Visualization of writing styles

Thesis Type Bachelor
Thesis Status
Student Thibault Gerrier
Thesis Supervisor
Research Field

Writing styles di?er from person to person. They can be analysed and captured with many stylometric features like the average word length or which words are commonly used. The data resulting from those features are plain values, and this makes it di?fficult to compare texts just by looking at numbers.

In this thesis, a web tool was created to make those features easily accessible by uploading texts and visualizing their styles with the help of those features. Additionally, the texts can be compared to one another in order to show similarities or di?fferences between them or between different authors. The tool was written in JavaScript using the Meteor platform with a React front end and communicates with a Java Library for the feature computation.
This thesis documents and explains the various aspects of the tool as well as the diff?erent technologies used to build it.