الفهرس 
Enhancing multi-factor friend recommendation in location-based social networks
ContentsOnly 14 pages are availabe for public view
 |  | from | 67 |  |  |
| | | from | 67 | | |
Abstract
Recently, location features have become available by the most popular on- line social networks such as Facebook, Twitter, and Foursquare.These networks are called location-based social networks (LBSN), which allow users to share their loca- tions and location-related content. One of the services that LBSNs present is friend recommendation. This service recommends new friends to users based on their posts, media, opinions, locations or social ties. Several studies have been conducted in the area of friend recommendation by LBSNs. They built their models/frameworks based on a combination of two or three features: social, spatial and textual. Nevertheless, there are some limitations or issues in previous studies that may undermine recommendation accuracy. In some previous studies, the users{u2019} topics and/or opinions were not considered as a textual feature. It re{uFB02}ects users{u2019} interests correctly and avoids con{uFB02}icts that can happen if two users have di{uFB00}erent opinions on the same topic. In spite of that, one of them will be recommended to the other. Another limitation is not considering social relations while calculating the social feature. Social relations on Twitter that based on three directed links followers, friends and mutual friends treated as undirected links as on Facebook.That leads to overlap users{u2019} interests which re{uFB02}ects on recommendation results. For spatial feature, the authors claimed that some places should be {uFB01}ltered or be ignored to enhance the recommendation process. from within these {uFB01}ltered places, there are some popular places and/or faraway places where users can meet and make friends, therefore, increasing the recommendation accuracy