ML: Recommend video

[jo-elimu]

Description

Code a machine learning model for recommending a video based on a student's previously watched videos.

Why?

We want to guide the student towards educational content that is more likely to keep the student engaged, so a video recommender system would be helpful for this.

Tasks

1. Prepare data:

• Import video datasets from webapp ml-video-recommender#1
• Clean the imported data ml-video-recommender#3

2. Train model:

• Split data into training set and test set ml-video-recommender#4
• Perform feature engineering ml-video-recommender#2

3. Make prediction:

• Make predictions on unseen data ml-video-recommender#6
• Compare the accuracy score across algorithms, features, train/test splits ml-video-recommender#7

4. Deploy in production:

• Export model to format compatible with Android apps ml-video-recommender#5
• Add video recommender model filamu#2