Workshop 'Population Reconstruction'
Date: 19 February 2014 to 21 February 2014
Location: IISH, Amsterdam
The digital availability of the very large data collections from various historical and modern sources describing jigsaw pieces of the lives of individuals, challenges for new methods and approaches to realize large-scale life cycle and population reconstruction. Methods are needed that are capable of weighting all pieces of available information from a wide range of sources while considering contemporary and local circumstances. Such an approach may resemble the intelligent human reasoning in handling fuzzy, variable, erroneous and missing data.
Whereas for several decades the focus in reconstruction studies largely was on data matching of record pairs, promising results on group linking are now emerging. At the dawn of the development of these new and more complex population reconstruction methods, with activities in many countries – sometimes isolated, but also in research networks –, it is timely to bring experts together again to discuss the current state of the art, and to look for opportunities for continued or renewed cooperation.
Workshop Topics (not limited to)
Data pre-processing and cleaning
Standardization (of person names, places, addresses, occupational titles)
Learning from rich data sources
Data structures for multi-source record linkage
Reasoning strategies
Decision rules
Group linking
Probabilistic data matching
Graph-based linking
Historical data mining
Methods for spatio-temporal reasoning
Computational efficiency
Evaluation and benchmarking
Visualisations
Key-note Speakers
Peter Christen, Research School of Computer Science, Australian National University, Canberra
Peter Christen is an Associate Professor in the Research School of Computer Science at the Australian National University in Canberra. His research interests are in data mining and data matching (entity resolution). He is especially interested in the development of scalable and real-time algorithms for data matching, and privacy and confidentiality aspects of data matching and data mining. He has published over 80 papers in these areas, including the book `Data Matching' (2012, Springer), and he is the principle developer of the Febrl (Freely Extensible Biomedical Record Linkage) open source data cleaning, deduplication and record linkage system.
Kris Inwood, University of Guelph, Canada
Kris Inwood is Professor of Economics and History at the University of Guelph (Canada) whose interests lie in the borderlands between history and the social sciences. He contributes to the reconstruction of past populations in order to investigate inequality, productivity and the standard of living. Inwood currently directs the People In Motion project in collaboration with the University of Alberta, University of Minnesota, Laval Université and Université de Montréal.
Arno Knobbe, Leiden Institute of Advanced Computer Science, Leiden, The Netherlands
Arno Knobbe is a senior researcher in the Data Mining Group at the Leiden Institute of Advanced Computer Science. Additionally, he is the owner of a Dutch Data Mining consulting company called Kiminkii. His research is focused on projects that revolve around large collections of data. In these projects, novel and scalable Data Mining techniques are employed to effectively unearth the useful knowledge in these rich sources of data.
Knobbe is involved in the international ChartEx project related to the mining of some ten thousand medieval property records. The project is concerned with the automated annotation of Old English and Latin by means of NLP techniques. The annotated records are then linked to form business and family networks, allowing the subsequent mining of the medieval network for statistical patterns of land ownership in the period.
Paper Submission
We welcome the submission of full papers (with a maximum of 8 pages) and demonstration descriptions. Papers will be peer-reviewed. All accepted papers will be made available electronically at the start of the workshop. Papers will be presented in oral sessions of 20 minutes, followed by 10 minutes of discussion. There will be ample time for demonstrations and general discussions.
Important Dates
Paper submission deadline 15 October 2013
Paper notification 15 November 2013
Final paper 15 January 2014
Workshop 19-21 February 2014
Organizing Committee
Gerrit Bloothooft, Utrecht University
Marijn Schraagen, Leiden University
Kees Mandemakers, IISH Amsterdam / Erasmus University Rotterdam
Contact
Gerrit Bloothooft
Utrecht University, Trans 10, 3512 JK Utrecht, The Netherlands
email: g.bloothooft [at] uu.nl
phone: +31.30.2536042
Kees Mandemakers (kma [at] iisg.nl)
Marijn Schraagen (schraage [at] liacs.nl)