User talk:Ahsh

Welcome to Wiki Course Notes! We hope you will contribute much and well. You will probably want to read the help pages. Again, welcome and have fun! WikiSysop 20:18, 25 July 2009 (UTC)

Problem Statement:

Webshoppers such as Amazon,Netflix,and Apple learn a suggestion function based on the the current user's and the others rating in order to offer personalized recommendations to this user. This user might be known (have been seen in the traing stage) or unknown for the system. To this end, collaborative filtering has provided a promising approach in which the rating patterns (of the products) by the current user and the others are used to estimate rates ( or ranking) for unrated items. The task is more challenging once the user is unknown for the system. Ranking is different from rating in which the set of recommendation is obtaind directly, rather than first finding the rates and then sort them accordingly. For collaborativ eratings, Maximum Marging Matrix Factorization (MMMF) had a promising result for estimating the unknown rate. This paper extends the use of MMMF for collaborative ranking.

Objectives:

1- diresct optimization of ranking scores, 2- being adaptable to different scores, 3- no need to extract features besides the actual ranking, 4- being scalable and parralizable with large number of items and users