I have problem with construction recommended system. I have a DataFrame with columns of users , items (books) and the order in users read the book -

df_interact = pd.DataFrame({user_id : [1, 1, 2, 2, 2], item_id : [A, B, C, A, B], order : [1, 2, 1, 2, 3]})

, also I have some information about users (anonymized features)

df_users = pd.DataFrame({user_id : [1, 2], featurer_1 : [0.333, 0.213], featurer_1 : [0.8879, 0.123]})

The problem is that I don't know how users rated for definite item, there is information only user read the book or didn't read. I construct the feature for every book (for example how many users read the book, how many users read the book in first and so on) and tried to build the recommendation system by LightFM but the prediction is too bad, can anyone help me with that? What kind of mashine learning algorithm I have to use? I use python



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.