{"id":2022,"date":"2022-03-25T17:56:44","date_gmt":"2022-03-25T17:56:44","guid":{"rendered":"https:\/\/salarydistribution.com\/machine-learning\/2022\/03\/25\/predicting-the-2022-oscar-winners-with-machine-learning\/"},"modified":"2022-03-25T17:56:44","modified_gmt":"2022-03-25T17:56:44","slug":"predicting-the-2022-oscar-winners-with-machine-learning","status":"publish","type":"post","link":"https:\/\/salarydistribution.com\/machine-learning\/2022\/03\/25\/predicting-the-2022-oscar-winners-with-machine-learning\/","title":{"rendered":"Predicting the 2022 Oscar Winners with Machine Learning"},"content":{"rendered":"
\n

The 94th Academy Awards<\/strong><\/a> ceremony is taking place this weekend as the event finally returns to the Dolby Theatre in Los Angeles. As we have done since 2018<\/strong><\/a>, we continued collecting the relevant data points about this year\u2019s nominees and updated our Machine Learning models with some tweaks as described later in this post. As a reminder, the dataset<\/strong><\/a> is openly available for anyone to clone to their BigML account (get a free one here<\/strong><\/a>) and come up with your own predictions if you haven\u2019t done so already.<\/p>\n

\"\"<\/figure>\n

The Models<\/strong><\/h2>\n

We\u2019ve used multiple modeling approaches including\u00a0<\/span>OptiML<\/a><\/strong>, the\u00a0popular AutoML feature<\/strong> on BigML that automatically finds the best performing supervised models<\/strong> and individual Deepnets<\/strong><\/a>, which powered most of our past Oscars predictions. We also built Fusions,<\/a> <\/strong>by combining up to the top 10 best performing models from OptiML results per category, e.g., Best Director.<\/span><\/p>\n

\"\"
OptiML Results<\/figcaption><\/figure>\n

Once our candidate models were created, we made Batch Predictions<\/strong> against the movies produced in 2021 that we had set aside in a separate dataset. Although both OptiML Fusions<\/strong> and Deepnets<\/strong> agreed on most categories, we observed some differences in rankings and relative scores. In the end, we chose to average out the two approaches to come up with our final predictions and rankings.<\/span><\/p>\n

\"\"
Fusion Model for Best Costume Design<\/figcaption><\/figure>\n

As usual, given BigML\u2019s emphasis on white-box models with explainability<\/strong><\/a> we can dig deeper into models and predictions for added introspection as needed. For example, the partial dependence plot<\/strong> of the above Fusion<\/strong> for Best Costume Design shows how International Cinephile Society<\/em><\/strong> and Costume Designer Guild<\/em><\/strong> awards (top two predictors for the category) interplay in determining whether a given movie will win the Oscar for Best Costume Design.<\/span><\/p>\n

The 2022 Predictions<\/strong><\/h2>\n

Drum rolls please! Below, we predict the most likely winner along with other nominees in the same category sorted by decreasing scores. These scores are a blend of the OptiML<\/strong> top performing model Fusions<\/strong> and our old faithful Deepnets<\/strong> we\u2019ve been using over the years and they aren\u2019t supposed to add up to 100. In other words, the models are telling us how a movie\/artist with a given set of characteristics will do in a given award category based on 20+ years of historical data on that award independent of the other nominees for the same award this year.<\/span><\/p>\n

Intuitively, Fusion<\/strong> models are likely to balance different techniques to make predictions more robust and in-line with past years as opposed to the individual Deepnets<\/strong> being more likely to pick surprise winners at the expense of missing some favorites. Combining both may be interpreted as a middle-of-the-road trade off that we used to come up with our picks this year.<\/p>\n

One more thing, in addition to revealing our model picks, we take note of the leading choices from the BigML Oscars Challenge<\/strong><\/a> participants so we can compare and contrast against a mini \u201cwisdom of the crowds<\/strong><\/a>\u201d approach.<\/p>\n

\"\"
2022 Best Picture Pick: THE POWER OF THE DOG<\/figcaption><\/figure>\n