We wonder if it’s possible to predict movie’s ratings regarding several features.

What are the features?

• gender_speaker_pct : proportion of male (for the speakers)
• gender_movie_pct : proportion of male (for all participants in the creation of the film)
• nb_quoteID_norm : number of quotes occurences (standardized)
• memorability:
  • gender_movie_pct : proportion of male (for all participants in the creation of the film)
  • days_mean_norm_before: average number of days before the release date (standardized)
  • days_max_norm_before : number of days from the first day we talk about it until the release (standardized)
  • nb_quote_month_norm_before: the number of situations one month before the exit (standardized)
  • days_mean_norm: average number of days after the release date (where it is discussed) (standardized)
  • days_max_norm : number of days until the last day that it is discussed (standardized)
  • nb_quote_month_norm : the number of situations one month after discharge (standardized)

What is the response?

• The rating of a film

As we can see on the graph and the value of the r2=-0.10, it does not seem possible to predict the score with the proposed features