MoST-Rec 2019: Workshop on Model Selection and Parameter Tuning in Recommender Systems Beijing, China, November 4-7, 2019 |
Conference website | http://most-rec.gt-arc.com |
Submission link | https://easychair.org/conferences/?conf=mostrec2019 |
Abstract registration deadline | August 14, 2019 |
Submission deadline | August 14, 2019 |
The MoST-Rec Workshop @ CIKM 2019 is a place for discussing and exchanging recent advances and open challenges between the Model-Selection and Parameter Tuning community and the Recommendation Systems community.
Recommender systems (RS) have attracted strong attention of the machine learning community, especially within the last decades. Researchers have developed various algorithms proven to have good performance in the laboratory environment; however, applying them to real business cases is typically more difficult.
This workshop addresses the issues of algorithm selection and parameter tuning for recommender systems. The goal is to bring together researchers from the machine learning community with the industry representatives in order to exchange information on current challenges, constraints and ideas from both domains.
Submission Guidelines
All papers must be original and not simultaneously submitted to another journal or conference. Submissions should be made through the EasyChair Conference System and must be formatted according to ACM SIG proceedings format and prepared as a PDF file. Both regular papers and demo/poster papers are welcome:
- Long papers: Max 8 pages
- Short papers (including demo/poster): Max 4 pages
List of Topics
MoST-Rec Workshop aims to connect the domains of Recommender Systems (RS), Model Selection (MS) and Parameter Tuning (PT), and to facilitate knowledge exchange between the communities of these research areas. Therefore, we are looking for topics that explain methods, challenges and insights at the intersection of these domains. In particular, topics of solicited papers include, but are not limited to:
- Model Selection and Parameter Tuning for Recommender Systems
- Ensemble methods
- Online model selection / ensembles
- Online boosting
- Parameter tuning
- High noise model selection / tuning
- Sparsely labeled model selection / tuning
- Distributing model selection or parameter tuning
- Recommender Systems applying model selection and parameter tuning methods
- Short term temporal dynamics (Item popularity, trends)
- Long term temporal dynamics (user tastes)
- Continously changing sets of users and items
- Scenarios with sparse rewards
- Tuning of robustness, convergence, lerning-rate
- Considerations of popularity bias (in the evaluation metrics, learning procedure)
New: Selected papers will be considered for inclusion in a special issue of the Springer International Journal of Data Science and Analytics (JDSA).
Contact
All questions about submissions should be emailed to most-rec@gt-arc.com.