![](bg6.png)
tagging Portuguese language sentences was
introduced in this paper. The model showed itself as
an alternative to feedforward neural network part-of-
speech taggers as the former is faster and had better
accuracy rate than the latter. WANN-Tagger also
showed that its accuracy rate is as high as of Hidden
Markov Models in some cases. However, the
WANN-Tagger does not show itself as being
competitive in training time with the Hidden Markov
Model, as it needs a context window and a large
number of inputs for each of its RAMs.
In order to make this model more competitive
with the Hidden Markov Model, the possibility of
adding Markov information to the WANN-Tagger
inputs, for instance the tag obtained during the
tagging phase of the previous word, will be left for
future work. This way, the model would be able to
know more information about the whole sentence and
not only about a small part of it that is within the
context window.
In addition, it is worth noting that WANN-Tagger
performed well with both a free and a fixed word
order language and showed itself as a good model to
be used when training is required during runtime.
ACKNOWLEDGEMENTS
This work was partially supported by CNPq
(306070/2007-3) and FAPERJ (E-26/102.957/2008)
Brazilian research agencies.
REFERENCES
Alexander, I., Thomas, W. V., Bowden, P. A., 1984.
WiSARD: a radical step forward in image recognition.
In Sensor Review, pp. 120-124.
Alexander, I., Kan, W. W., 1987. A probabilistic logic
neuron network for associative learning. In IEEE
Proceedings of the First International Conference on
Neural Networks, pp. 541-548.
Alexander, I., 1990. Ideal neurons for neural computers. In
Parallel Processing in Neural Systems and
Computers, North-Holland, Amsterdam, pp. 225-228.
Alexander, I., Morton, H., 1991. General neural unit:
retrieval performance. In IEE Electronics Letters, 27,
pp. 1176-1178.
Alias-i, 2009. LingPipe 3.9.2 http://alias-i.com/lingpipe
(accessed October 1, 2009).
Baum, L. E., 1972. An inequality and associated
maximization technique in statistical estimation for
probabilistic functions of Markov processes. In
Shinsha, O. (Ed.), Inequalities III: Proceedings of the
3
rd
Symposium on Inequalities, University of
California, Los Angeles, pp. 1-8. Academic Press.
Bick, E., 2000. The Parsing System Palavras – Automatic
Grammatical Analysis of Portuguese in a Constraint
Grammar Framework. Århus University Press.
Brill, E., 1995. Transformation-based error-driven
learning and natural language processing: a case study
in part-of-speech tagging. In Computational
Linguistics, 21(4), pp. 543-566.
Carvalho Filho, E. C. B., Fairhurst, M. C., Bisset, D. L.,
1991. Adaptive pattern recognition using goal-seeking
neurons. In Pattern Recognition Letters, 12, pp. 131-
138.
Dempster, A. P., Laird, N. M., Rubin, D. B., 1977.
Maximum likelihood from incomplete data via the EM
algorithm. In Journal of the Royal Statistical Society,
39(1), pp 1-21.
Francis, W. N., Kučera, H. (1982). Frequency Analysis of
English Usage. Houghton Mifflin, Boston.
Grieco, B. P. A., Lima, P. M. V., De Gregorio, M., França,
F. M. G., 2010. Producing pattern examples from
“mental” images. In Neurocomputing, 73, pp. 1057-
1064.
Kanerva, P., 1988. Sparse Distributed Memory. MIT
Press.
Karlsson, F., Vuotilainen, A., Heikkilä, J., Anttila, A.,
1995. Constraint Grammar: A Language-Independent
System for Parsing Unrestricted Text. Mouton de
Gruyter.
Lafferty, J., McCallum, A., Pereira, F., 2001. Conditional
random fields: probabilistic models for segmenting
and labeling sequence data. In Proceedings of the 18
th
International Conference on Machine Learning.
Linguateca, 2009. CETENFolha http://www.linguateca.pt/
CETENFolha/index_info.html (accessed October 1,
2009).
Marques, N. C., Lopes, G. P., 1996. Using neural nets for
portuguese part-of-speech tagging. In Proceedings of
the 5
th
International Conference on the Cognitive
Science of Natural Language Processing.
McCallum, A., Freitag, D., Pereira, F., 2000. Maximum
entropy Markov models for information extraction and
segmentation. In Proceedings of the 17
th
International
Conference on Machine Learning.
Rabiner, L. R., 1989. A tutorial on hidden Markov models
an selected applications in speech recognition. In
Proceedings of the IEEE, 77(2), pp. 257-286.
Schmid, H., 1994. Part-of-speech tagging with neural
networks. In Proceedings of the 15
th
Conference on
Computational Linguistics, 1, pp. 172-176.
Soares, C. M., da Silva, C. L. F., De Gregorio, M., França,
F. M. G., 1998. Uma implementação em software do
classificador WiSARD. In Proceedings of the V
Simpósio Brasileiro de Redes Neurais (SBRN 1998),
2, pp. 225-229. (In Portuguese).
Villavicencio, A., Marques, N. M., Lopes, J. G. P.,
Villavicencio, F., 1995. Part-of-speech tagging for
portuguese texts. In Proceedings of the 12
th
Symposium on Artificial Intelligence, 991, pp. 323-
332.
WANN-TAGGER - A Weightless Artificial Neural Network Tagger for the Portuguese Language
335