Geoinference predicts the location from which a piece of text was written. The Network Dynamics Geoinference Library is a collection of state-of-the-art geoinference methods for predicting the locations of posts in Twitter. This repository hosts the source code for the reference implementations evaluated in Jurgens et al. (2015), all documentation for the project, and the issue tracker for bugs and feature requests.

• Reference implmentations for many highly-cited geoinference techniques
• A flexible API that makes it easy to build new geoinference methods
• Support for both Java and Python implementations

See our project page for full details of the project. The Installation page has additional for detailed instructions on how to use and extend the software library. Also, see our Frequently Asked Questions for additional details documentation.

This repository connects to a multi-part effort for geoinference in social media. The Network Dynamics FREESR project (described here) aims to allow social media researchers anywhere to test and evaluate their methods on the same datasets. See the [FREESR] website (forthcoming) for full details.

The Geoinference library was made possible through the development efforts of many people.

• David Jurgens, McGill University
• Tyler Finethy, McGill University
• James McCorriston, McGill University
• Yi Tian Xu, McGill University
• Derek Ruths, McGill University

We especially thank the original authors of the papers which are implemented in the library for their work and occasional feedback on how the algorithms were originally implemented.

We kindly ask that if you use this library in a piece of academic work, that you cite the paper associated with it.

@inproceedings{jurgens2015geolocation,
    title={Geolocation Prediction in Twitter Using Social Networks: A Critical Analysis and Review of Current Practice},
    author={David Jurgens and Tyler Finethy and James McCorriston and Xu, Yi Tian and Derek Ruths},
    booktitle={Proceedings of the 9th International AAAI Conference on Weblogs and Social Media (ICWSM)},
    year={2015}
}

If you have discovered a bug in the build software or want to report an error in the library, please create a new Issue on our github page.