LARS: An Efficient and Scalable Location-Aware Recommender System

LARS: An Efficient and Scalable Location-Aware Recommender System
Feb 2013

Technology Used: Java/J2EE

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, nonspatial 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

Existing recommendation techniques assume ratings are represented by the (user, rating, item) triple, thus are ill-equipped to produce location-aware recommendations.

Location-based services
Current location-based services employ two main methods to provide interesting destinations to users. (1) KNN techniques and variants (e.g., aggregate KNN) simply retrieve the k objects nearest to a user and are completely removed from any notion of user personalization. (2) Preference methods such as skylines and location-based top-k methods require users to express explicit preference constraints.

Traditional recommenders
A wide array of techniques are capable of producing recommendations using non-spatial ratings for non-spatial items represented as the triple (user, rating, item). The closest these approaches come to considering location is by incorporating contextual attributes into statistical recommendation models. However, these are not personalized to each user; rather, this list is built using aggregate rental data for a particular city.

Location-aware recommenders
The CityVoyager system mines a user’s personal GPS trajectory data to determine her preferred shopping sites, and provides recommendation based on where the system predicts the user is likely to go in the future. The spatial activity recommendation system mines GPS trajectory data with embedded user-provided tags in order to detect interesting activities located in a city. It uses this data to answer two query types: (a) given an activity type, return where in the city this activity is happening, and (b) given an explicit spatial region, provide the activities available in this region. Geo-measured friend-based collaborative filtering produces recommendations by using only ratings that are from a querying user’s social-network friends that live in the same city. This technique only addresses user location embedded in ratings.

Disadvantages
 Does not personalize answers to the querying user
 No traditional approach has studied explicit location-based ratings.

PROPOSED SYSTEM

 LARS, a novel location-aware recommender system built specifically to produce high-quality location-based recommendations in an efficient manner.
 LARS produces recommendations using a taxonomy of three types of location-based ratings within a single framework
 Spatial ratings for non-spatial items, represented as a four-tuple (user, ulocation, rating, item), where ulocation represents a user location, for example, a user located at home rating a book
 Non-spatial ratings for spatial items, represented as a four-tuple (user, rating, item, ilocation), where ilocation represents an item location, for example, a user with unknown location rating a restaurant
 Spatial ratings for spatial items, represented as a five-tuple (user, ulocation, rating, item, ilocation), for example, a user at his/her office rating a restaurant visited for lunch.

Advantages
 Helps users discover new and interesting items
 LARS, produces personalized recommendations influenced by location-based ratings and a querying user location.

 

Related posts

Leave a Reply