On Using Temporal Networks to Analyze User Preferences Dynamics

• Type: Article in International Conference Proceedings Book
• Date: 2016

Authors

Pereira, FSF (Author) Other; de Amo, S (Author) Other; João Gama (Author) FEP

Conference proceedings International

Title: DISCOVERY SCIENCE, (DS 2016)

Pages: 408-423

19th International Conference on Discovery Science (DS) / 27th International Conference on Algorithmic Learning Theory (ALT)

Bari, ITALY, OCT 19-21, 2016

Indexing

ISI Web of Knowledge - 3 Citations

Scopus - 4 Citations

Other information

Authenticus ID: P-00M-3X6

DOI: 10.1007/978-3-319-46307-0_26

Abstract (EN): User preferences are fairly dynamic, since users tend to exploit a wide range of information and modify their tastes accordingly over time. Existing models and formulations are too constrained to capture the complexity of this underlying phenomenon. In this paper, we investigate the interplay between user preferences and social networks over time. We propose to analyze user preferences dynamics with his/her social network modeled as a temporal network. First, we define a temporal preference model for reasoning with preferences. Then, we use evolving centralities from temporal networks to link with preferences dynamics. Our results indicate that modeling Twitter as a temporal network is more appropriated for analyzing user preferences dynamics than using just snapshots of static network.

Language: English

Type (Professor's evaluation): Scientific

No. of pages: 16