i have a data set with personal properties (age, education etc.), and each entry in the dataset is associated with a list of websites and daily visit.

my goal is to predict for a given person the list of websites he will visit, and to predict, for each website, how many daily visit it will have from that person.

the number of websites each person visits is very limited, usually between 3 and 20 (same websites for the whole dataset). the number of different persons is limited (a few hundreds) but the number of entries per person is large (a few thousands, each representing a day).

currently my thoughts are to have two stage method - first stage is for each possible website a person might visit (like mentioned before, usually 3-20), classify that website as "visited" or "not visited" (using binary classifiers for each website).

stage two is for each website that was classified as "visited", use a linear regression to predict the number of daily visits.

I am very new to the field of ML so dont spare your words! your comments and tips are very appreciated :) Thank you!


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.