With the movie dataset, we want to predict the success of movies based on the given attributes. This will allow producers and investors to estimate how well films will do in the market using attributes such as runtime, release date, and actors. Success will be defined by the return on investment. We will also apply clustering techniques to group the movies to find out what are the important attributes for the movie to be successful.
We will start by preprocessing the data. Then, we will use dimension reduction and feature selection methods to break the attributes down further. Here, we will transform the dataset that includes multi-valued attributes in a useful way using a new feature engineering technique. We will then divide the dataset into training and testing datasets. Finally, we will apply various classification techniques to compare and select the best model to predict the return on investments of movies. After we establish the baseline, we can find ways to improve the existing models.