Webinar | Recommender Systems: Using Fuzzy Fingerprints for “Proper” Recommendations by Prof Joao Paulo Carvalho
Abstract: Most Recommender Systems rely exclusively on ratings and are known as Memory-based Collaborative Filtering systems. This is the currently dominant approach outside of academia due to the low implementation effort and service maintenance, when compared with more complex Model-based approaches.
Traditional Memory-based systems have as their main goal to predict ratings, using similarity metrics to determine similarities between the users’ (or items) rating patterns. In this talk, we propose item and user-based Fuzzy Collaborative Filtering approaches that do not necessarily rely on rating prediction, instead leveraging on Fuzzy Fingerprints to create a novel similarity based recommendation approach. Fuzzy Fingerprints provide a concise and compact representation of users allowing the reduction of the dimensionality usually associated with user-based collaborative filtering.
Biography: Prof João Paulo Carvalho has a PhD (2002), a MSc (1996) and an Electrical and Computer Engineer (1992) degree from Instituto Superior Técnico, University of Lisbon, Portugal, where he is currently a Tenured Associate Professor at the Department of Electrical Engineering and Computers. He has lectured courses on Computational Intelligence, Distributed Systems, Entrepreneurship and Technology Transfer, Computer Architectures and Digital Circuits since 1999. He is in the Board of Directors of INESC-ID, where he is a senior researcher and has coordinated 5 funded research projects and participated in more than a dozen national and European funded projects.
Prof João’s current main research interest involves developing and applying new Computational Intelligence techniques to natural language processing, text mining, social network analysis, social sciences and earth sciences. He has authored over 130 papers in international scientific Journals, books and peer-reviewed conferences. He is Area Editor of Fuzzy Sets and Systems and was Associate Editor of 2 other international Journals. He will be the General Chair of IPMU2020, was program co-chair and organizer of IFSA-EUSFLAT2009, Web chair of the 2010 IEEE World Congress on Computational Computation, FUZZ-IEEE2015 and FUZZ-IEEE2017 Publicity-chair, IPMU2016 program-chair, IEEE-WCCI2017 PR and Publicity-chair, and is PC member of more than 30 international conferences.
