« Peter Lindbergs reboot7-föreläsning: The Three Pillars of Software Architecture | Main | Kort om George Lakoff & Mark Turner: "Metaphors we live by" och lite annat relaterat till Lakoff »

juni 15, 2005

Nature News: Modellers measure 'word of mouth' for films

Nature News: Modellers measure 'word of mouth' for films

It's official, says one group of researchers: Blade II is a bad film. Their study turns patterns of attendance into a single number that claims to grade a film's quality.

The number attempts to gauge of how good the 'word of mouth' was around a given film, based on the behaviour of the harshest critics of all, the paying public.

César Hidalgo, now a graduate student in physics at the University of Notre Dame, Indiana, and his colleagues, decided to study the 'word of mouth' effect in the film world simply because reviews often have a huge impact on audience numbers and there are copious data on ticket sales.


Se vidare
Cesar A. Hidalgo, Alejandra Castro, Carlos Rodriguez-Sickert: The effect of social interactions in the primary life cycle of motion pictures (arXiv, PDF)

Abstract:

We model the consumption life cycle of theater attendance for single movies by taking into account the size of the targeted group and the effect of social interactions. We contrast the model with empirical data from the film industry obtaining good agreement with the diverse types of behaviors empirically found in the data. The model grants a quantitative measure of the valorization of this cultural good based on the relative values of the coupling between agents who have watched the movie and the ones who have not.

Posted by hakank at juni 15, 2005 06:22 EM Posted to Social Network Analysis/Complex Networks

Comments

Forskningsvärlden skulle nog vara lite fattigare utan alla dessa teoretiska fysiker och ekonomer som gör roliga modeller. Jag var också tvungen att blogga om den här :)

Egentligen tror jag nog att modellen är lite för enkel. De verkar ha kollat på en mer intressant variant av informationsspridning i grupper som fanns i deras referenslista (Information Flow in Social Groups), men inte använt den.

Posted by: Malin Sandström at juni 17, 2005 03:21 EM