UNIGE document Chapitre d'actes
previous document  unige:4542  next document
add to browser collection
Title

Mining Event Histories: A social Scientist View

Author
Published in IASC 2007 Conference on Statistics for data mining, learning and knowledge extraction. Aveiro, Portugal - August 30 - September 1 2007 - . 2007
Abstract Individual longitudinal or sequence data are common to many fields. For instance, they are essential for understanding and predicting the evolution of a patient's disease after it has been diagnosed (survival analysis), the behavior of a visitor of a web site (web log mining), but also for categorizing or clustering signal sequences in domains such as telecommunication. This paper focuses on the analysis of individual longitudinal data within social sciences, especially in population science where we are interested in describing and understanding life courses. A life event can be seen as the change of state of some discrete variable, e.g. the marital status, the number of children, the job, the place of residence. Such life history data are collected in mainly two ways: As a collection of time stamped events or as state sequences. The former is used for instance by survival analysis that focuses on a given type of event and is concerned with its hazard rate or equivalently the duration until it happens. Sequence analysis on the other hand is concerned with the sequencing of the events and is best suited for characterizing whole life trajectories. We consider using datamining-based approaches borrowed from other fields for analysing life courses with both a survival and a sequence perspective. We put stress on the social scientist's expectations and address some of the statistical challenges they raise.
Keywords Event historiesSequencesMining frequent episodesSurvival trees
Note on CDROM
Full text
Proceedings chapter - public document Free access
Structures
Citation
(ISO format)
RITSCHARD, Gilbert. Mining Event Histories: A social Scientist View. In: IASC 2007 Conference on Statistics for data mining, learning and knowledge extraction. Aveiro, Portugal. [s.l.] : [s.n.], 2007. https://archive-ouverte.unige.ch/unige:4542

202 hits

1500 downloads

Update

Deposited on : 2009-12-01

Export document
Format :
Citation style :