We present a method combining semantic inferencing and
visual analysis for nding automatically media (photos and
videos) illustrating events. We report on experiments vali-
dating our heuristic for mining media sharing platforms and
large event directories in order to mutually enrich the de-
scriptions of the content they host. Our overall goal is to
design a web-based environment that allows users to explore
and select events, to inspect associated media, and to dis-
cover meaningful, surprising or entertaining connections be-
tween events, media and people participating in events. We
present a large dataset composed of semantic descriptions of
events, photos and videos interlinked with the larger Linked
Open Data cloud and we show the bene ts of using semantic
web technologies for integrating multimedia metadata.