Modern online platforms for user modeling and recommendation require complex data infrastructures to collect and process data. Some of this data has to be kept to later be used in batches to train personalization models. However, since user activity data can be generated at very fast rates it is also useful to have algorithms able to process data streams online, in real time. Given the continuous and potentially fast change of content, context and user preferences or intents, stream-based models, and their synchronization with batch models can be extremely challenging. Therefore, it is important to investigate methods able to transparently and continuously adapt to the inherent dynamics of user interactions, preferably over long periods of time. Models able to continuously learn from such flows of data are gaining attention in the recommender systems community, and are being increasingly deployed in online platforms. These challenges motivate the research on online methods that are able to adapt over time, preferably without retraining models from scratch. Techniques available from online machine learning, data stream mining and continual learning have been gaining maturity in recent years, however this has happened mostly for conventional machine learning tasks. The application of this body of knowledge on online recommendation and user modeling is not direct or trivial, and needs further investigation and scientific contributions.
The objective of this workshop is to foster contributions and bring together a growing community of researchers and practitioners interested in online, adaptive approaches to user modeling, recommendation and personalization, and their implications regarding multiple dimensions, such as reproducibility, privacy, fairness, diversity, transparency, auditability, and compliance with recently adopted or upcoming legal frameworks worldwide.
We welcome two types of contributions: original research papers and brainstorming ideas.
Relevant topics include, but are not limited to:
- Stream-based and incremental algorithms
- Continual learning
- Mechanisms for forgetting and retaining
- Long term and lifelong user modeling
- Preference change detection and adaptation
- Context change detection and adaptation
- Session-based and sequential learning
- Online distributed and decentralized models
- Online learning with bandits and reinforcement learning
- Learning from evolving graphs
- Online automation and self-tuning
- Time-sensitive recommendation
- Online monitoring, evaluation and benchmarking
- Privacy and user agency in incremental models
- Reproducibility, auditability and legal compliance
- Scalability of online algorithms
- Platforms, software, data, and architectures
- Industrial case studies
We welcome original, unpublished work in the form of paper submissions via EasyChair at:
https://easychair.org/conferences/?conf=orsum2023.
Papers must not exceed 16 pages (plus up to 2 pages for references), and should report research at a mature stage.
We also welcome the submission of preliminary results of ongoing research in the form of shorter papers with a maximum length of 8 pages (plus up to 2 pages for references). Papers must be formatted in single column format according to the guidelines in the
ACM RecSys 2023 call for papers (template available
from ACM).
The review process is double-blind, so authors are required to remove any content that allows author identification.
2023-07-12 2023-08-03: Paper submission deadline
2023-08-24: Paper acceptance notification
2023-09-14: Camera-ready paper deadline
2023-09-19: Workshop date
All deadlines are at 11:59pm AoE.
Call for Brainstorming Ideas
We also welcome research and application ideas related to the above topics in the form of extended abstracts. The purpose of such ideas is to hold brainstorming sessions at the workshop. Examples include ideas for papers, projects, software tools, student projects (MSc or PhD level), events, competitions. The most promising ideas will be selected by the workshop organizers based on originality and match with the workshop topics. Selected ideas will be briefly presented by authors and discussed in brainstorming sessions involving all workshop participants.
Brainstorming ideas can be submitted via EasyChair at:
https://easychair.org/conferences/?conf=orsum2023.
Extended abstracts must not exceed 3 pages (excluding references), and must be formatted in single column format according to the guidelines in the
ACM RecSys 2023 call for papers (template available
from ACM).
The review process is single-blind, so authors are not required to omit their identification.
Authors of selected ideas will be invited to submit a post-event report with the main outcomes of the sessions, to be published with the workshop proceedings.
Early batch:
2023-07-19 2023-08-10: Ideas early submission deadline
2023-08-17: Ideas early acceptance notification
Late batch:
2023-09-05: Ideas late submission deadline
2023-09-08: Ideas late submission notification
2023-09-19: Workshop date
TBA (after workshop): Ideas discussion report
All deadlines are at 11:59pm AoE.
Papers will be first published on the workshop website and will serve two purposes: (1) as a reference to attendants and (2) proofs for the proceedings. Workshop proceedings will be available after the workshop as a dedicated volume in an open repository (e.g. CEUR-WS, PMLR, arXiv), and will include keynote abstracts, original papers, and brainstorming session reports.
Contributions will be published under the Creative Commons Attribution 4.0 International License, as specified
at http://creativecommons.org/licenses/by/4.0/legalcode
(human readable summary here).