Special Issue on Advances in Data Mining and Robust Statistics

The main aim of data mining is to extract knowledge from, usually very large, datasets. Data mining techniques are often applied to gain initial insights about the data and complement statistical models. This special issue focuses on the interface between data mining and statistical modelling, with special emphasis on robust statistics. Very large datasets, especially those that are machine generated and undergo limited quality control, are likely to contain outliers and anomalous measurements. The analysis of such datasets requires statistical approaches that are both computationally efficient and robust against outliers and mild departures from model assumptions. The broad scope includes, but is not limited to, visualization techniques for very large and complex data, including relational data, data analysis algorithms including optimisation and search techniques, methodologies to draw inference on patterns and subgroups, robust models, outlier detection methods, and the analysis of dependencies.