This repository contains code associated with the following paper.
P. Shoemark*, F. F. Liza*, D. Nguyen, S. A. Hale, B. McGillivray. Room to glo: A systematic comparison of semantic change detection approaches with word embeddings. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP 2019). *These authors contributed equally to the study
The code is documented in the wiki: https://github.com/alan-turing-institute/room2glo/wiki
Trained Twitter embeddings
Trained word embeddings created from the tweets available via the statuses/sample endpoint of the Twitter Streaming API in monthly intervals from 2012 to 2018 are available on Zenodo: https://doi.org/10.5281/zenodo.3527982 . Please refer to the paper and the wiki in this repository for details.
Please open an issue in this repository.