An incremental approach for mining all closed intervals from an interval database

In this paper we present an incremental algorithm for mining all the closed intervals from interval dataset. Previous methods for mining closed intervals assume that the dataset is available at the starting of the process, whereas in practice, the data in the dataset may change over time. This paper describes an algorithm, which provides efficient method for mining closed intervals by using a data-structure called CI-Tree (Closed Interval Tree) in dynamically changing datasets. If a new interval is added in the dataset the algorithm modifies the CI-Tree without looking at the dataset. The proposed method is tested with various real life and synthetic datasets.