Robust Outlier Detection Technique in Data Mining: A Univariate Approach
This work addresses outlier detection challenges in data mining for domain-specific applications, but appears incremental as it combines existing methods.
The paper tackles outlier detection in data mining by proposing a univariate approach as a pre-processing step, followed by K-means clustering to analyze outlier effects on cluster analysis, but does not report specific numerical results.
Outliers are the points which are different from or inconsistent with the rest of the data. They can be novel, new, abnormal, unusual or noisy information. Outliers are sometimes more interesting than the majority of the data. The main challenges of outlier detection with the increasing complexity, size and variety of datasets, are how to catch similar outliers as a group, and how to evaluate the outliers. This paper describes an approach which uses Univariate outlier detection as a pre-processing step to detect the outlier and then applies K-means algorithm hence to analyse the effects of the outliers on the cluster analysis of dataset.