Contact at mumbai.academics@gmail.com or 8097636691
Responsive Ads Here

Tuesday, 13 February 2018

LARS An Efficient and Scalable Location Aware Recommender System(2014)


LARS An Efficient and Scalable

Location Aware Recommender System(2014)


ABSTRACT:
This paper proposes LARS*, a location-aware recommender system that uses location-based ratings to produce recommendations. Traditional recommender systems do not consider spatial properties of users nor items; LARS*, on the other hand, supports a taxonomy of three novel classes of location-based ratings, namely, spatial ratings for non-spatial items, non-spatial ratings for spatial items, and spatial ratings for spatial items. LARS* exploits user rating locations through user partitioning, a technique that influences recommendations with ratings spatially close to querying users in a manner that maximizes system scalability while not sacrificing recommendation quality. LARS* exploits item locations using travel penalty, a technique that favors recommendation candidates closer in travel distance to querying users in a way that avoids exhaustive access to all spatial items. LARS* can apply these techniques separately, or together, depending on the type of location-based rating available. Experimental evidence using large-scale real-world data from both the Foursquare location-based social network and the MovieLens movie recommendation system reveals that LARS* is efficient, scalable, and capable of producing recommendations twice as accurate compared to existing recommendation approaches.
EXISTING SYSTEM:
Recommender systems make use of community opinions to help users identify useful items from a considerably large search space. The technique used by many of these systems is collaborative filtering (CF), which analyzes past community opinions to find correlations of similar users and items to suggest k personalized items (e.g., movies) to a querying user u. Community opinions are expressed through explicit ratings represented by the triple (user, rating, item) that represents a user providing a numeric rating for an item. Myriad applications can produce location-based ratings that embed user and/or item locations. Existing recommendation techniques assume ratings are represented by the (user, rating, item) triple.
DISADVANTAGES OF EXISTING SYSTEM:
·        The existing systems are ill-equipped to produce location aware recommendations.
·        The existing system provides more expensive operations to maintain the user partitioning structure.
·        The existing system does not provide spatial ratings.
PROPOSED SYSTEM:
We have proposed LARS*, a location-aware recommender system that uses location-based ratings to produce recommendations. LARS*, supports a taxonomy of three novel classes of location-based ratings, namely, spatial ratings for non-spatial items, non-spatial ratings for spatial items, and spatial ratings for spatial items. LARS* exploits user rating locations through user partitioning, a technique that influences recommendations with ratings spatially close to querying users in a manner that maximizes system scalability while not sacrificing recommendation quality. LARS* exploits item locations using travel penalty, a technique that favors recommendation candidates closer in travel distance to querying users in a way that avoids exhaustive access to all spatial items. LARS* can apply these techniques separately, or together, depending on the type of location-based rating available. Within LARS*, we propose:
(a) A user partitioning technique that exploits user locations in a way that maximizes system scalability while not sacrificing recommendation locality
(b) A travel penalty technique that exploits item locations and avoids exhaustively processing all spatial recommendation candidates.
ADVANTAGES OF PROPOSED SYSTEM:
·        LARS*, supports a taxonomy of three novel classes of location-based ratings, namely, spatial ratings for non-spatial items, non-spatial ratings for spatial items, and spatial ratings for spatial items.
·        LARS* achieves higher locality gain using a better user partitioning data structure and algorithm.
·        LARS* exhibits a more flexible tradeoff between locality and scalability.
·        LARS* provides a more efficient way to maintain the user partitioning structure
MODULE:
1.     Spatial ratings for non-spatial items
2.     Non-spatial ratings for spatial items
3.     Spatial ratings for spatial items
MODULES DESCRIPTION:
Spatial ratings for non-spatial items:
This section describes how LARS* produces recommendations using spatial ratings for non-spatial items represented by the tuple (user, ulocation, rating, item). The idea is to exploit preference locality, i.e., the observation that user opin-ions are spatially unique. We identify three requirements for producing recommendations using spatial ratings for non-spatial items: (1) Locality: recommendations should be influenced by those ratings with user locations spatially close to the querying user location (i.e., in a spatial neighborhood); (2) Scalability: the recommendation procedure and data structure should scale up to large number of users; (3) Influence: system users should have the ability to control the size of the spatial neighborhood (e.g., city block, zip code, or county) that influences their recommendations.
Non-spatial ratings for spatial items:
This section describes how LARS* produces recommendations using non-spatial ratings for spatial items represented by the tuple (user, rating, item, ilocation). The idea is to exploit travel locality, i.e., the observation that users limit their choice of spatial venues based on travel distance. Traditional (non-spatial) recommendation techniques may produce recommendations with burdensome travel distances (e.g., hundreds of miles away). LARS* produces recommendations within reasonable travel distances by using travel penalty, a technique that penalizes the recommendation rank of items the further in travel distance they are from a querying user. Travel penalty may incur expensive computational overhead by calculating travel distance to each item. Thus, LARS* employs an efficient query processing technique capable of early termination to produce the recommendations with-out calculating the travel distance to all items.
Spatial ratings for spatial items:
This section describes how LARS* produces recommendations using spatial ratings for spatial items represented by the tuple (user, ulocation, rating, item, ilocation). A salient feature of LARS* is that both the user partitioning and travel
penalty techniques can be used together with very little change to produce recommendations using spatial user ratings for spatial items. The data structures and maintenance techniques remain exactly the same as discussed in Sections 4and5; only the query processing frame-work requires a slight modification. Query processing uses Algorithm2 to produce recommendations. However, the only difference is that the item-based collaborative filtering prediction score P(u,i) used in the recommendation score calculation (Line16in Algorithm 2) is generated using the (localized) collaborative filtering model from the partial pyramid cell that contains the querying user, instead of the system-wide collaborative filtering model as was used.
SYSTEM REQUIREMENTS:
HARDWARE REQUIREMENTS:
Ø System                          :         Pentium IV 2.4 GHz.
Ø Hard Disk                      :         40 GB.
Ø Floppy Drive                 :         1.44 Mb.
Ø Monitor                         :         15 VGA Colour.
Ø Mouse                            :         Logitech.
Ø Ram                               :         512 Mb.
SOFTWARE REQUIREMENTS:
Ø Operating system           :         Windows XP/7.
Ø Coding Language         :         C#.net
Ø Tool                                  :          Visual Studio 2010
Ø Database                        :         SQL SERVER 2008

No comments:

Post a Comment