IJMIR: Special Issue on Multimedia Recommendation Systems

Multimedia Recommendation Systems

A Special Issue of the International Journal of Multimedia Information Retrieval

Edited by Yashar Deldjoo and Markus Schedl

Contact for inquiries: Yashar Deldjoo (deldjooy@acm.org)

Recommendation systems have become a crucial means to manage the ever increasing amount of multimedia content available today, and to help users discover interesting new items. While the recommender systems and the multimedia communities have researched great tools to address problems in their areas, the innovative combination of state-of-the-art recommender systems technology and multimedia content analysis to build content-based and hybrid recommendation systems for media or other items has not been subject of a wider discussion yet. With this Special Issue we aim to bridge this gap between communities and provide a venue for exciting new research on recommender systems that leverage multimedia content.

Call for Papers

We solicit original research that either use multimedia content (e.g., audio, visual, textual content) to recommend media items (e.g., movies, music, images) or non-media items (e.g., fashion products or e-commerce products). Also hybrid and context-aware recommendation approaches are welcome, as long they leverage at least one content modality, irrespective of its representation, e.g., including raw signal data as well as semantic descriptors extracted from knowledge bases or graphs. Purely CF-based methods are out-of-scope.

We encourage original submissions of excellent quality that are not submitted to or accepted by any other journal or conference. Substantially extended versions of conference or workshop papers (at least 30% novel content) are welcome as well. Papers should not exceed 14 pages in the Springer double-column format.

All submissions to this Special Issue will be peer-reviewed by at least three members of the Guest Advisory Board. The review process will be single-blind. After a first review cycle, we will select according to the reviewing results a small number of submissions which might be considered for acceptance. In a second review cycle the authors of the selected submissions will have the chance to modify their submissions according to the reviewers suggestions, before a final decision for acceptance or rejection will be made.

Topics of interest include the following:

Hybrid recommendation systems for multimedia content
Deep learning from multimedia signals for recommendation systems
Improving session-based recommendation systems by content models
Combating cold-start by leveraging multimedia content
User modeling and profiling for multimedia recommendation (including use of knowledge bases or graphs)
Improving beyond-accuracy performance of recommender systems through multimedia (e.g., diversity, coverage, serendipity)
Studies on the human understanding and perception of multimedia content with direct implications on recommender systems
Predicting and integrating user intent into multimedia recommendation
Using multimedia content for transparent and/or fair recommendations
Privacy-aware recommendation (complying to the general data protection regulation)
New evaluation metrics for content-based multimedia recommender systems
New datasets accompanied by solid case studies of their application
Novel (or under-researched) applications areas of content-based recommender systems (e.g., podcast, speech, health, art, or fashion recommendation)

Important Dates

Paper Submission Deadline:	November 24, 2019
Notification After First Review Cycle:	January 15, 2020
Paper Revisions Deadline:	February 28, 2020
Final Notification:	April 15, 2020
Submission of Camera Ready Paper:	May 15, 2020

Submission

Submissions will be managed by Springer Editorial Manager. Please create a user account if you have not already done so, login and follow the instructions to submit a new contribution.

Guest Editorial Board

Yashar Deldjoo	Politecnico di Bari, Italy
Markus Schedl	Johannes Kepler University Linz, Austria