Rundensteiner, Elke A.
Rankings are commonly used to evaluate everyday decisions. However, constructing rankings can be difficult because building a ranking requires extensive knowledge about the observed dataset. To overcome this problem, we present RANKIT, an online ranking application that provides solutions for analyzing and exploring rankings. The system uses machine learning to automatically construct rankings based on partial input from users. The system was evaluated with a comprehensive two part user study consisting of online surveys and in-person interviews. Results reveal that RANKIT succeeds at both informing users and sparking their interest about personalized rankings. Future improvements have also been identified to make RANKIT even better.
Worcester Polytechnic Institute
Interactive Qualifying Project
All authors have granted to WPI a nonexclusive royalty-free license to distribute copies of the work, subject to other agreements. Copyright is held by the author or authors, with all rights reserved, unless otherwise noted.