Chairs:

Workshop Description

The computational analysis of time is a challenging and very topical problem, as the needs of applications based on information extraction techniques expand to include varying degrees of time stamping and temporal ordering of events and/or relations within a narrative. The challenges derive from the combined requirements of a mapping process (text to a rich representation of temporal entities), representational framework (ontologically-grounded temporal graph), and reasoning capability (combining common-sense inference with temporal axioms).

Usually contextualized in question-answering applications (with obvious dependencies of answers on time), temporal awareness directly impacts numerous areas of NLP and AI: text summarization over events and their participants; making inferences from events in a text; overlaying timelines on document collections; commonsense reasoning in narrative and story understanding.

Interest in temporal analysis and event-based reasoning has spawned a number of important meetings, particularly as applied to IE and QA tasks (cf. at COLING 2000; ACL 2001; LREC 2002; TERQAS 2002; TANGO 2003, Dagstuhl 2005). Significant progress has been made in these meetings, leading to developing a standard for a specification language for events and temporal expressions and their orderings (TimeML). While recent research in the broader community (as indicated, for instance, in the most recent symposium on Annotating and Reasoning about Time and Events) highlights TimeML's status as an interchange format, this workshop, however, is not intended to focus on TimeML exclusively. Likewise, while the ultimate goal of temporal analysis is to facilitate reasoning about time and events, the formal aspects of this problem are being addressed by other meetings (see, for instance, the TIME 2006 Symposium). Instead, the workshop will explore largely the linguistic implications for temporal-analytical frameworks.

The goal of the meeting, therefore, is to address issues already raised, but not fully explored---including but not limited to the following: