While most prior studies in Location-Based Social Networks (LSBNs)
have mainly centered around areas such as Point-of-Interest
(POI) recommendation and place tag annotation, there exists no
works looking at the problem of associating place-type to venues
in LBSNs. Determining the type of places in location-based social
networks may contribute to the success of various downstream
tasks such as Point-of-Interest recommendation, location search,
automatic place name database creation, and data cleaning.
In this paper, we propose a multi-objective ensemble learning
framework that (i) allows the accurate tagging of places into one
of the three categories: public, private, or virtual, and (ii) identifying
a set of solutions thus oering a wide range of possible
applications. Based on the check-in records, we compute two types
of place features from (i) specic paerns of individual places and
(ii) latent relatedness among similar places. e features extracted
from specic paerns (SP) are derived from all check-ins at a specic
place. e features from latent relatedness (LR) are computed
by building a graph of related places where similar types of places
are connected by virtual edges. We conduct an experimental study
based on a dataset of over 2.7M check-in records collected by crawling
Foursquare-tagged tweets from Twier. Experimental results
demonstrate the eectiveness of our approach to this new problem
and show the strength of taking various methods into account
in feature extraction. Moreover, we demonstrate how place type
tagging can be benecial for place name recommendation services.