I have a dataset that is basically like the MovieLen dataset. It has 3 columns: userGUID, card, action. User GUID is userId, card is contentId, action is ratings of contents by users. It looks like as below:
I have 16536 users and 16734 cards And only little portion of them is rated (action).
What I want to do is to is data augmentation where I get userGUID and card rows which are not rated at all.
So I would ultimately like to have each user's all card that is not rated (action-ed').
Each user will have augmented data excluding already rated ones. So data I expect would look like below:
This table is only for not rated augmented ones.
I wish someone can provide python code to do this.
Thank you in advance.
与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…