Identifying (and fixing) homonymous and synonymous author profiles is one of the major tasks of curating personalized bibliographic metadata repositories like the dblp computer science bibliography. In this paper, we present and evaluate a machine learning approach to identify homonymous author bibliographies using a simple multilayer perceptron setup. We train our model on a novel gold-standard data set derived from the past years of active, manual curation at the dblp computer science bibliography.
Cite
@article{arxiv.1806.06017,
title = {Homonym Detection in Curated Bibliographies: Learning from dblp's Experience (full version)},
author = {Marcel R. Ackermann and Florian Reitz},
journal= {arXiv preprint arXiv:1806.06017},
year = {2018}
}
Comments
13 pages, 5 figures, short paper version accepted at TPDL 2018