Metatone Touch-Screen Performance Analysis

This repository contains data from collaborative touch-screen performances on iPads that were recorded from 2013-2015 as well as scripts in R and Python for performing statistical analysis on these performances.

Important files:

• data contains the touch-screen data and post-hoc gestural classifications for each performance in the archive
• metadata contains meta-data about each performance included in the repository
• analysis contains python scripts for generating & plotting the transition sequences
  • metatone_post_hoc_analysis.py is the main analysis script, which performs gestural classification on these performance logs, and measure the transition probabilities between these gestures
  • metatone-performance-data.csv is the output file from the analysis script and contains flux and entropy measures on the transition matrices of each performance.
• legacy analysis scripts are in R

Jupyter Notebooks

Some jupyter notebooks in the top level of the repository (e.g., generate_metatone_tinyperf_corpus_differential.ipynb) are used to transform the corpus data and use for other projects.