Identifying points of interest by self-tuning clustering

Abstract

Deducing trip related information from web-scale datasets has received very large amounts of attention recently. Identifying points of interest (POIs) in geo-tagged photos is one of these problems. The problem can be viewed as a standard clustering problem of partitioning two dimensional objects. In this work, we study spectral clustering which is the first attempt for the POIs identification. However, there is no unified approach to assign the clustering parameters; especially the features of POIs are immensely varying in different metropolitans and locations. To address this, we are intent to study a self-tuning technique which can properly assign the parameters for the clustering needed.Besides geographical information, web photos inherently store rich information. These information are mutually influenced each others and should be taken into trip related mining tasks. To address this, we study reinforcement which constructs the relationship over multiple sources by iterative learning. At last, we thoroughly demonstrate our findings by web scale datasets collected from Flickr.

Extraction of Tourist Destinations and Comparative Analysis of Preferences Between Foreign Tourists and Domestic Tourists on the Basis of Geotagged Social Media Data

“…Crandall et al [5], Yang et al [6], and Zhou et al [7] propose methods for extracting landmarks by clustering locations where photos are taken based on data from Flickr. Wei et al [8] propose a method to construct popular routes from uncertain trajectories on the basis of data obtained from Foursquare.…”

“…Many studies attempt to research tourist mobility based on geotagged social media data; for example, the extraction of the locations of popular touristic sites [5][6][7], the extraction of popular routes for tourists [8], and recommendations of touristic sites and routes [9][10][11]. Other studies compare inbound tourist mobility and domestic tourist mobility [12,13].…”

A Novel Popular Tourist Attraction Discovering Approach Based on Geo-Tagged Social Media Big Data

“…Another more popular spatial clustering method is the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm, which has the advantage of requiring less field knowledge and enables finding irregularly-shaped clusters [12]; Kisilevich et al introduced an adaptive density-based clustering method named P-DBSCAN [13], which is designed on the basis of the DBSCAN algorithm. Moreover, Yang used a spectral clustering method in attraction mining [14,15]; the advantage of this method is that the number of clusters can automatically be adjusted. Zheng and Yuan studied how to mine points of interest and popular attractions using GPS trajectory data and proposed a hierarchical algorithm [16][17][18].…”

Georeferencing Flickr resources based on textual meta-data

“…Finally, in [16], GeoSR is presented as a way of measuring the semantic relatedness of Wikipedia articles based on their geographic context, allowing users to explore information in Wikipedia that is relevant to a particular location. In [41], one would like to discover points of interests based on geotagged photos by applying a form of spectral clustering. The problem with this approach is that there is no unified way for determining the appropriate parameters for the clustering algorithm.…”