The department of Mathematics and Statistics at the University of Limerick invites you to a seminar by Dr Aonghus Lawlor (University College Dublin):
Title: The Effectiveness of Recommender Systems.
Abstract: The Effectiveness of Recommender Systems Recommender Systems are ubiquitous in many areas of e-commerce and have become essential for many online activities such as movies, books, consumer devices, fashion, travel, etc. In this talk I will summarise some of our recent work in Recommender Systems, with a particular focus on building explainable recommendations. Explanations accompany the recommendation and provide additional information to the user about how the recommendation was arrived at. Building explainable recommender systems is an increasingly important challenge, not only because they are known to increase the likelihood of acceptance by the user, but also because there is upcoming legislation which will give users rights to accurate and fair explanations. My work has focused on building explainable recommender systems, but little is known about the effect of recommender systems on the choices that users make about products, or content. This problem is increasingly important in the diffusion of information over social networks which is often driven or controlled in part by recommender systems. All sort of difficulties are encountered in studying how recommender systems affect the diffusion of information over the network- the context, network effects, community structure, timing and dynamics, saturation points , are all important features. If we can exploit this information we can build better recommender systems, and more importantly, new and better ways of explaining to the user why the recommendation was made. These explanations go beyond the usual attempts to explain the details of the model, and explain something more global about the adoption patterns of the recommendations.
This seminar will take place on Friday 23rd March, at 4p.m, in Room A2-002.
If you have any questions regarding this seminar, please direct them to Iain Moyles (061 233726, firstname.lastname@example.org).
A full list of upcoming seminars can be found at http://www.ulsites.ul.ie/macsi/node/48011
Supported by Science Foundation Ireland funding, MACSI - the Mathematics Applications Consortium for Science and Industry (www.macsi.ul.ie), centred at the University of Limerick, is dedicated to the mathematical modelling and solution of problems which arise in science, engineering and industry in Ireland.