Table of Contents

1. Data Understanding and Project Proposal
2. Models Implementation and Evaluation
   1. Movie Profitability Regression
      1. Additional Features Engineering
      2. Regressors
      3. Experiments
      4. Testcases
   2. Movie Profitability Classification
      1. Additional Features Engineering
      2. Handling Unbalanced Data
      3. Classifiers
      4. Experiments
      5. Testcases
3. Conclusion

1. Data Understanding and Project Proposal (to top)

https://jovian.ai/lilanpei/data-understanding-and-project-proposal

Although We have already performed extensive analysis of our data and haven't done a lot with respect to determining factors that make a movie a success. We all know how the economic return a producer can get from a film is important. So what factors influence whether a movie will be able to recoup its investments? To answer this question, we shall attempt to build :

• A Regressor that can, to a certain degree, predict the economic return of a particular movie starting from its characteristics.
• A Classifier that identifies if a movie will be a hit or will make the producers lose money.

2. Models Implementation and Evaluation (to top)

Regarding the suggestion of model implementation, we dropped the features ('popularity', 'vote_average', 'vote_count') that are still not available before a movie is released. Then, we redid the features engineering, data preprocessing and retrained all the models for regression and classification tasks.