Graph-based techniques relying on neural networks and embeddings have gained attention as a way to develop Recommender Systems (RS) with several papers on the topic presented at SIGIR 2022 and 2023. Given the importance of ensuring that published research is methodologically sound and reproducible, in this paper, we analyze 10 graph-based RS papers, most of which were published at SIGIR 2022, and assess their impact on subsequent work published in SIGIR 2023. Our analysis reveals several critical points that require attention: (i) the prevalence of bad practices, such as erroneous data splits or information leakage between training and testing data, which call into question the validity of the results; (ii) frequent inconsistencies between the provided artifacts (source code and data) and their descriptions in the paper, causing uncertainty about what is actually being evaluated; and (iii) the preference for new or complex baselines that are weaker compared to simpler ones, creating the impression of continuous improvement even when, particularly for the Amazon-Book dataset, the state-of-the-art has significantly worsened. Due to these issues, we are unable to confirm the claims made in most of the papers that we examined and attempted to reproduce.

Countless new machine learning models are published every year and are reported to significantly advance the state-of-the-art in top-n recommendation. However, earlier reproducibility studies indicate that progress in this area may be quite limited, due to widespread methodological issues, e.g., comparisons with untuned baseline models, creating an illusion of progress . In this work, we examine whether these problems persist in today’s research by attempting to reproduce nine SIGIR 2023 and 2024 recommendation algorithms based on Denoising Diffusion Probabilistic Models, a recent but rapidly expanding research area. Only 25% of reported results are fully reproducible and, since the original papers relied on weak baselines , they do not establish the superiority of diffusion models over state-of-the-art methods. In our controlled evaluations, well-tuned simpler baselines consistently exceed the diffusion-based models’ effectiveness reported in the original papers. Furthermore, we identify key mismatches between the characteristics of diffusion models and those of the traditional top-n recommendation task, raising doubts about their suitability for recommendation. Moreover, in the analyzed papers, the generative capabilities of these models are constrained to a minimum. Overall, our results call for greater scientific rigor and a disruptive change in the research and publication culture in this area.

Lately, we have observed a growing interest in intent-aware recommender systems (IARS). The promise of such systems is that they are capable of generating better recommendations by predicting and considering the underlying motivations and short-term goals of consumers. From a technical perspective, various sophisticated neural models were recently proposed in this emerging and promising area. In the broader context of complex neural recommendation models, a growing number of research works unfortunately indicates that (i) reproducing such works is often difficult and (ii) that the true benefits of such models may be limited in reality, e.g., because the reported improvements were obtained through comparisons with untuned or weak baselines. In this work, we investigate if recent research in IARS is similarly affected by such problems. Specifically, we tried to reproduce five contemporary IARS models that were published in top-level outlets, and we benchmarked them against a number of traditional non-neural recommendation models. In two of the cases, running the provided code with the optimal hyperparameters reported in the paper did not yield the results reported in the paper. Worryingly, we find that all examined IARS approaches are consistently outperformed by at least one traditional model. These findings point to sustained methodological issues and to a pressing need for more rigorous scholarly practices.

Novel data sources bring new opportunities to improve the quality of recommender systems and serve as a catalyst for the creation of new paradigms on personalized recommendations. Impressions are a novel data source containing the items shown to users on their screens. Past research focused on providing personalized recommendations using interactions and occasionally using impressions when such a data source was available. Interest in impressions has increased due to their potential to provide more accurate recommendations. Despite this increased interest, research in recommender systems using impressions is still dispersed. Many works have distinct interpretations of impressions and use impressions in recommender systems in numerous different manners. To unify those interpretations into a single framework, we present a systematic literature review on recommender systems using impressions, focusing on three fundamental perspectives: recommendation models , datasets , and evaluation methodologies . We define a theoretical framework to delimit recommender systems using impressions and a novel paradigm for personalized recommendations, called impression-aware recommender systems. We propose a classification system for recommenders in this paradigm, which we use to categorize the recommendation models, datasets, and evaluation methodologies used in past research. Last, we identify open questions and future directions, highlighting missing aspects in the reviewed literature.

The field of Quantum Computing (QC) has gained significant popularity in recent years, due to its potential to provide benefits in terms of efficiency and effectiveness when employed to solve certain computationally intensive tasks. In both Information Retrieval (IR) and Recommender Systems (RS) we are required to build methods that apply complex processing on large and heterogeneous datasets, it is natural therefore to wonder whether QC could also be applied to boost their performance. The tutorial aims to provide first an introduction to QC for an audience that is not familiar with the technology, then to show how to apply the QC paradigm of Quantum Annealing (QA) to solve practical problems that are currently faced by IR and RS systems. During the tutorial, participants will be provided with the fundamentals required to understand QC and to apply it in practice by using a real D-Wave quantum annealer through APIs.

Recommender systems typically rely on past user interactions as the primary source of information for making predictions. However, although highly informative, past user interactions are strongly biased. Impressions, on the other hand, are a new source of information that indicate the items displayed on screen when the user interacted (or not) with them, and have the potential to impact the field of recommender systems in several ways. Early research on impressions was constrained by the limited availability of public datasets, but this is rapidly changing and, as a consequence, interest in impressions has increased. Impressions present new research questions and opportunities, but also bring new challenges. Several works propose to use impressions as part of recommender models in various ways and discuss their information content. Others explore their potential in off-policy-estimation and reinforcement learning. Overall, the interest of the community is growing, but efforts in this direction remain disconnected. Therefore, we believe that a workshop would be useful in bringing the community together.

The Off-Policy Evaluation (OPE) problem consists in evaluating the performance of new policies from the data collected by another one. OPE is crucial when evaluating a new policy online is too expensive or risky. Many of the state-of-the-art OPE estimators are based on the Inverse Propensity Scoring (IPS) technique, which provides an unbiased estimator when the full support assumption holds, i.e., when the logging policy assigns a non-zero probability to each action. However, there are several scenarios where this assumption does not hold in practice, i.e., there is deficient support, and the IPS estimator is biased in the general case.In this paper, we consider two alternative estimators for the deficient support OPE problem. We first show how to adapt an estimator that was originally proposed for a different domain to the deficient support setting.Then, we propose another estimator, which is a novel contribution of this paper.These estimators exploit additional information about the actions, which we call side information, in order to make reliable estimates on the unsupported actions. Under alternative assumptions that do not require full support, we show that the considered estimators are unbiased.We also provide a theoretical analysis of the concentration when relaxing all the assumptions. Finally, we provide an experimental evaluation showing how the considered estimators are better suited for the deficient support setting compared to the baselines.

In Recommender Systems, impressions are a relatively new type of information that records all products previously shown to the users. They are also a complex source of information, combining the effects of the recommender system that generated them, search results, or business rules that may select specific products for recommendations. The fact that the user interacted with a specific item given a list of recommended ones may benefit from a richer interaction signal, in which some items the user did not interact with may be considered negative interactions. This work presents a preliminary evaluation of recommendation models with impressions. First, impressions are characterized by describing their assumptions, signals, and challenges. Then, an evaluation study with impressions is described. The study’s goal is two-fold: to measure the effects of impressions data on properly-tuned recommendation models using current open-source datasets and disentangle the signals within impressions data. Preliminary results suggest that impressions data and signals are nuanced, complex, and effective at improving the recommendation quality of recommenders. This work publishes the source code, datasets, and scripts used in the evaluation to promote reproducibility in the domain.

After decades of being mainly confined to theoretical research, Quantum Computing is now becoming a useful tool for solving realistic problems. This work aims to experimentally explore the feasibility of using currently available quantum computers, based on the Quantum Annealing paradigm, to build a recommender system exploiting community detection. Community detection, by partitioning users and items into densely connected clusters, can boost the accuracy of non-personalized recommendation by assuming that users within each community share similar tastes. However, community detection is a computationally expensive process. The recent availability of Quantum Annealers as cloud-based devices, constitutes a new and promising direction to explore community detection, although effectively leveraging this new technology is a long-term path that still requires advancements in both hardware and algorithms. This work aims to begin this path by assessing the quality of community detection formulated as a Quadratic Unconstrained Binary Optimization problem on a real recommendation scenario. Results on several datasets show that the quantum solver is able to detect communities of comparable quality with respect to classical solvers, but with better speedup, and the non-personalized recommendation models built on top of these communities exhibit improved recommendation quality. The takeaway is that quantum computing, although in its early stages of maturity and applicability, shows promise in its ability to support new recommendation models and to bring improved scalability as technology evolves.

Feature selection is a common step in many ranking, classification, or prediction tasks and serves many purposes. By removing redundant or noisy features, the accuracy of ranking or classification can be improved and the computational cost of the subsequent learning steps can be reduced. However, feature selection can be itself a computationally expensive process. While for decades confined to theoretical algorithmic papers, quantum computing is now becoming a viable tool to tackle realistic problems, in particular special-purpose solvers based on the Quantum Annealing paradigm. This paper aims to explore the feasibility of using currently available quantum computing architectures to solve some quadratic feature selection algorithms for both ranking and classification.

The evaluation of recommender systems is frequently focused on accuracy metrics, but this is only part of the picture. The diversity of recommendations is another important dimension that has received renewed interest in recent years. It is known that accuracy and diversity can be conflicting goals and finding appropriate ways to combine them is still an open research question. Several ways have been proposed to measure the diversity of recommendations and to include its optimization in the loss function used to train the model. Methods optimizing list based diversity suffer from two drawbacks: the high computational cost of the loss function and the lack of an efficient way to optimize them. In this paper we show the equivalence of the list based diversity metrics Hamming and Mean Inter-List diversity to the aggregate diversity metric measured with the Herfindahl index, providing a formulation that allows to compute and optimize them easily.

It has been long known that quantum computing has the potential to revolutionize the way we find solutions of problems that are difficult to solve on classical computers. It was only recently that small but functional quantum computers have become available on the cloud, allowing to test their potential. In this paper we propose to leverage their capabilities to address an important task for recommender systems providers, the optimal selection of recommendation carousels. In many video-on-demand and music streaming services the user is provided with a homepage containing several recommendation lists, i.e., carousels, each built with a certain criteria (e.g., artist, mood, Action movies etc.). Choosing which set of carousels to display is a difficult problem because it needs to account for how the different recommendation lists interact, e.g., avoiding duplicate recommendations, and how they help the user explore the catalogue. We focus in particular on the adiabatic computing paradigm and use the D-Wave quantum annealer, which is able to solve NP-hard optimization problems, can be programmed by classical operations research tools and is freely available on the cloud. We propose a formulation of the carousel selection problem for black box recommenders, that can be solved effectively on a quantum annealer and has the advantage of being simple. We discuss its effectiveness, limitations and possible future directions of development.

It is common for video-on-demand and music streaming services to adopt a user interface composed of several recommendation lists, i.e., widgets or swipeable carousels, each generated according to a specific criterion or algorithm (e.g., most recent, top popular, recommended for you, editors’ choice, etc.). Selecting the appropriate combination of carousel has significant impact on user satisfaction. A crucial aspect of this user interface is that to measure the relevance a new carousel for the user it is not sufficient to account solely for its individual quality. Instead, it should be considered that other carousels will already be present in the interface. This is not considered by traditional evaluation protocols for recommenders systems, in which each carousel is evaluated in isolation, regardless of (i) which other carousels are displayed to the user and (ii) the relative position of the carousel with respect to other carousels. Hence, we propose a two-dimensional evaluation protocol for a carousel setting that will measure the quality of a recommendation carousel based on how much it improves upon the quality of an already available set of carousels. Our evaluation protocol takes into account also the position bias, i.e., users do not explore the carousels sequentially, but rather concentrate on the top-left corner of the screen.

The design of algorithms that generate personalized ranked item lists is a central topic of research in the field of recommender systems. In the past few years, in particular, approaches based on deep learning (neural) techniques have become dominant in the literature. For all of them, substantial progress over the state-of-the-art is claimed. However, indications exist of certain problems in today’s research practice, e.g., with respect to the choice and optimization of the baselines used for comparison, raising questions about the published claims. To obtain a better understanding of the actual progress, we have compared recent results in the area of neural recommendation approaches based on collaborative filtering against a consistent set of existing simple baselines. The worrying outcome of the analysis of these recent works—all were published at prestigious scientific conferences between 2015 and 2018—is that 11 of the 12 reproducible neural approaches can be outperformed by conceptually simple methods, e.g., based on the nearest-neighbor heuristic or linear models. None of the computationally complex neural methods was actually consistently better than already existing learning-based techniques, e.g., using matrix factorization or linear models. In our analysis, we discuss common issues in today’s research practice, which, despite the many papers that are published on the topic, have apparently led the field to a certain level of stagnation. 1

In recent years, algorithm research in the area of recommender systems has shifted from matrix factorization techniques and their latent factor models to neural approaches. However, given the proven power of latent factor models, some newer neural approaches incorporate them within more complex network architectures. One specific idea, recently put forward by several researchers, is to consider potential correlations between the latent factors, i.e., embeddings, by applying convolutions over the user-item interaction map. However, contrary to what is claimed in these articles, such interaction maps do not share the properties of images where Convolutional Neural Networks (CNNs) are particularly useful. In this work, we show through analytical considerations and empirical evaluations that the claimed gains reported in the literature cannot be attributed to the ability of CNNs to model embedding correlations, as argued in the original papers. Moreover, additional performance evaluations show that all of the examined recent CNN-based models are outperformed by existing non-neural machine learning techniques or traditional nearest-neighbor approaches. On a more general level, our work points to major methodological issues in recommender systems research.

In this article, we introduce the \dataset dataset, a collection of implicit interactions and impressions of movies and TV series from an Over-The-Top media service, which delivers its media contents over the Internet. The dataset is distinguished from other already available multimedia recommendation datasets by the availability of impressions, \idest the recommendations shown to the user, its size, and by being open-source. We describe the data collection process, the preprocessing applied, its characteristics, and statistics when compared to other commonly used datasets. We also highlight several possible use cases and research questions that can benefit from the availability of user impressions in an open-source dataset. Furthermore, we release software tools to load and split the data, as well as examples of how to use both user interactions and impressions in several common recommendation algorithms.

The development of continuously improved machine learning algorithms for personalized item ranking lies at the core of today's research in the area of recommender systems. Over the years, the research community has developed widely-agreed best practices for comparing algorithms and demonstrating progress with offline experiments. Unfortunately, we find this accepted research practice can easily lead to phantom progress due to the following reasons: limited reproducibility, comparison with complex but weak and non-optimized baseline algorithms, over-generalization from a small set of experimental configurations. To assess the extent of such problems, we analyzed 18 research papers published recently at top-ranked conferences. Only 7 were reproducible with reasonable effort, and 6 of them could often be outperformed by relatively simple heuristic methods, e.g., nearest neighbors. In this paper, we discuss these observations in detail, and reflect on the related fundamental problem of over-reliance on offline experiments in recommender systems research.

Deep learning techniques have become the method of choice for researchers working on algorithmic aspects of recommender systems. With the strongly increased interest in machine learning in general, it has, as a result, become difficult to keep track of what represents the state-of-the-art at the moment, e.g., for top-n recommendation tasks. At the same time, several recent publications point out problems in today's research practice in applied machine learning, e.g., in terms of the reproducibility of the results or the choice of the baselines when proposing new models.

This paper focuses on recommender systems based on item-item collaborative filtering (CF). Although research on item-based methods is not new, current literature does not provide any reliable insight on how to estimate confidence of recommendations. The goal of this paper is to fill this gap, by investigating the conditions under which item-based recommendations will succeed or fail for a specific user. We formalize the item-based CF problem as an eigenvalue problem, where estimated ratings are equivalent to the true (unknown) ratings multiplied by a user-specific eigenvalue of the similarity matrix. We show that the magnitude of the eigenvalue related to a user is proportional to the accuracy of recommendations for that user. We define a confidence parameter called the eigenvalue confidence index, analogous to the eigenvalue of the similarity matrix, but simpler to be computed. We also show how to extend the eigenvalue confidence index to matrix-factorization algorithms. A comprehensive set of experiments on five datasets show that the eigenvalue confidence index is effective in predicting, for each user, the quality of recommendations. On average, our confidence index is 3 times more correlated with MAP with respect to previous confidence estimates.

As of today, most movie recommendation services base their recommendations on collaborative filtering (CF) and/or content-based filtering (CBF) models that use metadata (e.g., genre or cast). In most video-on-demand and streaming services, however, new movies and TV series are continuously added. CF models are unable to make predictions in such a scenario, since the newly added videos lack interactions—a problem technically known as new item cold start (CS). Currently, the most common approach to this problem is to switch to a purely CBF method, usually by exploiting textual metadata. This approach is known to have lower accuracy than CF because it ignores useful collaborative information and relies on human-generated textual metadata, which are expensive to collect and often prone to errors. User-generated content, such as tags, can also be rare or absent in CS situations. In this paper, we introduce a new movie recommender system that addresses the new item problem in the movie domain by (i) integrating state-of-the-art audio and visual descriptors, which can be automatically extracted from video content and constitute what we call the movie genome ; (ii) exploiting an effective data fusion method named canonical correlation analysis , which was successfully tested in our previous works Deldjoo et al. (in: International Conference on Electronic Commerce and Web Technologies. Springer, Berlin, pp 34–45, 2016b ; Proceedings of the Twelfth ACM Conference on Recommender Systems. ACM, 2018b ), to better exploit complementary information between different modalities; (iii) proposing a two-step hybrid approach which trains a CF model on warm items (items with interactions) and leverages the learned model on the movie genome to recommend cold items (items without interactions). Experimental validation is carried out using a system-centric study on a large-scale, real-world movie recommendation dataset both in an absolute cold start and in a cold to warm transition; and a user-centric online experiment measuring different subjective aspects, such as satisfaction and diversity. Results show the benefits of this approach compared to existing approaches.