Abstract

Trajectory data is ubiquitous in many real time areas. Sequential collections of movements of objects is
called trajectory. Trajectory data clustering is one of the most important and useful area of many modern
trajectory data applications. Trajectory data represents the actual mobility of a diversity of dynamically
moving objects, such as people, flights, birds, vehicles and animals. Many techniques have been
proposed in the literature of trajectory data mining for processing, managing and mining trajectory data
in the past. Most important tasks of trajectory data mining are - trajectory data preprocessing, trajectory
data management, and also different varieties of trajectory data mining tasks such as trajectory pattern
mining, outlier detection, and trajectory classification. Time complexity of many clustering data
algorithms is O(n2
). A new trajectory similarity measure is proposed for clustering trajectory data sets.
The time complexity of proposed trajectory data clustering algorithm is O(n) which is very much better
than time complexities of many real time clustering algorithms