On Authorship Attribution via Markov Chains and Sequence Kernels
Sanderson, C.
Guenter, S.
Australian Nat. Univ., Canberra, ACT;
This paper appears in: Pattern Recognition, 2006. ICPR 2006. 18th International Conference on
Publication Date: 0-0 0
Volume: 3,
On page(s): 437-440
Location: Hong Kong,
ISSN: 1051-4651
ISBN: 0-7695-2521-0
INSPEC Accession Number: 9209990
Digital Object Identifier: 10.1109/ICPR.2006.899
Posted online: 2006-09-18 09:26:48.0
Abstract
We investigate the use of recently proposed character and word sequence kernels for the task of authorship attribution and compare their performance with two probabilistic approaches based on Markov chains of characters and words. Several configurations of the sequence kernels are studied using a relatively large dataset, where each author covered several topics. Utilising Moffat smoothing, the two probabilistic approaches obtain similar performance, which in turn is comparable to that of character sequence kernels and is better than that of word sequence kernels. The results further suggest that when using a realistic setup that takes into account the case of texts which are not written by any hypothesised authors, about 5000 reference words are required to obtain good discrimination performance
Index
Terms
Available to subscribers and IEEE members.
References
Available to subscribers and IEEE members.
Citing Documents
Available to subscribers and IEEE members.
Cart
