Efficient Approximation Algorithms for the Diameter-Bounded Max-Coverage Group Steiner Tree Problem

<p>&nbsp;It contains all the data used in our experiments, including 5 real graphs (<code>MONDIAL</code>,&nbsp;<code>OpenCyc</code>,&nbsp;<code>LinkedMDB</code>,&nbsp;<code>YAGO</code>, and&nbsp;<code>DBpedia</code>) and 3 synthetic graphs (<code>LUBM-50K</code>,&nbsp;<code>LUBM-500K</code>, and&nbsp;<code>LUBM-5M</code>).</p> <p>Each real KG directory contains 8&nbsp;files, including:</p> <ul> <li><code>graph.txt</code>: The first value is the number of vertices. Then each line &#39;u v&#39; means there is an undirected edge between &#39;u&#39; and &#39;v&#39;.</li> <li><code>Weightgraph.txt</code>: The first value is the number of vertices. Then each line &#39;u v w&#39; means there is an undirected edge between &#39;u&#39; and &#39;v&#39; weighted by &#39;w&#39; which is computed by the Informativeness-based Weighting (IW) scheme.</li> <li><code>nodeName.txt</code>: Mapping from vertex ID to vertex name (i.e., entity URI).</li> <li><code>query.txt</code>: Each line is a keyword query containing a set of keyword names.</li> <li><code>kwName.txt</code>: Mapping from keyword ID to keyword name.</li> <li><code>kwMap.txt</code>: Mapping from keyword ID to vertex IDs. The first value of each line is keyword ID, and the rest are vertex IDs.</li> <li><code>UWHBLL.txt</code>: The HBLL index file which was built based on the Unit Weighting.</li> <li><code>IWHBLL.txt</code>: The HBLL index file which was built based on the Informativeness-based Weighting.</li> </ul> <p>Each synthetic directory contains 6&nbsp;files, including:</p> <ul> <li><code>graph.txt</code>: same as above.</li> <li><code>Weightgraph.txt</code>: same as above.</li> <li><code>nodeName.txt</code>: same as above.</li> <li><code>queryList.txt</code>: Each line contains a set (separated by &#39;,&#39;) of sets of vertex IDs.</li> <li><code>UWHBLL.txt</code>: same as above.</li> <li><code>IWHBLL.txt</code>: same as above.</li> </ul> <p>Apart from that, <code>Dbpedia</code> and <code>LUBM-5M</code> also contain a <code>PLLlabel.txt</code> file which was the supplementary file for the HBLL index.</p>