Abstract not found

Abstract not found

In this paper we discuss several metrics of coherence defined using Centering Theory and investigate the usefulness of such metrics for information ordering in automatic text generation. We estimate empirically which is the most promising metric and how useful this metric is using a general methodology applied on several corpora. Our main result is that the simplest metric (which relies exclusively on NOCB transitions) sets a robust baseline that cannot be outperformed by other metrics which make use of additional Centering-based features. This baseline can be used for the development of both text-to-text and concept-to-text generation systems. 1.

Although there is an increasing shift towards evaluating Natural Language Generation (NLG) systems, there are still many NLG-specific open issues that hinder effective comparative and quantitative evaluation in this field. The paper starts off by describing a task-based, i.e., black-box evaluation of a hypertext NLG system. Then we examine the problem of glass-box, i.e., module specific, evaluation in language generation, with focus on evaluating machine learning methods for text planning.

In this thesis, we present a user-sensitive approach to text summarization. One domain which would highly benefit from tailoring summaries to both individual and class-based user characteristics is the medical domain, where physicians and patients access similar information, each with their own needs and abilities. Our framework is a medical digital library for physicians and patients. We describe a summarizer, which generates summaries of findings in an input set of clinical studies. When a physician is treating a specific patient, he’s looking for information relevant to the patient’s history and problems. The summarizer takes the user’s interests into account and presents only the findings pertaining to a user model, as approximated by an existing patient record. The same synthesis of information can also be of interest to the patient. The summarizer predicts which medical terms used in a text will be too technical for patients, and augments it with appropriate definitions when necessary. We adopt a generation-like architecture for our summarizer. However, be-cause our input is textual and not semantic, new challenges arise. We operate over

We investigate the problem of generating the structure of short domain independent abstracts. We apply a supervised machine learning approach trained over a set of abstracts collected from abstracting services and automatically annotated with a text analysis tool. We design a set of features for learning inspired from past research in content selection, information ordering, and rhetorical analysis for training an algorithm which then predicts the discourse structure of unseen abstracts. The proposed approach to the problem which combines local and contextual features is able to predict the local structure of the abstracts in just over 60 % of the cases. 1