International Journal of Applied Information Systems (IJAIS)
In data warehouses, views or summaries can be materialized to obtain better performance. In this paper, the near greedy algorithm for views selection is proposed. It is a generalization of the greedy algorithm for views selection and defines a class of solutions existing in the range between the optimal and the greedy solutions. At each of its iterations, the algorithm selects multiple views in a greedy manner instead of just one. The iterations continue until the number of desired views is reached.