Paper Title
An Approach For Supervised Distance Based Outlier Detection

Abstract
Outlier detection simply refers to the process of finding outlying object from a dataset. It is the object having different properties as compare to the entire set of data. Outlier detection became important subject in different knowledge domains. Size of data is getting doubled every years there is a need to detect outliers in large datasets as early as possible. It is crucial to find outliers from large dataset because of curse of dimensionality. If no of outliers present in dataset there is a possibility of increase false negative rate. This paper presents a brief description of support vector machines. Also we have tried to provide the broad and a comprehensive literature survey of outliers and outlier detection techniques. SVM mainly focus on high dimensionality of data, this method are a specific type of machine learning method which is most widely-used for many statistical learning problems. Keywords— Outliers, high dimensional data. Curse of dimensionality, SVM. False Negative rate.

Author - Vijaya Shambharkar, Vaishali Sahare

Published : Volume-3,Issue-10  ( Oct, 2016 )

	|		PDF	|	Viewed - 46	|	Published on 2016-11-12