=Paper=
{{Paper
|id=Vol-2960/preface
|storemode=property
|title=None
|pdfUrl=https://ceur-ws.org/Vol-2960/preface.pdf
|volume=Vol-2960
}}
==None==
https://ceur-ws.org/Vol-2960/preface.pdf
3rd Edition of Knowledge-aware and Conversational
Recommender Systems (KaRS) & 5th Edition of
Recommendation in Complex Environments (ComplexRec)
Joint Workshop
Vito Walter Anelli1 , Pierpaolo Basile2 , Tommaso Di Noia3 , Francesco Maria Donini4 ,
Cataldo Musto5 , Fedelucio Narducci6 , Markus Zanker7 , Himan Abdollahpouri 8 ,
Toine Bogers9 , Bamshad Mobasher10 , Casper Petersen11 and Maria Soledad Pera12
1
Polytechnic University of Bari
2
University of Bari Aldo Moro
3
Polytechnic University of Bari
4
University of Tuscia
5
University of Bari Aldo Moro
6
Polytechnic University of Bari
7
Free University of Bozen-Bolzano
8
Spotify
9
Aalborg University, Copenhagen
10
DePaul University
11
Sampension
12
Boise State University
Abstract
This is the preface for the joint workshop between KaRS and ComplexRec: two workshops co-located with the 15th ACM
RecSys 2021 conference.
Keywords
Recommender Systems, Workshop, Proceedings
1. Introduction ronments that have no simple one-size-fits-all
solution. In particular, we envisioned deepening
In this volume, we include the contributions presented the community understanding on complex inputs–
at the Joint KaRS & ComplexRec Workshop, co-located e.g., active user inputs (interaction), implicit user
with the 15𝑡ℎ edition of the ACM Conference on inputs (task, context, preferences), item inputs
Recommender Systems (RecSys) in Amsterdam. (features or attributes), domain inputs (eligibility,
availability)–and complex outputs–e.g., package
This joint workshop adopted a hybrid format aligned recommendation, composite items, interface com-
with the goal of this year’s main conference – congregat- plexity, constraint-based recommendation.
ing to continue to build community around recommeder • Providing a meeting forum for stimulating and
systems research and development. In this joint work- disseminating research in Knowledge-aware and
shop, we merged the main objectives envisioned for the Conversational Recommender Systems, where
3𝑟𝑑 Edition of the KaRS Workshop and the 5𝑡ℎ edition of researchers can network and discuss their re-
the Workshop on Recommendation in Complex Environ- search results in an informal way. In particular,
ments: we aimed to expand community understanding
• Providing an interactive venue for discussing ap- on knowledge-aware recommenders–from models
proaches to recommendation in complex envi- and feature engineering issues to beyond accu-
racy recommendation quality with a particular
3rd Edition of Knowledge-aware and Conversational Recommender focus on real-world applications–and conversa-
Systems (KaRS) & 5th Edition of Recommendation in Complex
tional recommenders– from the design of a con-
Environments (ComplexRec) Joint Workshop co-located with the 15th
ACM Conference on Recommender Systems (RecSys 2021) versational agent and its interface to the user mod-
© 2021 Copyright for this paper by its authors. Use permitted under Creative
Commons License Attribution 4.0 International (CC BY 4.0).
elling problems and evaluation issues.
CEUR
Workshop
Proceedings
http://ceur-ws.org
ISSN 1613-0073 CEUR Workshop Proceedings (CEUR-WS.org)
Overall, we accepted 17 contributions: 11 long as generating an explanation for the recommended items.
papers, 3 short papers, and 3 position papers. Each Furthermore, this side information becomes crucial
presentation was peer-reviewed by at least 3 program when a conversational interaction is implemented, in
committee (PC) members. The presentations of the particular for the preference elicitation, explanation, and
accepted contributions, along with the two keynote critiquing steps.
addressed by Edward C. Malthouse (during the virtual
component) and Gerard de Melo (during the in person The 3rd Knowledge-aware and Conversational Recom-
workshop component), sparked interactions among mender Systems (KaRS) Workshop focuses on all aspects
attendees and fostered ideas to continue to advance related to the exploitation of external and explicit knowl-
research focused around the topics of the joint workshop. edge sources to feed and build a recommendation engine,
and on the adoption of interactions based on the conver-
As KaRS & ComplexRec co-organizer, we want to sational paradigm. The aim is to go beyond the traditional
thank the RecSys 2021 workshop co-chairs, for their sup- accuracy goal and to start a new generation of algorithms
port regarding hybrid workshop organization. Last, but and approaches with the help of the methodological di-
not least, we would like to thank all authors and presen- versity embodied in fields such as Human–Computer
ters, as well as the members of the program committee Interaction, Conversational Recommender Systems, Se-
who selflessly shared their time and expertise in provid- mantic Web, and Knowledge Graphs. Consequently the
ing feedback to workshop authors. Finally, the workshop focus lies on works improving the user experience and
proceedings shall be submitted to CEUR-WS.org for on- following goals such as user engagement and satisfaction
line publication. or customer value.
The aim of this third edition of KaRS is to bring to-
gether researchers and practitioners around the topics
2. Workshop of Knowledge-aware of designing and evaluating novel approaches for recom-
and Conversational mender systems in order to:
Recommender Systems • Share research and techniques, including new de-
sign technologies and evaluation methodologies
2.1. Background and Goals • Identify next key challenges in the area
In the last few years, a renewed interest of the research • Identify emerging topics in the field
community on conversational recommender systems
(CRSs) is emerging. This is probably due to the great 2.2. Program
diffusion of Digital Assistants (DAs) such as Amazon
The program of the half-day workshop (part virtual, part
Alexa, Siri, or Google Assistant that are revolutionizing
in-person) consists of:
the way users interact with machines. DAs allow users
to execute a wide range of actions through an interaction • An invited keynote by Professor Gerard de Melo
mostly based on natural language messages. However, from the Hasso Plattner Institute for Digital Engi-
although DAs are able to complete tasks such as sending neering and the University of Potsdam, Germany.
texts, making phone calls, or playing songs, they are • The presentation of the selected research papers,
still at an early stage on offering recommendation
capabilities by using the conversational paradigm.
2.3. Website & Proceedings
In addition, we have been witnessing the advent of All workshop material including schedule and news
more and more precise and powerful recommendation will be found on the 2021 workshop website at
algorithms and techniques able to effectively assess https://kars-workshop.github.io/2021/.
users’ tastes and predict information that would prob-
ably be of interest to them. Most of these approaches
rely on the collaborative paradigm (often exploiting
machine learning techniques) and do not take into
2.4. Program Committee
account the huge amount of knowledge, both structured We thank the members of the PC for their thorough re-
and non-structured ones, describing the domain of views and their detailed feedback they gave to the authors.
interest of the recommendation engine. Although very The PC consisted of the following international experts.
effective in predicting relevant items, collaborative
approaches miss some very interesting features that • Vito Walter Anelli, POLITECNICO DI BARI
go beyond the accuracy of results and move in the • Azzurra Ragone, EY BUSINESS AND TECHNOL-
direction of providing novel and diverse results as well OGY SOLUTIONS
• Paolo Rosso, UNIVERSITAT POLITÈCNICA DE • Giovanni Semeraro, UNIVERSITY OF BARI
VALÈNCIA • Nourah Alrossais, UNIVERSITY OF YORK
• Andrea Iovine, UNIVERSITÀ DEGLI STUDI DI • Nicola Ferro, UNIVERSITY OF PADOVA
BARI ALDO MORO • Yashar Deldjoo, POLITECNICO DI BARI
• Fedelucio Narducci, POLITECNICO DI BARI • Claudio Pomo, POLITECNICO DI BARI
• Adir Solomon, BEN-GURION UNIVERSITY • Antonio Ferrara, POLITECNICO DI BARI
• Maurizio Ferrari Dacrema, POLITECNICO DI MI-
LANO
• Diego Antognini, ECOLE POLYTECHNIQUE 3. Workshop on Recommendation
FÉDÉRALE DE LAUSANNE in Complex Environments
• Marco Polignano, UNIVERSITÀ DEGLI STUDI
DI BARI ALDO MORO 3.1. Background and Goals
• Tommaso Di Noia, POLYTECHNIC UNIVERSITY
OF BARI During the past decade, recommender systems have
• Iván Cantador, UNIVERSIDAD AUTÓNOMA DE rapidly become an indispensable element of websites,
MADRID apps, and other platforms that seek to provide person-
• Marco de Gemmis, UNIVERSITY OF BARI ALDO alized interactions to their users. As recommendation
MORO technologies are applied to an ever-growing array of
non-standard problems and scenarios, researchers and
• Raffaele Perego, ISTI-CNR
practitioners are also increasingly faced with challenges
• Claudio Gennaro, ISTI-CNR
of dealing with greater variety and complexity in the
• Gianmaria Silvello, UNIVERSITY OF PADUA inputs to those recommender systems. For example,
• Cataldo Musto, DIPARTIMENTO DI INFORMAT- there has been more reliance on fine-grained user
ICA - UNIVERSITY OF BARI signals as inputs rather than simple ratings or likes.
• Davide Di Ruscio, UNIVERSITÀ DEGLI STUDI Applications require more complex domain-specific
DELL’AQUILA constraints on inputs to the recommender systems.
• Nicola Tonellotto, UNIVERSITY OF PISA Likewise, the outputs of recommender systems are
• Pierpaolo Basile, UNIVERSITY OF BARI moving towards more complex composite items, such as
• Alejandro Bellogin, UNIVERSIDAD AU- package or sequence recommendations. This increasing
TONOMA DE MADRID complexity requires smarter recommender algorithms
• Chiara Renso, ISTI-CNR, PISA, ITALY that can deal with this diversity in inputs and outputs.
• Pablo Sánchez, UNIVERSIDAD AUTÓNOMA DE
MADRID For the past four years, the ComplexRec workshop
• Benjamin Heitmann, RWTH AACHEN UNIVER- series has offered an interactive venue for discussing
SITY approaches to recommendation in complex scenarios
• Maria Maistro, UNIVERSITY OF COPENHAGEN that have no simple one-size-fits-all solution. For the
• Olga Marino, UNIVERSIDAD DE LOS ANDES fifth edition of ComplexRec we have narrowed the focus
• Francesco M. Donini, UNIVERSITA’ DELLA TUS- of the workshop and contributions to the workshop
CIA about topics related to one of the two main themes on
• Dietmar Jannach, UNIVERSITY OF KLAGEN- complex recommendation: complex inputs and complex
FURT outputs.
• Cristina Gena, UNIVERSITY OF TORINO
For the past four years [1, 2, 3, 4], the ComplexRec
• Giorgio Maria Di Nunzio, UNIVERSITY OF
workshop series has offered an interactive venue for dis-
PADUA
cussing approaches to recommendation in complex sce-
• Federica Cena, UNIVERSITY OF TORINO
narios that have no simple one-size-fits-all solution. For
• Markus Zanker, FREE UNIVERSITY OF BOZEN- the fifth edition of ComplexRec we have narrowed the
BOLZANO focus of the workshop and contributions to the work-
• Ludovico Boratto, UNIVERSITY OF CAGLIARI shop about topics related to one of the two main themes
• Franco Maria Nardini, ISTI-CNR on complex recommendation: complex inputs and com-
• Alain Starke, WAGENINGEN UNIVERSITY & RE- plex outputs. ComplexRec 2021 will take place as a joint
SEARCH workshop with KARS 2021.
• Toine Bogers, AALBORG UNIVERSITY
• Francesco Ricci, FREE UNIVERSITY OF BOZEN-
BOLZANO
3.2. Complex inputs 3.6. Program Committee
An important source of complexity comes from the vari- The ComplexRec 2021 organizers would like to thank the
ous types of inputs to the system beyond users and items, members of the program committee for their time and
such as features, queries and constraints. There are active effort to provide timely and constructive reviews of the
user inputs (interaction), implicit user inputs (task, con- submitted papers.
text, preferences), item inputs (features or attributes) and
• Panos Adamopoulos, EMORY UNIVERSITY
domain inputs (eligibility, availability). In group-based
• Ludovico Boratto, EURECAT
recommendation, the user input can be a combination • Christine Bauer, UNIVERSITY OF UTRECHT
of inputs for multiple individual users as well as group • Alejandro Bellogin, UNIVERSIDAD
aspects such as the composition of the group and how AUTÓNOMA DE MADRID
well they know each other. An additional challenge is • Iván Cantador, UNIVERSIDAD AUTÓNOMA DE
providing users with ways to have control over the in- MADRID
puts. For instance by selecting and weighting or ranking • Tommaso Di Noia, POLITECNICO DI BARI
user and item features, providing interactive queries to • Mehdi Elahi, UNIVERSITY OF BERGEN
steer the recommendation, or deal with longer narrative • Fabio Gasparetti, ROMA TRE UNIVERSITY
statements that require natural language understanding. • Pasquale Lops, UNIVERSITY OF BARI “ALDO
MORO”
3.3. Complex outputs • Mirko Marras, EPFL
• Cataldo Musto, UNIVERSITY OF BARI “ALDO
Another type of complexity that we wish to focus on in MORO”
ComplexRec 2020 is the complexity of the outputs of a • Fedelucio Narducci, UNIVERSITY OF BARI
recommender system to move away from a straightfor- • Markus Schedl, JOHANNES KEPLER UNIVER-
ward ranked list of items as output. An example of such SITY
complex output is package recommendation: suggest- • Peter Dolog, AALBORG UNIVERSITY
ing a set or combination of items that go well together • Cristina Gena, UNIVERSITA’ DEGLI STUDI DI
and are complementary on dimensions that matter to TORINO
the user. In many domains the sequence in which items • Hanna Schäfer, UNIVERSITÄT KONSTANZ
are recommended is also important. Moreover, different • Marco De Gemmis, UNIVERSITY OF BARI
users may want different information about items, so the • Bei Yu, SYRACUSE UNIVERSITY
output complexity goes beyond ranking and also man-
ifests itself in how the interface should allow the user
to view the type of information that is most relevant to References
them. Another example of complexity in recommender
systems output are environments where the system’s [1] T. Bogers, M. K. B. Mobasher, A. Said, A. Tuzhilin,
goal is to create new, composite items that must satisfy Complexrec 2017, in: Proceedings of the First Work-
certain constraints (such as menu recommendation, or shop on Recommendation in Complex Scenarios, vol-
recommendations for product designs). ume 1892, CEUR-WS, 2017, pp. 1–28.
[2] T. Bogers, M. Koolen, B. Mobasher, C. Pe-
tersen, A. Said, Complexrec 2018, in: Pro-
3.4. Program ceedings of the Second Workshop on Recom-
The program of the half-day workshop (completely vir- mendation in Complex Scenarios, 2018, pp.
tual) consists of: 1–37. URL: http://toinebogers.com/workshops/
complexrec2018/resources/proceedings.pdf.
• An invited keynote by Edward C. Malthous, Eras- [3] M. Koolen, T. Bogers, B. Mobasher, A. Tuzhilin, Com-
tus Otis Haven Professor at Northwestern Uni- plexrec 2019, in: Proceedings of the Third Workshop
versity, Illinois, United States. on Recommendation in Complex Scenarios, volume
• The presentation of the selected research papers, 2449, CEUR-WS, 2019, pp. 1–39.
[4] T. Bogers, M. Koolen, C. Petersen, B. Mobasher,
3.5. Website & Proceedings A. Tuzhilin, O. S. Shalom, D. Jannach, J. A. Konstan,
Recommendation in complex scenarios and the im-
All workshop material including schedule and news pact of recommender systems 2020, in: Proceedings
will be found on the 2021 workshop website at of the Workshops on Knowledge-aware and Conver-
https://complexrec2021.aau.dk/. sational Recommender Systems and Recommenda-
tion in Complex Scenarios, volume 2697, CEUR-WS,
2020, pp. 1–67.