الفهرس 
Recommendation systems in location based social network
AbstractOnly 14 pages are availabe for public view
 |  | from | 89 |  |  |
| | | from | 89 | | |
Abstract
Recommendation systems in location based social networks have been significantly improved recently, because of the advancements in the social networks{u2019} services, positioning techniques and the increase in the number of users. Previously recommendation systems in location aimed at recommending most fitting venues based on their distance to the target users{u2019} location. As a result for the upgrades in positioning techniques (i.e. GPS, mobile tracking services), the data sizes produced became tremendousalso these enhancements improved spatial data from single points to trajectory data to keep track of the target users{u2019} movements information, for that recommendation systems has a challenging problem to enhance effectivenessof recommendation model performance. Taking into consideration the upgrades in the social networks{u2019} services with the numbers of the web users which led to new types of social information (i.e. friendship, rates) instead of spatial data (i.e. latitude, longitude) which was a key pillar in satisfying the target users{u2019} requirements. For that Contributions in trajectory search are lately enhanced because of the gigantic road networks and LBSN applications which give expansive scale information of locations and related properties. In general, trajectory search depends on two fundamental segments: 1) similarity search; which is concerned with looking at trajectories and retrieving the ones that are similar to each other based on some distance function 2) trajectory mining; it concentrates on the best way to preprocess trajectories before searching within trajectories through performing a number of processes like removing noise, grouping similar trajectories, distinguishing trajectories` features. Some research works in trajectory search area took care of the possibility of activity trajectory which means to endorse best related trajectory(s) that meet users{u2019} needs through using trajectory metadata like (nearest direction, activities needed, rating, preferences, etc.)