Getting Started with Elixir Data Exploration

This document demonstrates how to use Elixir and Livebook for interactive data exploration. We'll learn basic data processing, visualization, and analysis techniques.

What you'll learn:

Processing collection data with the Enum module
Creating interactive tables with Kino
Data visualization with VegaLite

Setup

First, let's install the necessary dependencies:

Mix.install([
  {:kino, "~> 0.12"},
  {:vega_lite, "~> 0.1"},
  {:kino_vega_lite, "~> 0.1"}
])

:ok

Creating Sample Data

Let's create a dataset containing programming language popularity:

# Define programming language data
languages = [
  %{name:"Elixir", year:2011, type:"Functional", popularity:85},
  %{name:"Rust", year:2010, type:"Systems", popularity:92},
  %{name:"Go", year:2009, type:"Systems", popularity:88},
  %{name:"TypeScript", year:2012, type:"Scripting", popularity:95},
  %{name:"Kotlin", year:2011, type:"JVM", popularity:78},
  %{name:"Swift", year:2014, type:"Mobile", popularity:82}
]

[
  %{name: "Elixir", popularity: 85, type: "Functional", year: 2011},
  %{name: "Rust", popularity: 92, type: "Systems", year: 2010},
  %{name: "Go", popularity: 88, type: "Systems", year: 2009},
  %{name: "TypeScript", popularity: 95, type: "Scripting", year: 2012},
  %{name: "Kotlin", popularity: 78, type: "JVM", year: 2011},
  %{name: "Swift", popularity: 82, type: "Mobile", year: 2014}
]

Displaying Interactive Tables with Kino

Kino can render data as beautiful interactive tables:

Kino.DataTable.new(languages)

Name	Year	Type	Popularity
Elixir	2011	Functional	85
Rust	2010	Systems	92
Go	2009	Systems	88
TypeScript	2012	Scripting	95
Kotlin	2011	JVM	78
Swift	2014	Mobile	82

Data Processing and Analysis

Using the Enum module for data filtering and transformation:

# Filter languages released after 2010 with popularity > 80
languages
|> Enum.filter(fn lang -> lang.year >= 2010and lang.popularity > 80end)
|> Enum.map(fn lang -> "#{lang.name} (#{lang.year})"end)

["Elixir (2011)", "Rust (2010)", "TypeScript (2012)", "Swift (2014)"]

# Calculate average popularity
avg_popularity = 
  languages
  |> Enum.map(& &1.popularity)
  |> Enum.sum()
  |> Kernel./(length(languages))
  |> Float.round(1)

"Average popularity: #{avg_popularity}"

"Average popularity: 86.7"

Data Visualization

Creating charts with VegaLite:

alias VegaLite, as: Vl

Vl.new(width:500, height:300, title:"Programming Language Popularity")
|> Vl.data_from_values(languages)
|> Vl.mark(:bar)
|> Vl.encode_field(:x, "name", type::nominal, title:"Language")
|> Vl.encode_field(:y, "popularity", type::quantitative, title:"Popularity")
|> Vl.encode_field(:color, "type", type::nominal)

Summary

In this Livebook document, we learned:

How to create and process Elixir data structures
Using the Enum module for data filtering and transformation
Using Kino to display interactive tables
Using VegaLite to create data visualizations

Next steps: Try modifying the code, adding more data, or creating different types of charts!

Getting Started with Elixir Data Exploration

Getting Started with Elixir Data Exploration

Setup

Creating Sample Data

Displaying Interactive Tables with Kino

Data Processing and Analysis

Data Visualization

Summary

Related Resources