A Policy Analytics Platform
QuantGov is an effort to expand the frontiers of economic and legal research by providing an open-source framework to uncover the latent data in legal text. QuantGov grew out of the RegData project, which seeks to solve the longstanding problem of the quantification of federal regulation for use in economic analysis.
The progression from RegData to QuantGov reflects two realizations: first, that the kinds of tools used in RegData can be useful in other contexts, and second, that a clear framework could ease adoption of these methods and allow researchers to build off of one another’s work without constant reinvention of the wheel.
This documentation describes the QuantGov framework, which consists of three parts:
- The corpus, which defines a collection of documents,
- A set of natural language analysis functions that can be applied to the documents in a corpus, and
- A system for training, packaging, and using machine learning estimators to make predictions about the documents in a corpus.
Get the Code
All code for official QuantGov projects is hosted on the QuantGov GitHub page. QuantGov currently maintains four repositories:
- QuantGov/quantgov, the QuantGov Library
- QuantGov/corpus, which holds the default QuantGov corpus skeleton.
- QuantGov/estimator, which holds the default QuantGov estimator skeleton.
- QuantGov/documentation, which holds this documentation.
QuantGov is available on the Python Package Index. It can be installed with the following command:
pip install quantgov
QuantGov is compatible with Python versions 3.3 and up, and should be installable on all operating systems.