Euredit, a four-year research project using statistical methods for data mining, is funded with millions of dollars by the European Union and is creating exciting possibilities for many types of bioinformatics work, including microanalysis and forecasting of trends. Euredit's aim was analysis of census surveys, which generally contain missing or incorrect data. European governments backed research to determine and improve statistical techniques that could be used to clean the data by filling in omissions or highlighting errors. The cleaned data generated through the research were used to improve estimation of the need for services and programs in communities. Emerging from the research were various new algorithms pertinent to disciplines where data mining is significant, and bioinformatics was possibly one of the main beneficiaries. Methods developed by the Euredit research project eliminated the need to store n-by-n similarity matrices, which allowed researchers to study much larger data sets. Such developments could also be used in bioinformatics. Also described are Euredit's development of regression models with out-of-core optimization (which eliminated the need to store whole data sets in computer memory) and methods of identification of unusual cases hidden in data (outliers). The Numerical Algorithms Group (NAG) created the first commercially available data mining application toolkit that uses the new algorithms, which are provided as components that can be included in a user's exiting applications without difficulty.