Abstract | ||
---|---|---|
We propose a technique to generate non-projective word orders in an efficient statistical linearization system. Our approach predicts liftings of edges in an unordered syntactic tree by means of a classifier, and uses a projective algorithm for tree linearization. We obtain statistically significant improvements on six typologically different languages: English, German, Dutch, Danish, Hungarian, and Czech. |
Year | Venue | Keywords |
---|---|---|
2012 | EMNLP-CoNLL | non-projective word order,projective algorithm,typologically different language,tree linearization,efficient statistical linearization system,generating non-projective word order,significant improvement,unordered syntactic tree |
Field | DocType | Volume |
Czech,Word order,Computer science,Natural language processing,Artificial intelligence,Classifier (linguistics),Syntax,Danish,Linearization,Machine learning,German,Projective test | Conference | D12-1 |
Citations | PageRank | References |
0 | 0.34 | 29 |
Authors | ||
5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Bernd Bohnet | 1 | 524 | 38.19 |
Anders Björkelund | 2 | 180 | 13.06 |
Jonas Kuhn | 3 | 115 | 13.05 |
wolfgang seeker | 4 | 121 | 10.56 |
Sina Zarrieß | 5 | 35 | 8.65 |