Seminar: A Recommender System for Breast Cancer Patients

Momeneh Taban
M.Sc. Candidate
Supervisor: Dr. Jeffrey Parsons

A Recommender System for Breast Cancer Patients

Department of Computer Science
Tuesday, May 13, 2014, 14:00 p.m., Room EN 2022


An ongoing challenge in the information age is finding information relevant to a particular need. One area in which this is particularly problematic is the medical domain, where patients suffering from certain conditions seek advice on managing their health. Personalized recommendations can be useful in this context. A recommender system can assist users to locate relevant information and choose the best option that matches their needs.

This thesis developed a Breast Cancer Recommender System (BCRS) which recommends health related articles appropriate for patients confronting breast cancer. BCRS applies a hybrid algorithm which combines collaborative filtering and content based approach to generate recommendations. Article recommendations can be categorized in 4 main groups of life style, emotional concerns, risk factors and treatment. To analyse the article recommendations' usefulness an evaluation was conducted using female medical students of Memorial University.