Database Reference
In-Depth Information
enter their feedback about their own experiences into the system. This
unique combination of user-based text feedback and mobile sensing is
powerful combination, which provides unprecedented information and
flexibility in terms of combining location information with the social
opinions of other users.
For shopping applications, the ability to perform
recommendations
is
a useful functionality in a wide variety of scenarios. Since spatial lo-
cation is highly corrected to user-buying behavior, it is natural to use
GPS information for such applications. An important observation in
this work is that some items or products (eg. restaurants) are spatial
in nature, whereas others (eg. movies) are non-spatial in nature, since
the user-experience with the product in not locality dependent. Simi-
larly, ratings of a user may sometimes be spatial in nature, when some
locations (eg.
FourSquare
) allow location-based check in and ratings.
The work in [101], which uses location-based ratings for the recommen-
dation process. The
LARS
[101] suppots a taxonomy of three classes
of location-based ratings- (i) spatial ratings for non-spatial items, (ii)
non-spatial ratings for spatial items, and (iii) spatial ratings for spatial
items.
LARS
uses spatial partitioning in order to utilize spatially closer
users for the recommendation process. This maximizes system scalabil-
ity without affecting the recommendation quality. Furthermore, since
users prefer closer locations for the purpose of their buying behavior,
the spatial nature of items is used in order to recommend items which
are closer to querying users. This is modeled with the use of a
travel
penalty
. It has been shown in [101], that these features can be used
either separately or together in order to maximize the effectiveness of
the recommendation process.
8.1.6 Wikitude Augmented Reality Application.
The
wikitude application [191] is designed for mobile phones (such as
Black-
berry
and
iPhone
, and uses the GPS location and the compass within
mobile phones in order to provide an “augmented reality” experience
from the mobile phone, by pointing it in different directions. The appli-
cation is connected with social networking application such as
Facebook
and
Twitter
, and can collect messages, tweets and events from users
within a particular neighborhood, and can be made available to the
user. In addition, by pointing the device in a particular direction, it
may be possible to find useful points of interest such as restaurants,
shopping places, or movie theaters. It is even possible to determine mo-
bile coupons and discounts from shops within a particular neighborhood
with this kind of application.
