A New Density-based Spatial Clustering Algorithm for Extracting Attractive Local Regions in Georeferenced Documents

Nowadays, with the increasing attention being paid to social media, a huge number of georeferenced documents, which include location information, are posted on social media sites via the Internet. People have been transmitting and collecting information through these georeferenced documents. Georeferenced documents are usually related to not only personal topics but also local topics and events. Therefore, extracting “attractive” local regions associated with local topics from georeferenced documents is one of the most important challenges in different application domains. In this paper, a novel spatial clustering algorithm, called the (ǫ, σ)-densitybased spatial clustering algorithm, for extracting “attractive” local regions in georeferenced documents is proposed. We defined a new type of spatial cluster called an (ǫ, σ)-densitybased spatial cluster. The proposed clustering algorithm can recognize not only semantically-separated but also spatiallyseparated spatial clusters. To evaluate our proposed clustering algorithm, geo-tagged tweets posted on the Twitter site are used. The experimental results show that the (ǫ, σ)-densitybased spatial clustering algorithm can extract “attractive” local regions as (ǫ, σ)-density-based spatial clusters.