Abstract | ||
---|---|---|
In this paper, we present a systematic evaluation of a hybrid approach of combined rule-based filtering and machine learning to Dutch coreference resolution. Through the application of a selection of linguistically-motivated negative and positive filters, which we apply in isolation and combined, we study the effect of these filters on precision and recall using two different learning techniques: memory-based learning and maximum entropy modeling. Our results show that by using the hybrid approach, we can reduce up to 92% of the training material without performance loss. We also show that the filters improve the overall precision of the classifiers leading to higher F-scores on the test set. |
Year | DOI | Venue |
---|---|---|
2007 | 10.1007/978-3-540-71412-5_10 | DAARC |
Keywords | Field | DocType |
systematic evaluation,overall precision,performance loss,maximum entropy modeling,different learning technique,positive filter,hybrid approach,dutch coreference resolution,memory-based learning,higher f-scores,machine learning,maximum entropy model,rule based | Coreference,Data-driven,Pattern recognition,Computer science,Precision and recall,Filter (signal processing),Artificial intelligence,Principle of maximum entropy,Machine learning,Test set | Conference |
Volume | ISSN | Citations |
4410 | 0302-9743 | 3 |
PageRank | References | Authors |
0.43 | 17 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Iris Hendrickx | 1 | 285 | 30.91 |
Veronique Hoste | 2 | 138 | 16.92 |
Walter Daelemans | 3 | 2019 | 269.73 |