JRS at Search and Hyperlinking of Television Content Task

This paper describes the work done by the JRS team for the linking sub-task. We submitted eight pairs of runs: four with different textual resources only, two using reranking based on visual similarity, and two using concept detection results. Each of the pairs contains of one run using the anchor segment only, and one using a longer context segment. The results show higher variance between anchors than for the 2013 task, also the differences between runs using different textual resources are more salient. The use of the context does not generally improve results, and visual reranking provides small improvements.