Counting What Counts: Decompounding for Keyphrase Extraction

A core assumption of keyphrase extraction is that a concept is more important if it is mentioned more often in a document. Especially in languages like German that form large noun compounds, frequency counts might be misleading as concepts “hidden” in compounds are not counted. We hypothesize that using decompounding before counting term frequencies may lead to better keyphrase extraction. We identified two effects of decompounding: (i) enhanced frequency counts, and (ii) more keyphrase candidates. We created two German evaluation datasets to test our hypothesis and analyzed the effect of additional decompounding for keyphrase extraction.

Preview

Category

Conference:
ACL 2015 Workshop on Novel Computational Approaches to Keyphrase Extraction, July 30, 2015 Beijing, China
Date Issued:
2015
URN:
urn:nbn:de:hbz:464-20211022-111641-1
DOI:
10.18653/v1/W15-3603
Language:
English
Type of Resource:
Text
Collection:
E-Publications
Subject categories of the Deutsche Nationalbibliografie:
004 Data processing Computer sciences
URL:
https://www.aclweb.org/anthology/W15-3603
Hosting institution:
Faculty of Engineering, Computer Science and Applied Cognitive Science, Computer Science, Language Technology Lab
Info on primary publication:
Erbs, N., Santos, P.B., Zesch, T., Gurevych, I. (2015). Counting What Counts: Decompounding for Keyphrase Extraction. In: Proceedings of the ACL 2015 Workshop on Novel Computational Approaches to Keyphrase Extraction, pp. 10-17. Association for Computational Linguistics. https://doi.org/10.18653/v1/W15-3603

Cite

Citation style:
Could not load citation form.

Rights

Use and reproduction:
This work may be used under a
CC BY-NC-SA 3.0 LogoCreative Commons Attribution - NonCommercial - ShareAlike 3.0 License (CC BY-NC-SA 3.0)
.