PRICOMOB: Principle Component Analysis as Applied to Qualitative Analysis of Mobile Measurement and Monitoring Data Sets
NKS-B Research Area
Mobile measurement systems are the backbone of most countries response arsenal in the case of orphan sources and the variety of contexts in which they may be encountered. Conducting mobile measurement surveys, irrespective of the platform utilised, is a non-trivial task with respect to the nature of the data being accrued – large volumes of discrete, often highly variable, data points where the signal of interest may be weak, superimposed on a constantly fluctuating background and only present for a tiny proportion of the overall data set. The operator is often expected to find these signals of interest using purely visual means and where automatic means of signal location often lack the sensitivity to detect a weak signal or are prone to false positives due to the nature of the background radiation environment. Similar problems are encountered in fixed monitoring applications, where the variability of the background is due to changing environmental conditions. Principal Component Analysis (PCA), one of the most popular multivariate statistical technique, is a flexible statistical procedure that allows for the summarizing of the information content in large data tables by means of a smaller set of “summary indices” that can be more easily visualized and analyzed in order to observe trends, jumps, clusters and outliers. The PRICOMOB proposal will focus on the application of PCA to mobile measurement data to assess its performance in identifying source signals from a number of isotopes superimposed on a variable background signal typical of mobile measurement data. The application of PCA to this field brings a powerful statistical technique to bear on data sets that are, relative to laboratory based gamma ray spectrometry, somewhat complex and that present challenges on a number of fronts.