genealogy-extraction

NLP 482 Final Project

Authors:

    Richa Gadgil
    Nate Andre
    Mateo Ibarguen

Dependencies

pip install -r requirements.txt

If you get the following error:

ERROR: Could not find a version that satisfies the requirement en-core-web-sm==2.2.5 (from -r requirements.txt (line 10))

Make sure to run:

python -m spacy download en

python -m spacy download en_core_web_sm

Data (Optional)

This directory contains a subdirectory called: data which only contains the labels for the data, the trained models, and the features. It does not contain the actual data. The actual Wikipedia articles and data about our parsed entities are stored under the src/wiki_referencer directory. If you wish to re-append this data, download the genealogy-extraction_data.zip file and follow these steps:

1. Download and unzip the file genealogy-extraction_data.zip .
2. Rename the directory to: wiki_referencer
3. Add that directory under src

We have submitted our project so that it already includes the data in its correct location.

Running the genealogy-extraction

python3 main.py

There are two options when running our main.py script:

• If you wish to copy-paste an article, do so in the command line. After you are done, type:

--genealogy

• If you instead wish to test our model on the pre-loaded wikipedia articles, type the following:

--generate

After the features are calculated, a family tree will pop up. Once you close it, the gedcom file will be saved as gedcom.ged in this directory.

In order to exit the program, type:

--exit