Computational Linguistics with & for Corpora
Corpling@GU is Georgetown University's Computational Corpus Linguistics lab at the Department of Linguistics. We specialize in building:
We are also part of the larger Georgetown University Computational Linguistics (GUCL) community!
The lab maintains two corpus search interfaces, which offer students and the general public access to language data and statistical analysis tools, as well as an online dictionary:
We develop a number of NLP tools that help to build corpora automatically, or feed into manual correction loops:
We provide a number of freely available annotation tools:
Several of our corpora are freely available, open source projects:
The first release of GUM series 7 now adds four more genres to our multilayer corpus, in addition to brand new annotation layers, corrections, and more.
Read more about the addition of entity annotations to the Coptic Universal Dependencies Treebank.
Coptic Scriptorium’s Natural Language Processing (NLP) tools now support two new features...
... in search of discourse signals! We now know a lot about what cues people use to identify discourse relations, but can we teach computers to notice the same signals?
Our EMNLP 2024 paper presents a valuable genre-diverse PDTB-style dataset for English shallow discourse parsing across modalities, text types, and domains using a cascade of conversion modules leveraging enhanced RST annotations, thereby also enabling theoretical studies of discourse relation variation across frameworks
In our Machine Learning for Ancient Languages (ML4AL) workshop paper , we present a bidirectional RNN model for character prediction of Coptic characters in manuscript lacunae and use it to rank the likelihood of various textual reconstructions. A live demo of our models is available here !
Our EACL 2024 paper promotes a strict definition of entity salience by presenting GUMsley, a 12-genre challenge dataset for entity salience evaluation and shows how salient entities added to summarization models are beneficial for deriving higher-quality summaries with fewer hallucinated entities
Check our AACL-IJCNLP 2023 paper about incorporating singletons and mention-based features to improve coreference generalization
Our SIGDIAL 2023 paper on English RST parsing errors examines and models some of the factors associated with parsing difficulties
If you have any questions or feedback please let us know! If you'd like to join us: We accept new PhD and Masters students every year, please contact Amir Zeldes for more information.