An event is occurrence of an incident at a particular instant of time with a set of attributes describing it. A collection of similar events ordered by time is called an event timeline. This paper presents the notion and design of an event timeline datatype, which can be used to store, maintain and operate on a collection of events as a single abstraction. The datatype represents event occurrences as bitmaps with methods for accessing and updating the timeline. Two additional pair-wise operators union (OR) and intersection (AND) allow correlating timelines to discover new information. Semi-structuring event attributes of a timeline as (name, value) pairs provides a basis for search and indexing. The datatype can be used either to develop simple applications requiring timelines, such as genealogy, patient history, etc. or to develop more involved applications tracing the evolution of a field based on timelines depicting developments in related areas. © 2008 Springer-Verlag Berlin Heidelberg.
cited By (since 1996)0; Conference of org.apache.xalan.xsltc.dom.DOMAdapter@7e9b9bb2 ; Conference Date: org.apache.xalan.xsltc.dom.DOMAdapter@1a0b040e Through org.apache.xalan.xsltc.dom.DOMAdapter@4937407d; Conference Code:72572
Aa Biswas, Sagar, B. Va, and Srinivasan, Jb, “Managing and correlating historical events using an event timeline datatype”, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 4947 LNCS, pp. 604-612, 2008.