Given the "RF" in the filename, a Random Forest classifier is appropriate for predicting the likelihood of a user saving a product [2, 3].
Develop a user-item matrix (e.g., user X saved product Y) to identify preferences [3]. 2. Modeling (Random Forest Approach) Wanelo_RF.7z
The model generates a ranked list of product IDs predicted to have the highest probability of being saved by that user. 4. Evaluation Given the "RF" in the filename, a Random
Create vectors based on description, category, and seller [1, 3]. Modeling (Random Forest Approach) The model generates a
What is in the (e.g., user-save data, product metadata)?
Use Precision@K and Recall@K to evaluate how many of the top-K recommended products were actually relevant to the user [2, 3]. To help you develop this further, could you tell me:
Create vectors for users based on categories saved, price points, and interaction frequency.