This paper considers the problem of automatic assessment of local coherence. We present a novel entity-based representation of discourse which is inspired by Centering Theory and can be computed automatically from raw text. We view coherence assessment as a ranking learning problem and show that the proposed discourse representation supports the effective learning of a ranking function. Our experiments demonstrate that the induced model achieves significantly higher accuracy than a state-of-the-art coherence model. 1

In recent years tree kernels have been proposed for the automatic learning of natural language applications. Unfortunately, they show (a) an inherent super linear complexity and (b) a lower accuracy than traditional attribute/value methods.

Despite much recent progress on accurate semantic role labeling, previous work has largely used independent classifiers, possibly combined with separate label sequence models via Viterbi decoding. This stands in stark contrast to the linguistic observation that a core argument frame is a joint structure, with strong dependencies between arguments. We show how to build a joint model of argument frames, incorporating novel features that model these interactions into discriminative loglinear models. This system achieves an error reduction of 22 % on all arguments and 32 % on core arguments over a stateof-the art independent classifier for gold-standard parse trees on PropBank.

This paper focuses on semantic role labeling using automatically-generated syntactic information. A simple and robust strategy for system combination is presented, which allows to partially recover from input parsing errors and to significantly boost results of individual systems. This combination scheme is also very flexible since the individual systems are not required to provide any information other than their solution. Extensive experimental evaluation in the CoNLL-2005 shared task framework supports our previous claims. The proposed architecture outperforms the best results reported in that evaluation exercise.

In many prediction tasks, selecting relevant features is essential for achieving good generalization performance. Most feature selection algorithms consider all features to be a priori equally likely to be relevant. In this paper, we use transfer learning — learning on an ensemble of related tasks — to construct an informative prior on feature relevance. We assume that features themselves have meta-features that are predictive of their relevance to the prediction task, and model their relevance as a function of the meta-features using hyperparameters (called meta-priors). We present a convex optimization algorithm for simultaneously learning the meta-priors and feature weights from an ensemble of related prediction tasks that share a similar relevance structure. Our approach transfers the meta-priors among different tasks, allowing it to deal with settings where tasks have non-overlapping features or where feature relevance varies over the tasks. We show that transfer learning of feature relevance improves performance on two real data sets which illustrate such settings: (1) predicting ratings in a collaborative filtering task, and (2) distinguishing arguments of a verb in a sentence. 1.

Abstract not found

We present a four-step hierarchical SRL strategy which generalizes the classical two-level approach (boundary detection and classification). To achieve this, we have split the classification step by grouping together roles which share linguistic properties (e.g. Core Roles versus Adjuncts). The results show that the nonoptimized hierarchical approach is computationally more efficient than the traditional systems and it preserves their accuracy. 1

We present a model for semantic role labeling that effectively captures the linguistic intuition that a semantic argument frame is a joint structure, with strong dependencies among the arguments. We show how to incorporate these strong dependencies in a statistical joint model with a rich set of features over multiple argument phrases. The proposed model substantially outperforms a similar state-of-the-art local model that does not include dependencies among different arguments. We evaluate the gains from incorporating this joint information on the Propbank corpus, when using correct syntactic parse trees as input, and when using automatically derived parse trees. The gains amount to 24.1 % error reduction on all arguments and 36.8 % on core arguments for gold-standard parse trees on Propbank. For automatic parse trees, the error reductions are 8.3 % and 10.3 % on all and core arguments, respectively. We also present results on the CoNLL 2005 shared task data set. Additionally, we explore considering multiple syntactic analyses to cope with parser noise and uncertainty.

In this paper we introduce a semantic role labeling system constructed on top of the full syntactic analysis of text. The labeling problem is modeled using a rich set of lexical, syntactic, and semantic attributes and learned using one-versus-all AdaBoost classifiers. Our results indicate that even a simple approach that assumes that each semantic argument maps into exactly one syntactic phrase obtains encouraging performance, surpassing the best system that uses partial syntax by almost 6%. 1

This paper presents a new method for extracting meaningful relations from unstructured natural language sources. The method is based on information made available by shallow semantic parsers. Semantic information was used (1) to enhance a dependency tree kernel; and (2) to build semantic dependency structures used for enhanced relation extraction for several semantic classifiers. In our experiments the quality of the extracted relations surpassed the results of kernel-based models employing only semantic class information. 1