The Elder Scrolls V: Skyrim Special Edition — Analysis of Dialogues

Read this on my blog


Disclaimer: A lot of the code is shared between my previous project and this one


Getting the data

I am going to look into other data I can extract this way, and maybe make some other stuff.

Processing the data

Counting the Words

I counted both RESPONSE TEXT & TOPIC TEXT data separately and then merged them into a single file count.json

Additional Tip: progress is a great Python Package to show progress in your scripts.

Making the WordCloud

To make the WordCloud, I used the wordcloud package. For the mask, I used Skyrim Logo Vector. For the font, I used Sovngarde font.

Making the Graph

  1. Some of the data was weird. Argneir had the highest dialogue count due to dialogues of many NPCs (including General Tullius, I think) is assigned to him.
  2. Some of the data doesn’t; produce interesting visualisations. Nords have the highest dialogue count, and after the difference between the first few races and the remaining is so huge that a lot of the races aren’t visible.

Since I had already made this, I thought of sharing it here, in case someone is interested in the image or its code.

Future Plans


Student | Developer | Photographer