UCDCS Best Short Paper award 13th ACM Conference on Recommender Systems (RecSys)
UCD Computer Science researchers received the Best Short Paper award at the 13th ACM Conference on Recommender Systems (RecSys), for a paper entitled "Pace My Race: Recommendations for Marathon Running” by Jakim Berndsen, Barry Smyth, and Aonghus Lawlor. RecSys is the leading conference on recommender systems and this year was located at Copenhagen, Denmark. The conference attracted more than 900 participants with more than 70% from industry,
The paper describes work using ideas from machine learning and recommender systems to help marathon runners to race more effectively, by recognising when a runner might be struggling during a marathon, and by recommending corrective action (pacing changes) to improve their chances of completing the race without hitting the wall. The paper considers when and how such recommendations might be communicated to a runner during a race, and describes the results of an evaluation, based on historical marathon data, to show how such recommendations can help runners to improve their finish-times.
Recommender systems have been used for some time to learn about the preferences of users in domains such as music, books, and videos (TV and movies), and have proven to be a key feature of many online e-commerce stores. The work presented by Jakim Berndsen, a 3rd year PhD student in the School of Computer Science, is a novel departure from such traditional application domains, and demonstrates the potential for machine learning and recommender systems to support users in many aspects of their everyday lives. The paper is part of an ongoing research project that focuses on the role of recommender systems in supporting exercise with a particular emphasis on running.
---
Link to the paper: (opens in a new window)https://dl.acm.org/citation.cfm?id=3346991