Feature Design | The proposed parsing algorithms both rely on machine learning methods. |
Feature Design | The shift-reduce parser (SRP) trains a machine learning classifier as the oracle 0 E (C —> T) to predict a transition 75 from a parser configuration 0 2 (L1, L2, Q, E), using node features such as the heads of L1, L2 and Q, and edge features from the already predicted temporal relations in E. The graph-based maximum spanning tree (MST) parser trains a machine learning model to predict SCORE(e) for an edge e = (107;, rj, wk), using features of the nodes w, and wk. |
Parsing Models | The oracle 0 is typically defined as a machine learning classifier, which characterizes a parser configuration c in terms of a set of features. |
Parsing Models | The SCORE function is typically defined as a machine learning model that scores an edge based on a set of features. |
Abstract | Evidence from machine learning indicates that increasing the training sample size results in better prediction. |
Introduction | This contradicts theoretical and practical evidence from machine learning that suggests that larger training samples should be beneficial to improve prediction also in SMT. |
Related Work | The focus of many approaches thus has been on feature engineering and on adaptations of machine learning algorithms to the special case of SMT (where gold standard rankings have to be created automatically). |