A comparison of summarisation methods based on term specificity estimation
Constantin Orăsan, Viktor Pekar, Laura Hasler
Research Group in Computational Linguistics, School of Humanities, Languages and Social Sciences, University of Wolverhampton Stafford St., Wolverhampton, WV1 1SB, United Kingdom
In automatic summarisation, knowledge poor methods do not necessarily perform worse than those which employ several knowledge sources to produce a summary. This paper presents a comprehensive comparison of several summarisation methods based on term specificity estimation in order to find out which one performs best. Parameters such as quality of the summary produced and the resources required to produce accurate results are considered in order to find out which of these methods is more appropriate for a real world application. Intrinsic and extrinsic evaluation indicates that TF*RIDF, a variant of the commonly used TF*IDF, is the best performing method.
automatic summarization, evaluation, term specificity summarization