Showing 1–2 of 2 results for author: Wachenchauzer, R

Variations of the Similarity Function of TextRank for Automated Summarization

Authors: Federico Barrios, Federico López, Luis Argerich, Rosa Wachenchauzer

Abstract: This article presents new alternatives to the similarity function for the TextRank algorithm for automatic summarization of texts. We describe the generalities of the algorithm and the different functions we propose. Some of these variants achieve a significative improvement using the same metrics and dataset as the original publication. This article presents new alternatives to the similarity function for the TextRank algorithm for automatic summarization of texts. We describe the generalities of the algorithm and the different functions we propose. Some of these variants achieve a significative improvement using the same metrics and dataset as the original publication. △ Less

Submitted 10 February, 2016; originally announced February 2016.

Comments: 8 pages, 2 figures. Presented at the Argentine Symposium on Artificial Intelligence (ASAI) 2015 - 44 JAIIO (September 2015)

ACM Class: I.2.7

Journal ref: 44 JAIIO - ASAI 2015 - ISSN: 2451-7585, pages 65-72

The intrinsic complexity of parametric elimination methods

Authors: J. Heintz, G. Matera, L. M. Pardo, R. Wachenchauzer

Abstract: This paper is devoted to the complexity analysis of a particular property, called "algebraic robustness" owned by all known symbolic methods of parametric polynomial equation solving (geometric elimination). It is shown that any parametric elimination procedure which owns this property must neccessarily have an exponential sequential time complexity. This paper is devoted to the complexity analysis of a particular property, called "algebraic robustness" owned by all known symbolic methods of parametric polynomial equation solving (geometric elimination). It is shown that any parametric elimination procedure which owns this property must neccessarily have an exponential sequential time complexity. △ Less

Submitted 9 July, 1997; originally announced July 1997.

Comments: LaTeX, 12 pages. To appear in Proc. of WAIT'97, Buenos Aires, Argentina, 1997

Report number: Preprint University of Cantabria, Spain