Troubles in Deciding?  How Recommender Technology Can Help Consumers Choose

Speaker:  Pearl Pu – Preverenges, Switzerland
Topic(s):  Web, Mobile and Multimedia Technologies

Abstract

As online stores offer practically an infinite shelf space, recommender systems are playing an increasingly important role in helping users search and discover items that they truly want. But are they also helping them make better decisions?

In decision theory, a choice problem is defined as the task of choosing an option out of a set of alternatives. The larger the alternative set is, and the more features definine each option, the more difficult it is for users to choose.  I will give an overview  of recent  literature  in decision  theory  that explains  the discrepancies between normative models of how people should reason and empirical studies of how they in fact think and decide. This discrepancy points out that we have trouble deciding in ubiquitous situations such as finding something to buy in online stores. I will explain what regulates this difficulty such as decision accuracy vs. user effort.

I then describe a special type of recommender technology, critiquing based recommender systems, designed specifically  to  help  users  make  better  decisions  while  minimizing  their  cognitive  effort.  I will  describe methods  to  compute  critiques  as  well  as  how  to  present  them  to  users.  I illustrate  the  techniques  with examples from actual implementations and user studies.


About this Lecture

Number of Slides:  50
Duration:  40 minutes
Languages Available:  English
Last Updated: 

Request this Lecture

To request this particular lecture, please complete this online form.

Request a Tour

To request a tour with this speaker, please complete this online form.

All requests will be sent to ACM headquarters for review.