UMUAI Special Issue on
Conversational Recommender Systems:
Theory, Models, Evaluations, and Trends

Motivations and Scope

In the last few years, a renewed interest of the research community in conversational recommender systems (CRSs) is emerging. This is likely due to the massive proliferation of Digital Assistants (DAs) such as Amazon Alexa, Siri, or Google Assistant, which are revolutionizing how people interact with machines.

DAs allow users to execute various actions through interactions primarily based on natural language utterances. However, although DAs are able to complete tasks such as sending texts, making phone calls, or playing songs, they remain at an early stage in terms of their recommendation-offering capabilities using the conversational paradigm. Very few studies investigated the theory behind conversational recommendations in terms of strategies and algorithms, complexity analysis, and so on.

Furthermore, since the recommendation becomes an incremental process where the user refines her request at each step to reach the recommendation goal, CRSs need to be powered with knowledge about the domain of the recommended items. Knowledge can be represented in different forms, such as structured information, unstructured one, knowledge graphs, etc.

Finally, the evaluation of CRSs is a crucial and non-conventional step, since it goes beyond the accuracy of results and moves in the direction of providing novel and diverse results, generating explanations for recommended items, reducing the user effort for reaching a liked item (e.g., in terms of exchanged messages), and often requires designing a user study.

Special Issue Topics

We are interested in contributions focusing on different new and relevant aspects of conversational recommender systems, in particular, new developments on the algorithmic and user interface level as well as new applications, all accompanied by a corresponding evaluation (e.g., empirical study) that clearly shows significant improvements compared to the state of the art.

The general topics regarded as relevant for the special issue on Conversational Recommender Systems include but are not limited to

  • Methodological aspects of a Conversational Recommender Systems (CRS)
    • Design and implementation methodologies
    • Dialogue management (end-to-end, dialogue-state-tracker
    • models, retrieval-based, hybrid)
    • Recommendation algorithms for CRS
    • Multi-objective recommendations for CRS
    • Dialogue protocol design
    • Next-question problem
    • Knowledge bases and knowledge graphs for CRS
    • Explainable CRS
  • User Modeling and Interfaces for CRS
    • Short- and Long-term user profiling and modeling
    • Preference elicitation
    • Critiquing and user feedback elicitation
    • UX design
    • Natural language, multimodal, and voice-based interfaces
  • Theoretical aspects and conceptual aspects of CRS
  • Evaluation of CRS
    • User studies
    • Online experiments
    • Metrics beyond accuracy
    • Datasets

Important Dates

  • Abstract submission: February 15, 2023
  • Abstract notification: February 28, 2023
  • Paper submission: April 15, 2023
  • Author notification: July 1, 2023
  • Revised paper submission: July 25, 2023
  • Final notification: August 25, 2023
  • Camera-ready paper submission: October 15, 2023

All deadlines are 11:59pm, AoE time (Anywhere on Earth).

Submission and Review Details

This special issue follows a three-stage process. In the first stage, authors are invited to submit an extended abstract. The guest editors of the special issue will then screen all submitted extended abstracts and will invite authors of submissions that pass this screening to submit a full manuscript.A two-round peer review process will then follow. Accepted papers will be published in the Special Issue.

  • Abstract Submission: Extended abstracts (up to three pages in journal format, not counting references) should be submitted through EasyChair: https://easychair.org/conferences/?conf=umuaisicrs23. The abstract must be formatted according to UMUAI's template and guidelines.
  • Full Manuscript Submission: After abstract pre-screening, the authors of selected papers will be asked to submit a complete version of the paper through UMUAI journal submission system.

All submissions must be written in English. We expect authors, reviewers, and the organizing committee to adhere to the ACM’s Conflict of Interest Policy and the ACM’s Code of Ethics and Professional Conduct. Authors should use follow the journal's submission instructions.

Authors are welcome to submit extended versions of papers already published in conferences and workshops. Submissions must follow UMUAI's policy on previously published material. Authors must explicitly refer in the Extended Abstract submission that they are extending previous work and indicate the added value of their submission with respect to the work being extended.

Committee

Guest Editors

Contacts

For general enquiries on the workshop, please send an email to vitowalter.anelli@poliba.it, lichen@comp.hkbu.edu.hk, gerard.demelo@hpi.de, jmcauley@eng.ucsd.edu, fedelucio.narducci@poliba.it, and azzurra.ragone@uniba.it.