Recommender systems have become a ubiquitous part of modern web applications. They help users discover new and relevant items. Today's users, through years of interaction with these systems have developed an inherent understanding of how recommender systems function, what their objectives are, and how the user might manipulate them. We describe this understanding as the Theory of the Recommender. In this study, we conducted semi-structured interviews with forty recommender system users to empirically explore the relevant factors influencing user behavior. Our findings, based on a rigorous thematic analysis of the collected data, suggest that users possess an intuitive and sophisticated understanding of the recommender system's behavior. We also found that users, based upon their understanding, attitude, and intentions change their interactions to evoke desired recommender behavior. Finally, we discuss the potential implications of such user behavior on recommendation performance.
As a highly data-driven application, recommender systems could be affected by data bias, resulting in unfair results for different data groups, which could be a reason that affects the system performance. Therefore, it is important to identify and solve the unfairness issues in recommendation scenarios. In this paper, we address the unfairness problem in recommender systems from the user perspective. We group users into advantaged and disadvantaged groups according to their level of activity, and conduct experiments to show that current recommender systems will behave unfairly between two groups of users. Specifically, the advantaged users (active) who only account for a small proportion in data enjoy much higher recommendation quality than those disadvantaged users (inactive). Such bias can also affect the overall performance since the disadvantaged users are the majority. To solve this problem, we provide a re-ranking approach to mitigate this unfairness problem by adding constraints over evaluation metrics. The experiments we conducted on several real-world datasets with various recommendation algorithms show that our approach can not only improve group fairness of users in recommender systems, but also achieve better overall recommendation performance.
To solve the information explosion problem and enhance user experience in various online applications, recommender systems have been developed to model users preferences. Although numerous efforts have been made toward more personalized recommendations, recommender systems still suffer from several challenges, such as data sparsity and cold start. In recent years, generating recommendations with the knowledge graph as side information has attracted considerable interest. Such an approach can not only alleviate the abovementioned issues for a more accurate recommendation, but also provide explanations for recommended items. In this paper, we conduct a systematical survey of knowledge graph-based recommender systems. We collect recently published papers in this field and summarize them from two perspectives. On the one hand, we investigate the proposed algorithms by focusing on how the papers utilize the knowledge graph for accurate and explainable recommendation. On the other hand, we introduce datasets used in these works. Finally, we propose several potential research directions in this field.
In social media, recommender systems are responsible for directing the users to relevant content. In order to enhance the users' engagement, recommender systems adapt their output to the expected reactions of the users, which are in turn affected by the recommended contents. In this work, we model a single user that interacts with an online news aggregator, with the purpose of making explicit the feedback loop between the evolution of the user's opinion and the personalised recommendation of contents. We assume that the user has a scalar opinion on a certain issue: this opinion is influenced by all received news, which are characterized by a binary position on the issue at hand. The user has a confirmation bias, that is, a preference for news that confirm her current opinion. At the same time, we assume that the recommender has the goal of maximizing the number of user's clicks (as a measure of her engagement): in order to fulfil its goal, the recommender has to compromise between exploring the user's preferences and exploiting them. After defining suitable metrics for the effectiveness of the recommender systems and for its impact on the opinion, we perform both extensive numerical simulations and a mathematical analysis of the model. We find that personalised contents and confirmation bias do affect the evolution of opinions: the extent of these effects is inherently related to the effectiveness of the recommender. We also show that by tuning the amount of randomness in the recommendation algorithm, one can reduce the impact of the recommendation system on the opinions.
With the global metamorphosis of the beauty industry and the rising demand for beauty products worldwide, the need for an efficacious makeup recommendation system has never been more. Despite the significant advancements made towards personalised makeup recommendation, the current research still falls short of incorporating the context of occasion in makeup recommendation and integrating feedback for users. In this work, we propose BeautifAI, a novel makeup recommendation system, delivering personalised occasion-oriented makeup recommendations to users while providing real-time previews and continuous feedback. The proposed work's novel contributions, including the incorporation of occasion context, region-wise makeup recommendation, real-time makeup previews and continuous makeup feedback, set our system apart from the current work in makeup recommendation. We also demonstrate our proposed system's efficacy in providing personalised makeup recommendation by conducting a user study.
Recommendations Systems allow users to identify trending items among a community while being timely and relevant to the user's expectations. When the purpose of various Recommendation Systems differs, the required type of recommendations also differs for each use case. While one Recommendation System may focus on recommending popular items, another may focus on recommending items that are comparable to the user's interests. Content-based filtering, user-to-user & item-to-item Collaborative filtering, and more recently; Deep Learning methods have been brought forward by the researchers to achieve better quality recommendations. Even though each of these methods has proven to perform well individually, there have been attempts to push the boundaries of their limitations. Following a wide range of methods, researchers have tried to expand on the capabilities of standard recommendation systems to provide the most effective recommendations to users while being more profitable from a business's perspective. This has been achieved by taking a hybrid approach when building models and architectures for Recommendation Systems. This paper is a review of the novel models & architectures of hybrid Recommendation Systems. The author identifies possibilities of expanding the capabilities of baseline models & the advantages and drawbacks of each model with selected use cases in this review.
The task of item recommendation is to select the best items for a user from a large catalogue of items. Item recommenders are commonly trained from implicit feedback which consists of past actions that are positive only. Core challenges of item recommendation are (1) how to formulate a training objective from implicit feedback and (2) how to efficiently train models over a large item catalogue. This article provides an overview of item recommendation, its unique characteristics and some common approaches. It starts with an introduction to the problem and discusses different training objectives. The main body deals with learning algorithms and presents sampling based algorithms for general recommenders and more efficient algorithms for dot product models. Finally, the application of item recommenders for retrieval tasks is discussed.
Recent years have witnessed the progress of sequential recommendation in accurately predicting users' future behaviors. However, only persuading accuracy leads to the risk of filter bubbles where recommenders only focus on users' main interest areas. Different from other studies which improve diversity or coverage, we investigate the calibration in sequential recommendation. However, existing calibrated methods followed a post-processing paradigm, which costs more computation time and sacrifices the recommendation accuracy. To this end, we propose an end-to-end framework to provide both accurate and calibrated recommendations. We propose a loss function to measure the divergence of distributions between recommendation lists and historical behaviors for sequential recommendation framework. In addition, we design a dual-aggregation model which extracts information from two individual sequence encoders with different objectives to further improve the recommendation. Experiments on two benchmark datasets demonstrate the effectiveness and efficiency of our model.
Recommender systems (RSs) are software tools and algorithms developed to alleviate the problem of information overload, which makes it difficult for a user to make right decisions. Two main paradigms toward the recommendation problem are collaborative filtering and content-based filtering, which try to recommend the best items using ratings and content available. These methods typically face infamous problems including cold-start, diversity, scalability, and great computational expense. We argue that the uptake of deep learning and reinforcement learning methods is also questionable due to their computational complexities and uninterpretability. In this paper, we approach the recommendation problem from a new prospective. We borrow ideas from cluster head selection algorithms in wireless sensor networks and adapt them to the recommendation problem. In particular, we propose Load Balanced Recommender System (LBRS), which uses a probabilistic scheme for item recommendation. Furthermore, we factor in the importance of items in the recommendation process, which significantly improves the recommendation accuracy. We also introduce a method that considers a heterogeneity among items, in order to balance the similarity and diversity trade-off. Finally, we propose a new metric for diversity, which emphasizes the importance of diversity not only from an intra-list perspective, but also from a between-list point of view. With experiments in a simulation study performed on RecSim, we show that LBRS is effective and can outperform baseline methods.
As Recommender Systems (RS) influence more and more people in their daily life, the issue of fairness in recommendation is becoming more and more important. Most of the prior approaches to fairness-aware recommendation have been situated in a static or one-shot setting, where the protected groups of items are fixed, and the model provides a one-time fairness solution based on fairness-constrained optimization. This fails to consider the dynamic nature of the recommender systems, where attributes such as item popularity may change over time due to the recommendation policy and user engagement. For example, products that were once popular may become no longer popular, and vice versa. As a result, the system that aims to maintain long-term fairness on the item exposure in different popularity groups must accommodate this change in a timely fashion. Novel to this work, we explore the problem of long-term fairness in recommendation and accomplish the problem through dynamic fairness learning. We focus on the fairness of exposure of items in different groups, while the division of the groups is based on item popularity, which dynamically changes over time in the recommendation process. We tackle this problem by proposing a fairness-constrained reinforcement learning algorithm for recommendation, which models the recommendation problem as a Constrained Markov Decision Process (CMDP), so that the model can dynamically adjust its recommendation policy to make sure the fairness requirement is always satisfied when the environment changes. Experiments on several real-world datasets verify our framework's superiority in terms of recommendation performance, short-term fairness, and long-term fairness.