This repository contains the source code used to generate my model for the pan 2020 authorship verification shared task: https://pan.webis.de/clef20/pan20-web/author-identification.html
The training datasets are not included in the repository and can be downloaded from: https://pan.webis.de/data.html
I have included the trained models that were used in my submission.
Our approach is described in:
Janith Weerasinghe and Rachel Greenstadt. Feature Vector Difference based Neural Network and Logistic Regression Models for Authorship Verification—Notebook for PAN at CLEF 2020. In Linda Cappellato, Carsten Eickhoff, Nicola Ferro, and Aurélie Névéol, editors, CLEF 2020 Labs and Workshops, Notebook Papers, September 2020. CEUR-WS.org. https://pan.webis.de/downloads/publications/papers/weerasinghe_2020.pdf
@InProceedings{weerasinghe:2020,
author = {Janith Weerasinghe and Rachel Greenstadt},
booktitle = {{CLEF 2020 Labs and Workshops, Notebook Papers}},
crossref = {pan:2020},
editor = {Linda Cappellato and Carsten Eickhoff and Nicola Ferro and Aur{\'e}lie N{\'e}v{\'e}ol},
month = sep,
publisher = {CEUR-WS.org},
title = {{Feature Vector Difference based Neural Network and Logistic Regression Models for Authorship Verification---Notebook for PAN at CLEF 2020}},
url = {},
year = 2020
}