IS607 Project-3 Summary
Is Best Film Editing is the best predictor of Best Picture ?
Introduction
Data Acquisition & Transformation
Data Tidying & Reshaping for Analysis
Analysis 1 - Points based approach
- If Best Picture movie was also nominated on other categories then those categories get one point
- If the BEST Picture movie also won in other categories then those categories get two points. Categories with most points are the best indicators or predictors of Best picture movie.
Conclusion: Directing Category can be used a good indicator to PREDICT the BEST PICTURE.
Analysis 2 - Using 'Bayes' formula
- Probability of getting Best Movie, given that the movie is nominated in a paricular award category ?
Analysis 2 - Using 'Bayes' formula
- Probability of getting Best Movie, given that the movie wins in a paricular award category ?
Conclusion: Directing Category can be used a good indicator to PREDICT the BEST PRICTURE.
Analysis 3 - Using 'Decision Tree'
Conclusion: Directing Category can be used a good indicator to PREDICT the BEST PRICTURE.
Analysis 4 - Applying Modeling Techniques
Different modeling techniques ( CART, Bagged Trees, Random Forests ) have been applied in the above analysis.
Conclusion: No definitive conclusion may be drawn here.
Visualizations
Visualizations
Visualizations
Visualizations
Conclusion
- Majority of our analysis concludes that the Best Film Editing is NOT the best predictor of Best Picture
- Also, most of the approaches suggests that the Directing Category can be used a good indicator to predict the BEST PICTURE.
