•A
suitable technology option would be Collaborative filtering which is a common
web technique used in generating personalized recommendations.
•It is
used by Amazon, Netflix, iTunes and last FM. With iTunes CF works in that all
in needs is a database of who has bought what so that it can calculate “people
who are like you” so that anytime anyone “who is like you” buys something you
have not bought you get a personal recommendation of the song or album.
• CF
does not need built in subject knowledge and this makes it easier and less
complicated such as Pandora which needs music experts to individualise each
song and code a genome to each song.
No comments:
Post a Comment