A Survey on One Class Collaborative Filtering

PhD Qualifying Examination


Title: "A Survey on One Class Collaborative Filtering"

by

Mr. Zhongqi LU


Abstract:

The real world recommender systems are usually challenged by the lack of 
negative training samples. Especially for the online advertisement 
recommendation systems, only the positive ``click'' or ``like'' samples are 
collected. This leads to the One Class problems in the recommender systems. A 
branch of researches solve the one class recommendation problems by 
Collaborative Filtering. We classify these researches as One Class 
Collaborative Filtering (OCCF).

This survey provides a summary of the state-of-the-art OCCF models from four 
perspectives: memory-based methods, model-based methods, transfer learning 
methods, and reinforcement learning methods. We would like to cover a broad 
spectrum of the existing methods for the reference of the future researches.


Date:			Tuesday, 21 June 2016

Time:                  	2:00pm - 4:00pm

Venue:                  Room 2611
                         Lifts 31/32

Committee Members:	Prof. Qiang Yang (Supervisor)
 			Prof. Huamin Qu (Chairperson)
 			Dr. Xiaojuan Ma
 			Prof. Nevin Zhang


**** ALL are Welcome ****