A fast unified optimal route query evaluation algorithm

We investigate the problem of how to evaluate, fast and efficiently, classes of optimal route queries on a massive graph in a unified framework. To evaluate a route query effectively, a large network is partitioned into a collection of fragments, and distances of some optimal routes in the network are pre-computed. Under such a setting, we find a unified algorithm that can evaluate classes of optimal route queries. The classes that can be processed efficiently are called constraint preserving (CP) which include, among others, shortest path, forbidden edges/nodes, -autonomy, and some of hypothetical weight changes optimal route query classes. We prove the correctness of the unified algorithm. We then turn our attention to the optimization of the proposed algorithm. Several pruning and optimization techniques are derived that minimize the search time and I/O accesses. We show empirically that these techniques are effective. The proposed optimal route query evaluation algorithm, with all...

• Type: Conference
• Year: 2007
• Where: CIKM
• Authors: Edward P. F. Chan, Jie Zhang
• Keywords: CIKM 2007, Information Management, Optimal Route, Optimal Route Queries, Route Query