School of Technology

Permanent URI for this collectionhttp://192.168.8.146:4000/handle/123456789/18

Browse

Filter results by typing the first few letters
Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    An Extra Trees Regressor To Predict Content Popularity On The Netflix Platform In Kenya
    (KCA University, 2024) Karuku, Esther G
    This study aimed to develop an Extra Trees Regressor model to predict content popularity on the Netflix platform in Kenya. The data used in this study was collected from June 28, 2021 to March 24, 2024. The experimentation yielded compelling results, with the Extra Trees Regressor demonstrating superior performance compared to both Linear Regression and Ridge Regression. Extra Trees Regressor showed consistently lower error rates across all metrics (MAE, MSE, RMSE, and MAPE) except RMSLE suggesting a high degree of accuracy in predicting content popularity for Kenyan audiences. A high R² value (0.9140) indicates the Extra Trees Regressor model effectively captured the relationship between content attributes and content popularity. The study revealed the two most important predictors of content popularity are the show title and the director contributing to the ongoing investigation of the content popularity problem globally.