Top-k Nearest Keyword Search in Public Transportation Networks

Abstract

Top-k nearest keyword search is important for various applications. However, the existing methods are only applicable to static graphs, not public transportation networks. This is because unlike static graph, public transportation network is a temporal graph where the path in the temporal graph must satisfy the time constraint. Thus, the path which is reachable in the static graph, may not reachable in the temporal graph. Therefore, the methods applicable to static graphs cannot be applied to temporal graphs. In this paper, to solve the top-k nearest neighbor keyword search on public transportation networks, we propose two indexes and two algorithms called Temporal Forward Search (TFS) and Temporal Forward-Backward Search (TFBS) to improve the efficiency. Extensive experiments on the real-world datasets were conducted to show the efficiency of our proposed methods.