Social Networks mining for recommending users
The massive and global use of social networks has generated a source of information that is extremely large, constantly updated and which is mostly publicly accessible to all users. Due to this large volume of data, emerges the need of theories and tools that assist people in extracting useful information from the user-generated data, combined with the relationships among users and social interactions.
In this direction, the recommendation of users with specific features is vital for decision-making in different contexts. For example, knowing expert users or users that can be considered sources of information is useful to generate new connections for new users, while meeting influential users will identify nodes that propagate faster certain information.
The amount of information that can be obtained from social network users with respect to users in other domains, is greater at a lower cost. There are several aspects that make social networks interesting for building recommender systems. First, users are no longer mere consumers of information but become leading producers and broadcasters. Millions of people spend many hours on various social media sites (blogs, social networks, specialized forums, etc.), interacting and sharing information online the same way they do in the real world. This fact makes social networks a great source of knowledge. Second, publications and other user interactions often reflect the users' interests, that can be used to incorporate new preferences to the users' profiles without requiring direct user interaction with the system. Finally, useful information can be extracted from the established connections between users on social media to enrich the recommendation process. The overall objective of this project is to specify, adapt and apply techniques for identifying and recommending users with relevant features for other users in the network. These techniques can be applied to both to massive-use online social networks and to private social networks such as collaborative work environments, computer-supported learning, academic collaboration networks, e-commerce sites, among others.
Director: Marcelo G. Armentano
Responsible Group: Ariel Monteserin. Virginia Yannibelli
Collaborators: Luis Berdun. Eduardo Zamudio. Franco Berdun.