Structural variation in generated health reports We present a natural language generator that produces a range of medical reports on the clinical histories of cancer patients, and discuss the problem of conceptual restatement in generating various textual views of the same conceptual content. We focus on two features of our system: the demand for “loose paraphrases ” between the various reports on a given patient, with a high degree of semantic overlap but some necessary amount of distinctive content; and the requirement for paraphrasing at primarily the discourse level. 1

Maintainers often face the daunting task of wading through a collection of both new and old revisions, trying to ferret out those that warrant detailed inspection. Perhaps the most obvious way to rank revisions is by size in terms of lines of code (LOC); this technique has the advantage of being both simple and fast. However, it is well known that the vast majority of revisions are quite small, and so we would like a way of distinguishing between simple and complex changes of the same size. Classical complexity metrics, such as Halstead’s and McCabe’s, could be used but they are hard to apply to code fragments written in multiple programming languages. We propose using the statistical moments of indentation as a lightweight, language independent, revision/diff friendly metric as a proxy for classical complexity metrics. We have evaluated our approach against the entire CVS histories of the 278 of the most popular and most active SourceForge projects. We found that our results are linearly correlated and rank-correlated with traditional measures of complexity, suggesting that measuring indentation is a cheap and accurate proxy for code complexity of revisions. Thus ranking revisions by the standard deviation and summation of indentation yields results that are very similar to ranking revisions by complexity.

The aim of this research is to provide a principled account of the generation of embedded constructions (called parentheticals) and to implement the results in a natural language generation system. Parenthetical constructions are frequently used in texts written in a good writing style and have an important role in text understanding. We propose a framework to model the rhetorical properties of parentheticals based on a corpus study and develop a unified natural language generation architecture which integrates syntax, semantics, rhetorical and document structure into a complex representation, which can be easily extended to handle parentheticals. 1

This paper presents an argument against modularizing linguistic information in natural language generation systems. We argue that complex linguistic constructions require grammatical information to be located in the same module, in order to avoid over-complicating the system architecture. We demonstrate this point by showing how parenthetical constructions — which have only been generated in previous systems using an aggregation or revision module — can be generated by a surface realizer when using an integrated grammar. 1

Abstract not found

and other research outputs Generating multimedia presentations: from plain text to screenplay

and other research outputs Hyper-Document structure: maintaining discourse coherence in non-linear documents

and other research outputs

For guidance on citations see FAQs. c ○ [not recorded] Version: [not recorded] Link(s) to article on publisher’s website:

For guidance on citations see FAQs. c ○ [not recorded] Version: [not recorded] Link(s) to article on publisher’s website: