Conclusion and Outlook Joakim Nivre Uppsala University Department - - PowerPoint PPT Presentation

▶

Feb 02, 2024 1.03k likes •1.13k views

Conclusion and Outlook Joakim Nivre Uppsala University Department of Linguistics and Philology joakim.nivre@lingfil.uu.se Conclusion and Outlook 1(9) 1. Synthesis 2. Ensemble Systems 3. Domain Adaptation 4. Cross-Language Variation 5.

SLIDE 1

Conclusion and Outlook

Joakim Nivre

Uppsala University Department of Linguistics and Philology joakim.nivre@lingfil.uu.se

Conclusion and Outlook 1(9)

SLIDE 2

1. Synthesis
2. Ensemble Systems
3. Domain Adaptation
4. Cross-Language Variation
5. Unsupervised Parsing

Conclusion and Outlook 2(9)

SLIDE 3

Synthesis ◮ Feature-rich discriminative factored models ◮ Global learning (conditional or discriminative) ◮ Dynamic programming or beam search decoding

Conclusion and Outlook 3(9)

SLIDE 4

Ensemble Systems ◮ Parser outputs (for sentence x): y1, . . . , ym ◮ Arc scores: Score(i, l, j, x) = |{yk : (i, l, j, x) ∈ yk}| ◮ Maximum spanning tree parsing

Conclusion and Outlook 4(9)

SLIDE 5

Domain Adaptation

WSJ Brown Genia SWBD 89.7 84.1 76.2 76.7

Conclusion and Outlook 5(9)

SLIDE 6

Cross-Language Variation

Arabic Basque Catalan Chinese Czech 76.5 76.9 88.7 84.7 80.2 English Greek Hungarian Italian Turkish 89.6 76.3 80.3 84.4 79.8

Conclusion and Outlook 6(9)

SLIDE 7

Cross-Language Variation

Universal Dependencies (http://universaldependencies.github.io/docs/)

◮ Google universal part-of-speech tags ◮ Interset morphological features ◮ Stanford universal dependencies ◮ Guidelines, version 1, October 2014 ◮ First set of treebanks (10 languages), January 2015

Conclusion and Outlook 7(9)

SLIDE 8

Unsupervised Parsing

P(x|y) =

1≤i<j≤n

P(xi, . . . , xj|yij)P(xi−1, xj+1|yij)

Conclusion and Outlook 8(9)

SLIDE 9

Unsupervised Parsing

P(T(h)) =

d∈l,r

 

a∈D(h,d)

P!(¬!|h, d, ?)Pv(a|h, d)P(T(a))   P!(!|h, d, ?)

Conclusion and Outlook 9(9)