In Facebook, the set of pages liked by some users represents an important knowledge about their real life tastes. However, the process of classification, which is already hard when dealing with dozens of classes and genres, is made even more difficult by the very coarse information of Facebook pages. Our work originates from a large dataset of pages liked by users of a Facebook app. To overcome the limitations of multilabel automatic classification of free-form user-generated pages, we acquire data also from IMDb, a large public database about movies. We use it to associate with high accuracy a given cinema-related page on Facebook to the corresponding record on IMDb, which includes plenty of metadata in addition to genres. To this aim, we compare different approaches. The obtained results demonstrate that the highest accuracy is obtained by the combined use of different methods and metrics.

Guess the movie - Linking facebook pages to IMDb movies

Guidi B.
;
2017-01-01

Abstract

In Facebook, the set of pages liked by some users represents an important knowledge about their real life tastes. However, the process of classification, which is already hard when dealing with dozens of classes and genres, is made even more difficult by the very coarse information of Facebook pages. Our work originates from a large dataset of pages liked by users of a Facebook app. To overcome the limitations of multilabel automatic classification of free-form user-generated pages, we acquire data also from IMDb, a large public database about movies. We use it to associate with high accuracy a given cinema-related page on Facebook to the corresponding record on IMDb, which includes plenty of metadata in addition to genres. To this aim, we compare different approaches. The obtained results demonstrate that the highest accuracy is obtained by the combined use of different methods and metrics.
2017
978-3-319-71969-6
978-3-319-71970-2
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11568/1056192
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 2
  • ???jsp.display-item.citation.isi??? ND
social impact