Identification Of Outliers In Oxazolines AND Oxazoles High Dimension Molecular Descriptor Dataset Using Principal Component Outlier Detection Algorithm And Comparative Numerical Study Of Other Robust Estimators

摘要

From the past decade outlier detection has been in use. Detection of outliersis an emerging topic and is having robust applications in medical sciences andpharmaceutical sciences. Outlier detection is used to detect anomalousbehaviour of data. Typical problems in Bioinformatics can be addressed byoutlier detection. A computationally fast method for detecting outliers isshown, that is particularly effective in high dimensions. PrCmpOut algorithmmake use of simple properties of principal components to detect outliers in thetransformed space, leading to significant computational advantages for highdimensional data. This procedure requires considerably less computational timethan existing methods for outlier detection. The properties of this estimator(Outlier error rate (FN), Non-Outlier error rate(FP) and computational costs)are analyzed and compared with those of other robust estimators described inthe literature through simulation studies. Numerical evidence based Oxazolinesand Oxazoles molecular descriptor dataset shows that the proposed methodperforms well in a variety of situations of practical interest. It is thus avaluable companion to the existing outlier detection methods.