WikiRank: Improving Keyphrase Extraction based on Background Knowledge

Yang Yu and Vincent Ng.
Proceedings of the 11th International Conference on Language Resources and Evaluation, pp. 3723-3727, 2018.

Click here for the PDF version.

Abstract

Keyphrase is an efficient representation of the main idea of documents. While background knowledge can provide valulable information about documents, they are rarely incorporated in keyphrase extraction methods. In this paper, we propose WikiRank, an unsupervised method for keyphrase extraction based on background knowledge from Wikipedia. Firstly, we construct a semantic graph for the document. Then we transform the keyphrase extraction problem into an optimization problem on the graph. Finally, we get the optimal keyphrase set to be the output. Our method obtains improvements over other state-of-the-art models by more than 2% in F1-score.

BibTeX entry

@InProceedings{Yu+Ng:18a,
  author = {Yang Yu and Vincent Ng},
  title = {{WikiRank}: Improving Keyphrase Extraction Based on Background Knowledge},
  booktitle = {Proceedings of the 11th International Conference on Language Resources and Evaluation},
  pages = {3723--3727},
  year = 2018}