Call for Challenge: 2nd Linked Open Data-enabled Recommender Systems Challenge

Challenge Website: http://sisinflab.poliba.it/events/lod-recsys-challenge-2015

General Chair

- Fabien Gandon (Inria, Sophia Antipolis, France)

Challenge Coordinators

- Elena Cabrio (Inria, Sophia Antipolis, France)

- Milan Stankovic (SEPAGE, Paris, France)

Challenge Chairs

- Iván Cantador (Universidad Autónoma de Madrid, Spain)

- Tommaso Di Noia (Polytechnic University of Bari, Italy)

- Vito Claudio Ostuni (Pandora Media, Inc. USA)

- Matthew Rowe (University of Lancaster, UK)

Important Dates

  • March 25, 2015, 23:59 CET: Paper and Results Submission due
  • April 16, 2015, 23:59 CET: Notification of acceptance and submission of task results
  • May 31 - June 4, 2015: The Challenge takes place at ESWC-15​

Motivation and Objectives

People generally need more and more advanced tools that go beyond those implementing the canonical search paradigm for seeking relevant information. A new search paradigm is emerging, where the user perspective is completely reversed: from finding to being found. 

Recommender systems may help to support this new perspective, since they have the effect of pushing relevant objects, selected from a large space of possible options, to potentially interested users. To achieve this result, recommendation techniques generally rely on data referring to three kinds of entities: users, items and their relations.

Recent developments in the Semantic Web community offer novel strategies to represent data about users, items and their relations that might improve the current state of the art of recommender systems, in order to move towards a new generation of recommender systems which fully understand the items they deal with.

More and more semantic data are published following the Linked Data principles, which enable to set up links between entities in different data sources, by connecting information in a single global data space: the Web of Data. 

Today, the Web of Data includes different types of knowledge represented in a homogeneous form: a sedimentary one (encyclopedic, cultural, linguistic, and common-sense) and a real-time one (news, data streams, etc.). These data might be useful to interlink diverse information about users, items, and their relations, and implement reasoning mechanisms that can support and improve the recommendation process.

The primary goal of this challenge is twofold. On the one hand, we want to enforce the link between the Semantic Web and the Recommender Systems communities. On the other hand, we aim to showcase how Linked Open Data and semantic technologies can boost the creation of a new breed of knowledge-enabled and content-based recommender systems.

Target Audience

The target audience is the above communities, both academic and industrial, which are interested in personalized information access with a particular emphasis on Linked Open Data.

During the last ACM RecSys conference the vast majority of participants were from industry. This is for sure a witness of the actual interest of recommender systems for industrial applications ready to be released in the market.

Dataset

We collected data from Facebook profiles about personal preferences (“likes”) for items in three domains: movies, books and music. After a process of user anonymization, we reconciled the item data with DBpedia entities. This data will be made available to train proposed recommendation approaches. In order to emphasize the usefulness of content-based data, only “cold users” will be available in the dataset.

Tasks

Task 1: Top-N recommendations from unary user feedback

This task deals with the top-N recommendation problem, in which a system is requested to find and recommend a limited set of N items that best match a user profile, instead of correctly predict the ratings for all available items.

In order to favor the proposal of content-based, LOD-enabled recommendation approaches, and limit the use of collaborative filtering approaches, this task aims at generating ranked lists of items for which only unary feedback (LIKE) is provided.

For this task, we will concentrate only on the movie domain. 

Task 2: Diversity within recommended item sets

A very interesting aspect of content-based recommender systems and then of LOD-enabled ones is giving the possibility to evaluate the diversity of recommended items in a straight way. This is a very popular topic in content-based recommender systems, which usually suffer from over-specialization.

In this task, the evaluation will be made by considering a combination of both accuracy of the recommendation list, and the diversity of items belonging to it. 

Focusing on recommending musical artists, we will consider diversity with respect to the <http://dbpedia.org/ontology/genre> and <http://purl.org/dc/terms/subject> properties.

Task 3: Cross-domain recommendation

This task aims to address a cross-domain recommendation scenario in which user preferences and/or domain knowledge of a source domain are used to recommend items in a different target domain.

This may correspond with the following use cases. The first refers to the well known cold-start problem, which hinders the recommendation generation due to the lack of sufficient information about users or items. In a cross-domain setting, a recommender may draw on information acquired from other domains to alleviate such problem, e.g., a user’s favorite movie genres may be derived from her favorite book genres. The second refers to the generation of personalized cross-selling or bundle recommendations for items from multiple domains, e.g., a movie accompanied by a music album similar to the soundtrack of the movie. These relations may not be extracted from rating correlations within a joined movie-music rating matrix.

In this task, we will request participants to exploit user preferences and domain knowledge about movies, in order to provide book recommendations. 

Making this task highly challenging, we will provide the list of books available in the test set, but we will provide little information about the users’ book preferences. Thus, we encourage not (only) to use collaborative filtering strategies based on correlations between movie and book preferences, but to investigate approaches that exploit LOD relating both movies and books domains.

Judging and Prizes

After a first round of reviews, the Program Committee and the chairs will select a number of submissions that will have to satisfy the challenge’s requirements, and will have to be presented at the conference. Submissions accepted for presentation will receive constructive reviews from the Program Committee, and will be included in post-proceedings. All accepted submissions will have a slot in a poster session dedicated to the challenge. In addition, the winners will present their work in a special slot of the main program of ESWC’15, and will be invited to submit a chapter to a post-proceedings book published by Springer (Communications in Computer and Information Science series).

For each task we will select:

* the best performing tool, given to the paper which will get the highest score in the evaluation

* the most original approach, selected by the Challenge Program Committee with the reviewing process

How to Participate

The following information has to be provided:

* Abstract: no more than 200 words.

* Description: It should contain the details of the system, including why the system is innovative, how it uses Semantic Web, which features or functions the system provides, what design choices were made, and what lessons were learned. The description should also summarize how participants have addressed the evaluation tasks. Papers must be submitted in PDF format, following the style of the Springer’s Lecture Notes in Computer Science (LNCS) series (http://www.springer.com/computer/lncs/lncs+authors), and not exceeding 12 pages in length.

* Result evaluation: For the three tasks previously described, a Web-accessible service will be provided in order to evaluate the produced results. All the details about the format and the service URL will be provided on the website.

All submissions should be provided via EasyChair

We invite the potential participants to subscribe to our mailing list in order to be kept up to date with the latest news related to the challenge. 

[email protected]

Program Committee  [TBC]

  • * Pablo Castells, Universidad Autónoma de Madrid, Spain
  • * Oscar Corcho, Universidad Politécnica de Madrid, Spain
  • * Marco de Gemmis, University of Bari Aldo Moro, Italy
  • * Ignacio Fernández-Tobías, Universidad Autónoma de Madrid
  • * Frank Hopfgartner, Technische Universität Berlin, Germany
  • * Andreas Hotho, Universität Würzburg, Germany
  • * Dietmar Jannach, TU Dortmund University, Germany
  • * Pasquale Lops, University of Bari Aldo Moro, Italy
  • * Valentina Maccatrozzo, Delft University of Technology, The Netherlands
  • * Alexandre Passant, seevl.fm, Ireland
  • * Francesco Ricci, Free University of Bozen-Bolzano, Italy
  • * Mariano Rico, Universidad Politécnica de Madrid, Spain
  • * Giovanni Semeraro, University of Bari Aldo Moro, Italy
  • * Manolis Wallace, University of Peloponnese, Greece
  • * Markus Zanker, Alpen-Adria-Universitaet Klagenfurt, Austria

 

More information is available at http://sisinflab.poliba.it/events/lod-recsys-challenge-2015