
Ultimate access to all questions.
Your company operates a global video sharing platform with millions of daily active users who both view and upload videos. The platform's success heavily relies on user engagement, which is significantly influenced by the visibility of popular videos. You are tasked with developing a machine learning model to predict which newly uploaded videos will become the most popular within the first 30 days after upload. The goal is to feature these videos prominently on the website to maximize user engagement and ad revenue. Given the platform's scale, the model must efficiently process thousands of videos daily, considering factors like video content, uploader history, and initial engagement metrics. Which of the following metrics should you use to evaluate the success of your model, ensuring it aligns with the business objective of maximizing watch time and user engagement? (Choose one correct option)
A
The model accurately predicts 97.5% of the most popular clickbait videos based on the number of clicks they receive.
B
The model identifies videos as popular if the uploader has garnered more than 10,000 likes in the past, regardless of the current video's performance.
C
The Pearson correlation coefficient between the log-transformed views after 7 days and 30 days post-upload is 0, indicating no linear relationship.
D
The model predicts 95% of the most popular videos, measured by watch time, within the first 30 days after upload.