Sequence Labelling

The task "Emerging and Rare entity recognition" that is based on the Workshop on Noisy User-generated Text (W-NUT) was used to train a Name Entity Recognition (NER) classifier

Hyperparameter optimization and engineering of additional features were performed in order to experiment and evaluate their effect on the quality of labelling.

Running the experiments

In order to run the experiments for hyperparameter optimization, and feature engineering, use the following command:

$ python main.py

Statistical Analysis

For statistics related to the given data set, the following command should be executed:

$ python statistical.py

Collaborators

• Andreas Paraskeva
• Pavlos Zakkas