Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Conference contribution/Paper › peer-review
Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Conference contribution/Paper › peer-review
}
TY - GEN
T1 - Using a keyness metric for single and multi document summarisation
AU - El-Haj, Mahmoud
AU - Rayson, Paul
PY - 2013/8
Y1 - 2013/8
N2 - In this paper we show the results of our participation in the MultiLing 2013 summarisation tasks. We participated with single-document and multi-document corpus-based summarisers for both Arabic and English languages. The summarisers used word frequency lists and log likelihood calculations to generate singleand multi document summaries. The single and multi summaries generated by our systems were evaluated by Arabic and English native speaker participants and by different automatic evaluation metrics, ROUGE, AutoSummENG, MeMoG and NPowER. We compare our results to other systems that participated in the same tracks on both Arabic and English languages.Our single-document summarisers performed particularly well in the automatic evaluation with our English singledocumentsummariser performing better on average than the results of the other participants. Our Arabic multi-document summariser performed well in the human evaluation ranking second.
AB - In this paper we show the results of our participation in the MultiLing 2013 summarisation tasks. We participated with single-document and multi-document corpus-based summarisers for both Arabic and English languages. The summarisers used word frequency lists and log likelihood calculations to generate singleand multi document summaries. The single and multi summaries generated by our systems were evaluated by Arabic and English native speaker participants and by different automatic evaluation metrics, ROUGE, AutoSummENG, MeMoG and NPowER. We compare our results to other systems that participated in the same tracks on both Arabic and English languages.Our single-document summarisers performed particularly well in the automatic evaluation with our English singledocumentsummariser performing better on average than the results of the other participants. Our Arabic multi-document summariser performed well in the human evaluation ranking second.
M3 - Conference contribution/Paper
SP - 64
EP - 71
BT - Proceedings of the MultiLing 2013 Workshop on Multilingual Multi-document summarization
PB - Association for Computational Linguistics
CY - Sofia, Bulgaria
ER -