LC-QuAD 2.1

<p>A cleaned dataset version of LC-QuAD 2.0 including:</p> <ul> <li>Outliers and duplicated question removal</li> <li>Dummy SPARQL templates for Wikidata and DBpedia 2018</li> <li>Training/development/testing sets used on the paper&nbsp; &quot;A hereditary attentive template-based approach for Complex Knowledge Base Question Answering systems&quot; <ul> <li>Two training sets: one with and without paraphrased questions<span>&nbsp;</span></li> </ul> </li> </ul> <p>The original dataset (LC-QuAD 2.0)&nbsp;can be found at&nbsp;http://lc-quad.sda.tech/</p>