Taylor Ferebee is currently pursuing bachelors of science degrees in physics and mathematics at Roanoke College in Salem, Virginia. Having worked as an Aerospace Ambassador at NASA Langley Research Center, Taylor enjoys the many aspects of space exploration and technology development. Along with physics and mathematics, she studies the classics and has won numerous awards for excelling in Latin. In her leisure time, Taylor enjoys hiking, community service, running, and reviewing movies.
For many years, films have been one of the most valued ways of storytelling. Through previews and film posters, viewers judge whether or not to pay to see the film. With the advent of critic websites such as Rotten Tomatoes, viewers want to predict if the film is worth spending hard-earned money on. Today, with the demand for movies higher than ever, filmmakers need a way to understand what makes a movie profitable. I propose that by using machine learning techniques, an algorithm for predicting the profitability of a film can be found.
First, I will use machine learning techniques to understand what makes a movie profitable. I will use variables such as budget, director, and genre to find some infographical relationships between the variables and profitability by using the top 100 grossing films. Next, I will use the data to form a subset of values that can be placed together to understand what composes the most successful movies. I will also try to weight these values according to the strength of correlation. I will repeat this study with sequels to see if there is a significant difference when a film is a sequel. Finally, using new computations I will develop an algorithm that takes input of the most “important” values and predicts the profitability of a film.
Favorite Outer Totalistic r=1, k=2 2D Cellular Automaton