KNOWLEDGE MANAGEMENT

AND DISCOVERY LAB

In this course we elaborate on the role of recommenders as a primary means of improving a user's or customer's experience, while increasing company revenue. The course covers learning methods for the recommender core, approaches for the design and evaluation of recommenders, and specific application areas of recommenders.

This course is a followup of the bachelor course CRM/RecSys, which contains an introduction to Customer Relationship Management and to Recommendation Engines, including two core methods for model learning (collaborative filtering and content-based modeling) and basics on evaluating recommenders. In this course, we consider advanced learning methods and advanced instruments for guaranteeing the goodness of the recommender core.

Literature:

    F. Ricci, L. Rokach, B. Shapira (eds). Recommender Systems Handbook. Springer 2011, esp:

Learning methods and applications

• Ch4: A Comprehensive Survey of Neighborhood-based Recommendation Methods
• Ch5: Advances in Collaborative Filtering
• Ch19: Social Tagging Recommender Systems
• Ch22: Aggregation of Preferences in Recommender Systems

Recommender design and evaluation

• Ch11: Matching Recommendation Technologies and Domains
• Ch14: Creating More Credible and Persuasive Recommender Systems:
 The Influence of Source Characteristics on Recommender
 System Evaluations
• Ch15: Designing and Evaluating Explanations for
RecommenderSystems

2) Scientific articles, mainly from the ACM conferences:

• International Conf. on Recommender Systems (RecSys)
• International Conf. on Information & Knowledge Management (CIKM)

Background in data mining is of advantage. This course is also appropriate for students who have heard the CRM/RecSys bachelor course.