Representing Overlaps in Sequence Labeling Tasks with a Novel Tagging Scheme: Bigappy-Unicrossy


Berk G., Erden B., GÜNGÖR T.

20th International Conference on Computational Linguistics and Intelligent Text Processing, CICLing 2019, La Rochelle, Fransa, 7 - 13 Nisan 2019, cilt.13451 LNCS, ss.622-635, (Tam Metin Bildiri)

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Cilt numarası: 13451 LNCS
  • Doi Numarası: 10.1007/978-3-031-24337-0_44
  • Basıldığı Şehir: La Rochelle
  • Basıldığı Ülke: Fransa
  • Sayfa Sayıları: ss.622-635
  • Anahtar Kelimeler: Bigappy-unicrossy tagging scheme, Gappy 1-level tagging scheme, IOB tagging scheme, Long Short-Term Memory, Multiword expressions
  • Boğaziçi Üniversitesi Adresli: Evet

Özet

Multiword expression (MWE) identification can be handled by using sequence tagging approach accompanied with stochastic models and variants of IOB tagging scheme. In this paper, we introduce a new tagging scheme called bigappy-unicrossy to rise to the challenge of overlapping MWEs. The bigappy-unicrossy tagging scheme is compared with the two other well-known tagging schemes which are IOB2 and gappy 1-level in the verbal multiword expression (VMWE) identification task using bidirectional Long Short-Term Memory model with a Conditional Random Field layer on top (bidirectional LSTM-CRF). Both the bigappy-unicrossy and the gappy 1-level tagging schemes outperform the IOB2 tagging scheme. The bigappy-unicrossy tagging scheme competes with the gappy 1-level tagging scheme. We believe that our tagging scheme will show better performance on corpora with higher frequency of overlapping cases.