Scientific and reference databases on the Web are a primary source of information. However, the data is subject to continuous change and often only the most recent versions are preserved. For many database providers it has become common practice to overwrite existing database states when changes occur and regularly publish new releases of the data on the Web. Failure to archive earlier states of the data may lead to the loss of scientific evidence, and the basis of findings may no longer be verifiable.

Our archiving tool XArch is an archive management system that allows one to maintain, populate, and query archives of multiple database versions. XArch is based on a nested merge approach that efficiently stores multiple database versions in a compact archive. The system allows one to create new archives, to merge new versions of data into existing archives, and execute both snapshot and temporal queries using a declarative query language.

XArch - The XML Archiver project

High costs and loss of reputation caused by data of poor quality made quality assurance and data cleansing hot topics in the business world. Recently, data quality is gaining attention in the scientific community as well. Within this project, we review existing data cleansing methods. We classify data deficiencies that diminish the quality of existing data sources and quality criteria that are affected by these deficiencies. Based on these classifications, we show which cleansing approaches are capable of handling which data deficiencies and quality criteria. We show why existing data cleansing techniques fall short for the domain of genome data and argue that merging overlapping data has outstanding ability to increase data accuracy.

DBIS: Data Cleansing of Genome Data @ Humboldt-Universität zu Berlin

Information integration is often faced with the problem that different data sources represent the same set of real-world objects, but give conflicting values for specific properties of these objects. The main objective of this project is to provide methods that aid the developer of an integrated system over overlapping, but contradicting, sources in the task of resolving value conflicts. In many cases, conflicts between contradicting sources do not occur by chance, but follow some systematic reason. Our goal is to identify such systematic conflicts and outline regular data patterns that occur in conjunction with them. Evaluated by an expert user, the discovered regularities provide insights on possible conflict reasons and help assess the quality of inconsistent values.

DBIS: Conflicts in Data Integration @ Humboldt-Universität zu Berlin

• Wenfei Fan, Floris Geerts, Shuai Ma, Heiko Müller
  Detecting Inconsistencies in Distributed Data
  26th IEEE International Conference on Data Engineering, March 1-6, 2010, Long Beach, California, USA

• Peter Buneman, Heiko Müller, Chris Rusbridge
  Curating the CIA World Factbook
  The International Journal of Digital Curation, Issue 3, Volume 4, 2009 (pdf)

• Heiko Müller, Peter Buneman, Ioannis Koltsidas
  XArch: archiving scientific and reference data
  ACM SIGMOD international conference on Management of data, Vancouver, Canada, 2008, 1295-1298 (poster)
   
• Ioannis Koltsidas, Heiko Müller, Stratis Viglas
  Sorting hierarchical data in external memory for archiving
  Proceedings of the VLDB Endowment, Volume 1 , Issue 1, August 2008

• Heiko Müller, Johann-Christoph Freytag, Ulf Leser
  Describing Differences between Databases
  ACM 15th Conference on Information and Knowledge Management (CIKM), Arlington, VA, USA, 2006 (pdf)
   
• Heiko Müller, Ulf Leser, Johann-Christoph Freytag
  Classification of Contradiction Patterns
  30th Annual Conference of the German Classification Society (GfKl), Berlin, 2006 (pdf)

• Heiko Müller, Melanie Weis, Jens Bleiholder, Ulf Leser
  Erkennen und Bereinigen von Datenfehlern in naturwissenschaftlichen Daten
  Datenbank-Spektrum, Heft 15, November 2005, 26-35 (in german) (pdf)
   
• Michael Mielke, Heiko Müller, Felix Naumann
  Ein Data-Quality-Wettbewerb
  Datenbank-Spektrum, Heft 14, August 2005, 34-37 (in german)
   
• Kristian Rother, Silke Trißl , Heiko Müller, Thomas Steinke, Ina Koch, Robert Preissner, Cornelius Frömmel, Ulf Leser
  Columba: An Integrated Database of Proteins, Structures, and Annotations
  BMC Bioinformatics März 2005, 6(1):81 (pdf) (link)

• P. Rieger, S. Heymann, H. Müller
  Datenbankgestützte Wissensakquisition in den Lebenswissenschaften
  Datenbank-Spektrum, Heft 10, August 2004, pp. 14-21 (in german) (pdf)
   
• Heiko Müller, Peter Rieger, Katja Tham, Johann-Christoph Freytag
  Dynamic information fusion for genome annotation
  Informatik 2004 Workshop über Dynamische Informationsfusion, Ulm, Germany, 2004 (pdf)
   
• Heiko Müller, Ulf Leser, Johann-Christoph Freytag
  Mining for Patterns in Contradictory Data
  Proceedings of the SIGMOD International Workshop on Information Quality for Information Systems (IQIS'04), Paris, France, 2004 (pdf)
   
• Kristian Rother, Heiko Müller, Silke Trissl, Ina Koch, Thomas Steinke, Robert Preissner, Cornelius Frömmel, Ulf Leser
  COLUMBA: Multidimensional Data Integration of Protein Annotations
  International Workshop on Data Integration in Life Sciences (DILS 2004), Leipzig, Germany, 2004 (pdf)

• Heiko Müller
  Semantic Data Cleansing in Genome Databases
  Proc. of the VLDB 2003 PhD Workshop, Berlin, Germany, September 12-13, 2003
   
• Heiko Müller, Felix Naumann, Johann-Christoph Freytag
  Data Quality in Genome Databases
  Proceedings of the Conference on Information Quality (IQ 03), Boston (pdf)

• Heiko Müller
  Describing differences between overlapping databases
  Ph.D. Thesis, Humboldt-Universität zu Berlin, 2009 (link)
   
• Heiko Müller
  Archiving and Maintaining Curated Databases
  Grundlagen von Datenbanken, 2009, 135-139 (pdf)

• Heiko Müller
  XArch - An Archive Management System
  3rd International Digital Curation Conference, Washington, DC, 2007 (poster)
   
• Ioannis Koltsidas, Heiko Müller, Stratis Viglas
  Sorting Hierarchical Data in External Memory
  Technical Report, EDI-INF-RR-1217 (pdf)

• Heiko Müller, Ulf Leser, Johann-Christoph Freytag
  On the Distance of Databases
  HUB-IB-199, Humboldt University Berlin, March 2006 (pdf)

• Heiko Müller, Johann-Christoph Freytag
  Problems, Methods, and Challenges in Comprehensive Data Cleansing
  HUB-IB-164, Humboldt University Berlin, 2003 (pdf)

• Heiko Müller
  Realisierung eines einheitlichen Zugriffs auf molekular­biologische Genomkarten mit Hilfe von CORBA
  Diploma Thesis, Technische Universität Berlin, 2000 (in german)

• Heiko Müller, Ulf Leser
  Integration durch Standards: Erfahrungen mit CORBA in Life Science Research
  Föderierte Datenbanken, Berlin, 1999, 89-102 (in german)