Time-Constrained Continuous Subgraph Matching Using Temporal Information for Filtering and Backtracking
./TCM_network <data_graph_file> <query_graph_file> <window_size> # for Netflow
./TCM_snap <data_graph_file> <query_graph_file> <window_size> # for Wikitalk and Superuser
- Each script file in
scripts
directory reproduces the results of the experiments in the paper.- netflow_varying_query_size.sh: Figure 7a, 7b
- wikitalk_varying_query_size.sh: Figure 7c, 7d
- superuser_varying_query_size.sh: Figure 7e, 7f
- netflow_varying_density.sh: Figure 8a, 8b
- wikitalk_varying_density.sh: Figure 8c, 8d
- superuser_varying_density.sh: Figure 8e, 8f
- netflow_varying_window_size.sh: Figure 9a, 9b
- wikitalk_varying_window_size.sh: Figure 9c, 9d
- superuser_varying_window_size.sh: Figure 9e, 9f
- The results will be stored in
results/<exp_name>
directory.
Put graph files into datasets
directory.
Data graph file format is a text format to store a directed temporal data graph.
- The first line of the file should be "t # ID".
- Following lines of "v vertex-ID vertex-label" indicate the vertices in the graph.
- The vertices should be written in the file in ascending order of their IDs, and a vertex ID should be in [0, #vertices - 1].
- Edges are represented as follows:
- Following lines of "e source-ID destination-ID source-port destination-port edge-label sec msec" indicate the edges in the graph for Netflow.
- Following lines of "e source-ID destination-ID edge-label sec" indicate the edges in the graph for Wikitalk and Superuser.
Query graph file format is a text format to store a directed temporal query graph.
- The first line of a graph should be "t # s ID".
- Following lines of "v vertex-ID vertex-label" indicate the vertices in the graph.
- Edges are represented as follows:
- Following lines of "e source-ID destination-ID edge-label source-port destination-port" indicate the edges in the graph for Netflow.
- Following lines of "e source-ID destination-ID edge-label" indicate the edges in the graph for Wikitalk and Superuser.
- The ID of the
$i$ -th edge in the file is$i-1$ .
- Following lines of "b edge-ID1 edge-ID2" indicate that the strict partial order edge-ID1
$\prec$ edge-ID2 holds.