Prototype

Dataset Summary

My dataset comes from the pokemon universe. It has attributes concerning all the pokemon from all six generations. There are currently 721 of the creatures who fit into many different categories. The follow following information is contained for each:

Generation: Generation in which a pokemon first appears. (e.g., Red and Blue are first generation, Gold and Silver are second generation.)
id: ID Number.
Name: Creature name.
Primary Type: Type of pokemon, e.g. Psychic, Water, Normal.
Secondary Type: Secondary type of the same variety as the first. Not always present.
Health Points: Amount of points that can be lost before fainting.
Attack: Determines amount of damage given from attacks.
Defense: Determines amount of damage taken from attacks.
Special Attack: Determines amount of damage given from attacks related to primary and secondary types.
Special Defense: Determines amount of damage taken from attacks related to primary and secondary types.
Speed: Determines which pokemon attacks first. (The game is turn-based.)
Total Attribute Points: Sum of all attribute points.

If you are not familiar with the games: The idea is that there are a diverse set of creatures with different attributes that you catch and train. Then using a team of these pokemon you explore the world, see new pokemon, spar with your rival, defeat other prominent trainers, and save the world from an evil empire.

You may view the Dataset Writeup by clicking here.

Sketch Summary

     I chose to use parallel coordinates for the sketch because of how simple it is to use and how much exploration can be done with it. Initially I was hoping to color by Primary Type but the high number led to a confusing amount of colors in the plot.
     I plan to have this plot shown first so that people can explore
     For my second chart (which is not shown) I plan to run a Principal Component Analysis and plot the first two dimensions so that users can see a plotted explanation of the variance to get a better understanding of the dataset they just explored.
     I have also started another one of my charts: the radar plot. The radar plot has already been used throughout the pokemon universe so I know that people with a knowledge of the game will probably have seen this technique before, making it easier for them to understand what is happening. I'm hoping to have the ability to filter based on pokemon to allow the user to choose two teams of six pokemon and compare average statistics across the team.

You may view the Dataset Sketch by clicking here.

Visualization Prototype

Steven Rea - MSAN 622