Sentence Boundary Verification in Polish Text

Simiński, Krzysztof

doi:10.1007/978-3-540-75175-5_62

Krzysztof Simiński

Abstract:

In this paper the heuristic metod based on phrase analysis is proposed for sentence boundary verification in Polish texts. The decision rules, maximum entropy and neural network as reference methods are compared with the phrase analysis. The results elaborated by the proposed method are more acurate than the reference methods.

Reference:

Krzysztof Simiński, Sentence Boundary Verification in Polish Text, [chapter in] Computer Recognition Systems 2 (Marek Kurzyński, Edward Puchała, Michał Woźniak, Andrzej Żołnierek, eds.), Springer Verlag, 2008, volume 45, pp. 493-499. ([10 pkt])

Bibtex Entry:

@INCOLLECTION{id:Siminski2008Sentence,
  author = {Krzysztof Simi\'{n}ski},
  title = {Sentence Boundary Verification in {P}olish Text},
  booktitle = {Computer Recognition Systems 2},
  publisher = {Springer Verlag},
  year = {2008},
  editor = {Marek Kurzy\'{n}ski and Edward Pucha{\l}a and Micha{\l} Wo\'{z}niak and Andrzej \.{Z}o{\l}nierek},
  volume = {45},
  series = {Advances in Soft Computing},
  address = {Berlin, Heidelberg},
  timestamp = {2008.12.20},
  pages = {493--499},
  doi = {10.1007/978-3-540-75175-5_62},
  abstract = {In this paper the heuristic metod based on phrase analysis is proposed
for sentence boundary verification in Polish texts. The decision rules, maximum
entropy and neural network as reference methods are compared with the phrase
analysis. The results elaborated by the proposed method are more acurate than the
reference methods.},
note = {[10 pkt]},
}