pydistinto - a Python implementation of different measures of distinctiveness for contrastive text analysis

pydistinto is a further development of pyzeta

What is the purpose?

To facilitate the use of relevant measures that can be used for contrastive text analysis, we are developing a tool written in Python called pydistinto as part of the Zeta and Company project. The goal of our project is to achieve a deeper understanding of the different measures of distinctiveness and to propose improvements for their implementation and use. For this purpose, we created a conceptual framework of different measures of distinctiveness. Based on the framework and Codes from pyzeta, pydistinto has been developed.

Getting help

• The data folder contains some examples of what input pydistinto needs and the working_dir folder contains what output it produces
• The howto.md file contains a brief tutorial of sorts for running analyses with pydistinto.

Requirements and installation

• Requirements: Python 3 with pandas, numpy, sklearn, pygal, spacy