Bounded-Depth High-Coverage Search Space for Noncrossing Parses

Forskningsoutput: Kapitel i bok/rapport/konferenshandlingKonferensbidragVetenskapligPeer review


A recently proposed encoding for non- crossing digraphs can be used to imple- ment generic inference over families of these digraphs and to carry out first-order factored dependency parsing. It is now shown that the recent proposal can be substantially streamlined without information loss. The improved encoding is less dependent on hierarchical processing and it gives rise to a high-coverage bounded-depth approximation of the space of non- crossing digraphs. This subset is presented elegantly by a finite-state machine that recognises an infinite set of encoded graphs. The set includes more than 99.99% of the 0.6 million noncrossing graphs obtained from the UDv2 treebanks through planarisation. Rather than taking the low probability of the residual as a flat rate, it can be modelled with a joint probability distribution that is factorised into two underlying stochastic processes – the sentence length distribution and the related conditional distribution for deep nesting. This model points out that deep nesting in the streamlined code requires extreme sentence lengths. High depth is categorically out in common sentence lengths but emerges slowly at infrequent lengths that prompt further inquiry.
Titel på värdpublikationProceedings of the 13th International Conference on Finite State Methods and Natural Language Processing : FSMNLP 2017
RedaktörerFrank Drewes
Antal sidor11
FörlagThe Association for Computational Linguistics
Utgivningsdatum4 sep. 2017
ISBN (tryckt)978-1-5108-4746-0
StatusPublicerad - 4 sep. 2017
MoE-publikationstypA4 Artikel i en konferenspublikation
EvenemangInternational Conference on Finite State Methods and Natural Language Processing (FSMNLP) - Umeå, Umeå, Sverige
Varaktighet: 5 sep. 20177 sep. 2017
Konferensnummer: 13


  • 6121 Språkvetenskaper
  • 113 Data- och informationsvetenskap
  • 111 Matematik
  • 112 Statistik
  • Finnish TreeBank 1

    Bartis, I. (!!Other), FIN-CLARIN-konsortio, Nykykielten laitos, Helsingin yliopisto, 2015


Citera det här