Outlier Mining for Removing the Anomalies in High Dimensional Data Using ARVDH Algorithm
Krupa Mary Jacob, K.Sangeetha, S.Karthik?
Journal Title:International Journal of Computer Science and Mobile Computing - IJCSMC
In Data mining outliers are one of the main threats for efficient information retrieval from databases. Outliers are also known as Anomalies. Mining of outliers from the normal data is very important and scope of this is very high. Anomaly detection can be found in applications such as credit card fraud detection, intrusion and insider threat detection in cyber-security, detection of fault, or malignant diagnosis. Anomalous data present in database is harmful for the processing of information and usage of that information. Viscous data contain erroneous information and it may contain dangerous code for carking the whole system where it is stored. The main drawback of the existing system is, it does not support data with Multiclustering for removing viscous data. To avoid this problem we propose one algorithm which is Algorithm for Removing the Viscous data in High Dimensional data (ARVDH). Simple and efficient steps are used to remove outliers form information