Clustering of Concept Drift Categorical Data using Our-NIR Method

Abstract

In the clustering using Ming-Syan Chen NIR method has deficiency that is importance of data labeling and outlier detection. The Our-NIR method introduced to improve Ming-Syan Chen method. In this paper the newly introduced method is taken for comparison to improve the cluster efficiency. To improve the efficiency of clustering by the sampling techniques. However, with sampling applied, those sampled points that are not having their labels after the normal process. Even though there is straight forward method for numerical domain and categorical data. But still it has a problem that is how to allocate those unlabeled data points into appropriate clusters in efficient manner. In this paper the concept-drift phenomenon is studied, and we first propose an adaptive threshold for outlier detection, which is a playing vital role detection of cluster. Second, probabilistic approaches for detection of cluster are proposed using Our-NIR method. Index Terms—Clustering, concept-drift, threshold, weather prediction