Gain the skills to transform data into stories, using the DIKW pyramid and the Altair Python library.

tarball.tar.gz

Job Code Widget

Tool Z Familiarity

Altair, a versatile Python library, empowers you to craft visually compelling and interactive data visualizations. Acquiring proficiency in Altair unlocks new avenues for effectively conveying insights and narratives through data.

The course begins by introducing the concept of data storytelling and the DIKW pyramid. Then, you’ll learn why Altair is a valuable tool for developers interested in data storytelling and how it can enhance their ability to effectively communicate complex ideas. Next, the course takes a sequential approach to cover the DIKW model for data storytelling, which provides a structured framework for organizing and presenting data narratives. Finally, you’ll implement two practical use cases.

After taking this course, you’ll have acquired the skills and knowledge to build compelling data stories that combine Altair and the DIKW pyramid. You will be able to create engaging and impactful visualizations that effectively convey insights and build narratives from data.

Using Python Altair for Data Storytelling

`chart = alt.Chart(df
).mark_line(
).encode(
     x='x:Q'
)`

`chart = alt.Chart(df
).mark_line(
).encode(
     x='x:O'
)`

`chart = alt.Chart(df
).mark_line(
).encode(
     x='x:N'
)`

Which code correctly represents the `x` channel as a string column?

Which code correctly represents the <code>x</code> channel as a string column?

By using the `altair.native_data()` function

By creating an `InlineDataset` object and wrapping it with `InlineData`

How can we pass an inline data source to an Altair `Chart`?

How can we pass an inline data source to an Altair <code>Chart</code>?

`chart = alt.Chart(df).mark_line(point=True).encode(
    x='year:O',
    y='value:Q',
    color='category'
).properties(width=300)`

`chart = alt.Chart(df).mark_line(point=True).encode(
    x='year:O',
    y='value:Q',
    color='formatted_category:N'
).properties(width=300).transform_calculate(formatted_category='if(datum.category == "dog pet", "dog" ,datum.category)'
)`

`chart = alt.Chart(df).mark_line(point=True).encode(
    x='year:O',
    y='value:Q',
    color='category:N'
).properties(width=300).transform_calculate(formatted_category='if(datum.category == "dog pet", "dog" ,datum.category)'
)`

`chart = alt.Chart(df).mark_line(point=True).encode(
    x='year:O',
    y='value:Q',
    column='category'
).properties(width=300)`


Consider the following dataset:

`df = pd.DataFrame({
    'year': [2020,2020,2022,2022],
    'value': [1, 3,5,6],
    'category': ['cat', 'dog', 'cat', 'dog pet']
})`

To draw a line chart in Altair, one for `dog` and `cat` each, which code represents the data correctly?

Consider the following dataset:
<code>df = pd.DataFrame({ 'year': [2020,2020,2022,2022], 'value': [1, 3,5,6], 'category': ['cat', 'dog', 'cat', 'dog pet'] })</code>
To draw a line chart in Altair, one for <code>dog</code> and <code>cat</code> each, which code represents the data correctly?

Which is _not_ a data source supported by Altair for data visualization?

Which is not a data source supported by Altair for data visualization?

Test yourself on data definition, types, licenses, cleaning, and enrichment.

What is Data Storytelling?

Using Altair for Data Storytelling

The DIKW Pyramid

Introducing the Concept of Data

Getting Familiar with Unknown Data

Assessment 1

Introducing the Concept of Information

Data Modeling

Knowledge

Wisdom

Assessment 2

Case Studies

Wrap Up

Appendix

Quiz: Concept of Data