This repository is part of my bachelor’s thesis, which concludes my Bachelor’s degree programme in Information Science at the University of Groningen.

It provides details on the data preprocessing and includes the code used for developing or training and evaluating the cosine-based, dependency-based and hybrid systems. Additionally, it offers interactive demos.

Further details are presented in the respective directories.

To replicate the systems or run the demos, please follow the steps below.

git clone https://github.com/jasperkbos/bs-thesis-is

The systems support Python 3.7 or later. To install all required dependencies, simply run:

cd bs-thesis-is
pip3 install -r requirements.txt

To download the Dutch trained pipeline package, simply run:

python3 -m spacy download nl_core_news_lg