Alan Said

A month in the life of a production news recommender system

During the last decade, recommender systems have become a ubiquitous feature in the online world. Research on systems and algorithms in this area has flourished, leading to novel techniques for personalization and recommendation. The evaluation of recommender systems, however, has not seen similar progress---techniques have changed little since the advent of recommender systems, when evaluation methodologies were borrowed from related research areas. As an effort to move evaluation methodology forward, this paper describes a production recommender system infrastructure that allows research systems to be evaluated in situ, by real-world metrics such as user clickthrough. We present an analysis of one month of interactions with this infrastructure and share our findings.

Towards Health (Aware) Recommender Systems

People increasingly use the Internet for obtaining information regarding diseases, diagnoses and available treatments. Currently, many online health portals already provide non-personalized health information in the form of articles. However, it can be challenging to find information relevant to ones condition, interpret this in context, and understand the medical terms and relationships. Recommender Systems (RS) already help these systems perform precise information filtering. In this short paper, we look one step ahead and show the progress made towards RS helping users find personalized, complex medical interventions or support them with preventive healthcare measures. We identify key challenges that need to be addressed for RS to offer the kind of decision support needed in high-risk domains like healthcare.

Do recommendations matter?: news recommendation in real life

We present a study of how recommendations are received in real life by users across different news domains (traditional online newspapers, hobbyist websites, forums, etc.). Our analysis shows that readers of websites centered around specific topics are generally less likely to interact with recommendations than readers of traditional news websites.

Engendering Health with Recommender Systems

The first Workshop on Engendering Health with Recommender Systems was organized in conjunction with ACM RecSys 2016. The focus of the workshop was on bringing together researchers and practitioners from diverse areas of health, well-being, decision support, and behavioral change. Health-related issues in recommender systems have been a growing research topic in the recent years and this was a initial attempt at bringing together academics and practitioners to share their experiences on working on related issues.

Introduction to the Special Issue on Recommender System Benchmarking

Recommender systems addvalue to vast content resources by matching users with items of interest. In recent years, immense progress has been made in recommendation techniques. The evaluation of these systems is still based on traditional information retrieval and statistics metrics (e.g., precision, recall, RMSE), often not taking the use case and situation of the system into consideration. However, the rapid evolution of recommender systems in both their goals and their application domains fosters the need for new evaluation methodologies and environments. This special issue serves as a venue for work on novel, recommendation-centric benchmarking approaches taking the users’ utility, the business values, and the technical constraints into consideration. Building on the success of the Recommendation Utility Evaluation Workshop [Amatriain et al. 2012] held at Recsys 2012, the Workshop on Benchmarking Adaptive Retrieval and Recommender Systems [Castells et al. 2013a, 2013b] held at SIGIR 2013, the Workshop on Reproducibility and Replication in Recommender System Evaluation [Bellogı́n et al. 2013, 2014], the various Recommender System Challenges [Adomavicius et al. 2010; Said et al. 2011; Manouselis et al. 2012; Blomo et al. 2013; Said et al. 2014; Ben-Shimon et al. 2015], and other similar events, this special issue collects articles focusing on various aspects of challenges in benchmarking and evaluation or recommender systems. In the decade since the inception of the ACM RecSys conference and the decades since the first papers on this topic started to appear [Resnick et al. 1994], the field has

Replicable Evaluation of Recommender Systems

Recommender systems research is by and large based on comparisons of recommendation algorithms predictive accuracies: the better the evaluation metrics (higher accuracy scores or lower predictive errors), the better the recommendation algorithm. Comparing the evaluation results of two recommendation approaches is however a difficult process as there are very many factors to be considered in the implementation of an algorithm, its evaluation, and how datasets are processed and prepared. This tutorial shows how to present evaluation results in a clear and concise manner, while ensuring that the results are comparable, replicable and unbiased. These insights are not limited to recommender systems research alone, but are also valid for experiments with other types of personalized interactions and contextual information access.

User Nutrition Modelling and Recommendation: Balancing Simplicity and Complexity

In order to use and model nutritional knowledge in a food recommender system, uncertainties regarding the users nutritional state and thus the personal health value of food items, as well as conflicting nutritional theories need to be quantified, qualified and subsumed into falsifiable models. In this paper, we reflect on different error sources with respect to nutrition and consider how such issues can be tackled in future systems. We discuss the integration of general nutritional theories into information systems as well as user specific nutritional measures and different approaches to evaluating the utility of a given nutritional model.

Second Workshop on Health Recommender Systems: (HealthRecSys 2017)

The 2017 Workshop on Health Recommender Systems was held in conjunction with the 2017 ACM Conference on Recommender Systems in Como, Italy. Following the fists workshop in 2016, the focus of this workshop was on enhancing the results of the first workshop by elaborating discussions on the topics, attracting scientist from other domains, finding cross-domain collaboration, and establishing shared infrastructures.

A Short History of the RecSys Challenge

The RecSys Challenge is a yearly recurring competition focusing on creating the best performing recommendation approach for a specific scenario. Over the years, the competition has drawn many participants from industry and academia, and has become an key part of the ACM Conference on Recommender Systems series. This article presents a brief historical overview of the RecSys Challenge from its inception in 2010 until the seventh iteration in 2016.

Information Retrieval and Recommender Systems

This chapter provides a brief introduction to two of the most common applications of data science methods in e-commerce: information retrieval and recommender systems. First, a brief overview of the systems is presented followed by details on some of the most commonly applied models used for these systems and how these systems are evaluated. The chapter ends with an overview of some of the application areas in which information retrieval and recommender systems are typically developed.