Visual event mining from geo-tweet photos

In this paper, we propose a system to mine events visually from the Twitter stream by making use of “geo-tweet photos” which are tweets including both geotags and photos. Some works on event mining which utilize geotagged tweets have been proposed so far. However, they used no images but only textual analysis of tweet texts. In this work, we detect events using visual information as well as textual information. In the experiments, we show some examples of detected events and their photos such as “rainbow”, “fireworks” and “Tokyo firefly festival”.