OCCAM Open Source Project


The official site for the development of the OCCAM open source software.

View the Project on GitHub github.com/occam-ra/occam

What is OCCAM?

OCCAM is software that implements Reconstructibility Analysis. OCCAM combines data mining, machine learning, and statistical analysis in a unique way which lets a user quickly understand the structure of their data, evaluate many possible models, and compare those models under statistical and information-theoretic criteria. Once a model is selected, OCCAM will fit that model to the data, identifying a simplified representation of the data structure that still preserves its essential features, and allow the user to explore and understand variable relationships and make predictions.

How does it work?

A typical OCCAM workflow might look like this:

Who can use it?

Data scientists; genomics and health data researchers; economists, sociologists, and other social scientists; systems and complexity science researchers; and in general anyone who is working with categorical data. See the RA and OCCAM research page at Portland State University for more on applications and past and present research using OCCAM.