The lack of adequate bases of commonsense or even lexical knowledge is perhaps the main obstacle to the development of highperformance, robust tools for semantic interpretation. It is also generally accepted that, notwithstanding the increasing availability in recent years of substantial hand-coded lexical resources such as WordNet and EuroWordNet, addressing the commonsense knowledge bottleneck will eventually require the development of effective techniques for acquiring such information automatically, e.g., from corpora. We discuss research aimed at improving the performance of anaphora resolution systems by acquiring the commonsense knowledge require to resolve the more complex cases of anaphora, such as bridging references. We focus in particular on the problem of acquiring information about part-of relations.

In this paper we present an extension of a machine learning based coreference resolution system which uses features induced from different semantic knowledge sources. These features represent knowledge mined from WordNet and Wikipedia, as well as information about semantic role labels. We show that semantic features indeed improve the performance on different referring expression types such as pronouns and common nouns. 1

We report on experiments in reference resolution using a decision tree approach. We started with a standard feature set used in previous work, which led to moderate results. A closer

We compare two ways of obtaining lexical knowledge for antecedent selection in other-anaphora and definite noun phrase coreference. Specifically, we compare an algorithm that relies on links encoded in the manually created lexical hierarchy WordNet and an algorithm that mines corpora by means of shallow lexico-semantic patterns. As corpora we use the British National Corpus (BNC), as well as the Web, which has not been previously used for this task. Our results show that (a) the knowledge encoded in WordNet is often insufficient, especially for anaphor– antecedent relations that exploit subjective or context-dependent knowledge; (b) for otheranaphora, the Web-based method outperforms the WordNet-based method; (c) for definite NP coreference, the Web-based method yields results comparable to those obtained using WordNet over the whole data set and outperforms the WordNet-based method on subsets of the data set; (d) in both case studies, the BNC-based method is worse than the other methods because of data sparseness. Thus, in our studies, the Web-based method alleviated the lexical knowledge gap often encountered in anaphora resolution and handled examples with context-dependent relations between anaphor and antecedent. Because it is inexpensive and needs no hand-modeling of lexical knowledge, it is a promising knowledge source to integrate into anaphora resolution systems. 1.

Wikipedia provides a semantic network for computing semantic relatedness in a more structured fashion than a search engine and with more coverage than WordNet. We present experiments on using Wikipedia for computing semantic relatedness and compare it to WordNet on various benchmarking datasets. Existing relatedness measures perform better using Wikipedia than a baseline given by Google counts, and we show that Wikipedia outperforms WordNet on some datasets. We also address the question whether and how Wikipedia can be integrated into NLP applications as a knowledge base. Including Wikipedia improves the performance of a machine learning based coreference resolution system, indicating that it represents a valuable resource for NLP applications. Finally, we show that our method can be easily used for languages other than English by computing semantic relatedness for a German dataset.

GuiTAR is an anaphora resolution system designed to be modular and usable as an off-the-shelf component of a NL processing pipeline. We discuss the system’s design and a preliminary evaluation of the two algorithms implemented in the current version of the system – for definite descriptions and for pronoun resolution. 1.

We present a generative model for unsupervised coreference resolution that views coreference as an EM clustering process. For comparison purposes, we revisit Haghighi and Klein’s (2007) fully-generative Bayesian model for unsupervised coreference resolution, discuss its potential weaknesses and consequently propose three modifications to their model. Experimental results on the ACE data sets show that our model outperforms their original model by a large margin and compares favorably to the modified model. 1

Knowledge of the anaphoricity of a noun phrase might be profitably exploited by a coreference system to bypass the resolution of non-anaphoric noun phrases. Perhaps surprisingly, recent attempts to incorporate automatically acquired anaphoricity information into coreference systems, however, have led to the degradation in resolution performance. This paper examines several key issues in computing and using anaphoricity information to improve learning-based coreference systems. In particular, we present a new corpus-based approach to anaphoricity determination. Experiments on three standard coreference data sets demonstrate the effectiveness of our approach.

Vieira and Poesio (2000) proposed an algorithm for definite description (DD) resolution that incorporates a number of heuristics for detecting discoursenew descriptions. The inclusion of such detectors was motivated by the observation that more than 50 % of definite descriptions (DDs) in an average corpus are discourse new (Poesio and Vieira, 1998), but whereas the inclusion of detectors for non-anaphoric pronouns in algorithms such as Lappin and Leass ’ (1994) leads to clear improvements in precision, the improvements in anaphoric DD resolution (as opposed to classification) brought about by the detectors were rather small. In fact, Ng and Cardie (2002a) challenged the motivation for the inclusion of such detectors, reporting no improvements, or even worse performance. We re-examine the literature on the topic in detail, and propose a revised algorithm, taking advantage of the improved discourse-new detection techniques developed by Uryupina (2003). 1

In this paper we address the issue of automatically assigning information status to discourse entities. Using an annotated corpus of conversational English and exploiting morpho-syntactic and lexical features, we train a decision tree to classify entities introduced by noun phrases as old, mediated, or new. We compare its performance with hand-crafted rules that are mainly based on morpho-syntactic features and closely relate to the guidelines that had been used for the manual annotation. The decision tree model achieves an overall accuracy of 79.5%, significantly outperforming the hand-crafted algorithm (64.4%). We also experiment with binary classifications by collapsing in turn two of the three target classes into one and retraining the model. The highest accuracy achieved on binary classification is 93.1%. 1