The Segmentation of Data Set Area method in the clustering of uncertain data

The clustering of uncertain objects is a well researched field. This paper is concerned with the clustering of uncertain objects with 2D location uncertainties, due to object movements. The location of a moving object is reported periodically, thus the location is uncertain and is described using a probability density function. Data on moving objects and their locations is placed in distributed databases. The number of objects in a database can be large, thus their proper clustering is a challenging task. A survey of existing clustering methods is given in this paper and a new clustering method is proposed. This method is called Segmentation of Data Set Area. Using this method the execution time of clustering objects is shortened, compared to previous methods. In this method, the data set area is divided into sixteen segments. Each segment is observed separately and only the clusters and objects in a given segment and its neighbouring segments are observed. Experiments were conducted to evaluate the effectiveness of the new method. These experiments proved that this method outperformed previous methods by up to 28% in computing time whilst using the same memory space.