Recommendation System—Explainability, Evaluation and Rank with User Behavior Dynamics

Dear Colleagues,

Recommendation systems have been becoming increasingly more essential in people’s lives. Hence, more research and development efforts have been invested by researchers in academic institutes and R&D departments in various businesses.

The application of effective recommendation systems is present in many aspects of people’s day-to-day lives, such as being used by Netflix, YouTube, and TikTok users to help decide which movie or video to watch, Spotify for letting people know what music to listen to, LinkedIn for discovering matched candidates or jobs, and Stitch Fix for becoming your personalized stylist. At the same time, people can almost search for anything on Google, and it will recommend relevant information. Meanwhile, recommendation systems are still facing a series of challenges. In this Special Issue, we would like to see topics such as model explainability, evaluation, and learning to rank with the dynamics of user behaviors.

Model explainability: Addressing the problem of the building transparency of models, increasing the maintainability of models, and enhancing the decision-making process for end users. In recent years, explainability in AI has grown in importance for various reasons, such as fairness and transparency. It is specifically meaningful in recommendation systems as it helps system developers to understand the system better and, hence, can efficiently improve the system. It also can potentially better engage the end users by providing context as to why an item is recommended.

Model evaluation: Addressing the challenge of data sparsity, such as how to collect feedback for evaluation, how to use limited data for assessments, and how to choose which metrics to optimize in various scenarios.

Learning to rank with the dynamics of user behaviors: First, we need to address the general ranking problem—what is the first item a system would want the user to see or use; hence, bringing more user engagement and satisfaction. Second, how should ranking mechanisms be captured and adapted to changed use behaviors.

The above is a shortlist of recommended focus areas. However, in this Special Issue, topics are not limited to those mentioned above. We encourage researchers and developers to share and submit manuscripts on a broader scope of recommendations systems. In addition to algorithm improvements and vision expansions, domain-specific applications are also welcome.

Dr. Qi Liu
Dr. Yingying Zhao
Dr. Giorgio Maria Di Nunzio
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Information is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• recommendation system
• deep learning
• machine learning
• model explainability
• model evaluation
• applications