Abstract
Because the IP film box office is affected by a lot of factors, IP influencing factors have become an important basis for practitioners to make decisions. In the actual situation, the relationship between box office and various influencing factors is often uncertain and non-linear. The generalized additive model breaks through the limitations of traditional linear model assumptions and improves the accuracy of model fitting. Using the generalized additive Model to analyze the Influencing Factors of IP film Box Office, could show the degree of influence of various factors more accurately and intuitively, including IP type, IP influence, IP score and other IP influence factors, besides, film type, director influence, scriptwrite influence, starring influence and other non-IP influence factors. This paper selects the 237 film data released in 2013-2016 as a sample. The analysis results show that the IP score and IP influence have a nonlinear relationship with the box office, and play a key role in the change of the box office. The relationship between IP type and box office is not significant. Among non-IP factors, film type and director influence have a greater impact on box office.