This book provides an introduction to reciprocal recommendation. It starts with theory, and then moves on to concrete examples of the most successful algorithms in the field. Researchers and developers with a little background in machine learning will find many of the algorithms are straightforward to implement, and code samples are included to help with this.
In addition to accessible algorithms, the book also examines some more cutting-edge research such as the recent interest in applying matching theory to reciprocal recommendation. These parts will be of interest both to developers who are looking to optimize their systems, and to researchers who might find avenues to further advance the field and develop new methods of recommending people to people.
By the end of this book, the reader will have a comprehensive understanding of the state of the art in reciprocal recommendation and will be equipped to design and implement their own systems.
Preface.- 1. Introduction.- 2. Theoretical Background.- 3. Collaborative Filtering.- 4. Content-Based Filtering.- 5. Hybrid Filtering and Additional Approaches.- 6. Matching Theory.- 7. Ethical Concerns and Future Work.
James Neve is a machine learning researcher with Eureka Inc. in Tokyo, designing AI systems including Reciprocal Recommender Systems (RRSs) for online dating services. He has a PhD in Machine Learning from the University of Bristol, specialized in RRSs, and he has published multiple papers on reciprocal recommendation in competitive conferences such as ACM RecSys.
