Life Expectancy vs GDP per Capita & Continent')
fig.update_xaxes(type="log")
fig.show()

Multiple Information

Size: quantitative

Size for representing quantitative data.
Example:
- Size = pop.
- Color = continent.

Colors and size are common, and the resulting graphs are often called Bubble chart.

One shoule choose suitable max size to have a nice graph.

Code

fig = px.scatter(
    data2007, x="gdpPercap", y="lifeExp", color="continent",
    hover_name="country", size="pop", size_max=35)
fig.update_layout(width=500, height=370, title='Life Expectancy, GDP, Population & Continent')
fig.update_xaxes(type="log")
fig.show()

Multiple Information

3D: quantitative

All the previous options can be used with 3D scatter plot.
Example: Marketing
- X = Youtube.
- Y = Facebook.
- Z = Sales.
- Size = Newspaper.
- Color = Newspaper.

Avoid 3D if they are not interactive [Section: “Don’t go 3D” by Claus O. Wilke (2019)].

Time series data

Quantitative: lineplot

Let’s take a look at Oceania from 1952 to 2007.

Code

df_ocean = gapminder.query("continent == 'Oceania'")
fig = px.line(df_ocean, x='year', y='lifeExp',
    symbol="country", color="country",
    title="Evolution of Life Expectancy")
fig.update_layout(height=350, width=330)
fig.show()

Code

fig = px.line(df_ocean, x='year', y='pop',
    symbol="country", color="country",
    title="Evolution of Population")
fig.update_layout(height=350, width=330)
fig.show()

Code

fig = px.line(df_ocean, x='year', y='gdpPercap',
    symbol="country", color="country",
    title="Evolution of GDP per Capita")
fig.update_layout(height=350, width=330)
fig.show()

Is there any other country’s evolution you would like to see?

Time series data

Quantitative: lineplot

Is there any other country’s evolution you would like to see?

Time series data

Qualitative: Evoluational barplots

Let’s take a look at the evolution of GDP per capita categories for Asian countries over time.

Code

def cat_gdp(yearly_data):
    return pd.qcut(yearly_data, q=3, labels=['Developing', 'Emerging', 'Developed'])
df = gapminder
# Apply the function to each year
df['GDP_Category'] = df.groupby('year').apply(lambda x: cat_gdp(x.gdpPercap)).reset_index(level=0, drop=True)

df_Af = df.query("continent == 'Asia'")
# Aggregate the data
df_agg = df_Af.groupby(['year', 'GDP_Category']).size().reset_index(name='Count')

# Create the stacked bar chart
fig = px.bar(
    df_agg, x='year', y='Count', 
    color='GDP_Category', barmode='stack',
    title="Evolution of Asian Countries' GDP Categories from 1952 to 2007",
    labels={'Count': 'Number of Countries', 'year': 'Year'})

fig.update_layout(height=350, width=1000)
fig.show()

Animated Graphs

Animation with Plotly

Code

import plotly.express as px
df = px.data.gapminder()
fig_anime = px.scatter(df, x="gdpPercap", y="lifeExp", animation_frame="year", animation_group="country",
           size="pop", color="continent", hover_name="country",
           log_x=True, size_max=50, range_x=[100,100000], 
           range_y=[25,90])
fig_anime.update_layout(height=460, width=1000, 
    title="The Evolution of the World in a Single Graph")

“There is no such thing as information overload. There is only bad design.” – Edward Tufte.

Data Visualization

Outline

Motivation

Motivation

Gapminder dataset (1704, 5)

Motivation

Gapminder dataset (1704, 5)

Motivation

Gapminder dataset (1704, 5)

Bivariate Visualization

Bivariate Visualization

Quantitative vs quantitative: Scatterplot

Bivariate Visualization

Quantitative vs quantitative: Scatterplot

Bivariate Visualization

Quantitative vs quantitative: Scatterplot

Bivariate Visualization

Quantitative vs qualitative: Conditional

Bivariate Visualization

Quantitative vs qualitative: Conditional

Bivariate Visualization

Quantitative vs qualitative: Conditional

Bivariate Visualization

Qualitative vs qualitative: mosaic plot

Bivariate Visualization

Qualitative vs qualitative: mosaic plot

Bivariate Visualization

Qualitative vs qualitative: grouped barplots

Multiple Information

Multiple Information

Color: quantitative & qualitative

Multiple Information

Color: quantitative & qualitative

Multiple Information

Color: quantitative & qualitative

Multiple Information

Shape/Symbol: qualitative

Multiple Information

Size: quantitative

Multiple Information

3D: quantitative

Time series data

Time series data

Quantitative: lineplot

Time series data

Quantitative: lineplot

Time series data

Qualitative: Evoluational barplots

Animated Graphs

Animated Graphs

Animation with Plotly

🥳 Yeahhhh….

Let’s Party… 🥂

`Gapminder dataset` (1704, 5)

`Gapminder dataset` (1704, 5)

`Gapminder dataset` (1704, 5)