This module provides an IPython Notebook to analyze existing disambiguation with respect to metrics elaborated on in the COLING 2016 paper called: Semantic overfitting: what `world' do we consider when evaluating disambiguation of text?"
This module was written in Python 3. After cloning this repository, the only external dependency is Anaconda.
Please visit Demo.ipynb for an example of the analyses of the datasets.
In datasets, you find the datasets in the format that we use. Each row contains six fields:
- lexical expression
- document creation time
- resource ambiguity
- resource variance