Outlier Mining of Drug Database in Chemo Informatics

The task of outlier detection is to find small groups of data objects that are exceptional when compared with rest large amount of data. Recently, the problem of outlier detection in categorical data is defined as an optimization problem. The Clustering and Outlier Analysis are the important techniques in data mining Earlier we implemented clustering technique on GENERAL drugs and ANTI HIV (Human Immunodeficiency Virus) drug databases based on descriptors mentioned using WEKA tool and in this paper we have detected the outliers. This method reduces the number of drugs in the GENERAL drug database as well as ANTI HIV database that need not be clinically tested. The outlier drugs obtained as a result should not be tried as they exhibit exceptional behavior.