"The nature of these products means you're going to share a lot of personal information about yourself, and of course the dating apps say that you share that information in service of finding someone," MacDonald said. But they "take more information than just what you're conscious of sharing" and then use that information for purposes that aren't going to help you find a partner.

Modern large-scale recommender systems are built upon computation-intensive infrastructure and usually suffer from a huge difference in traffic between peak and off-peak periods. In peak periods, it is challenging to perform real-time computation for each request due to the limited budget of computational resources. The recommendation with a cache is a solution to this problem, where a user-wise result cache is used to provide recommendations when the recommender system cannot afford a real-time computation. However, the cached recommendations are usually suboptimal compared to real-time computation, and it is challenging to determine the items in the cache for each user. In this paper, we provide a cache-aware reinforcement learning (CARL) method to jointly optimize the recommendation by real-time computation and by the cache. We formulate the problem as a Markov decision process with user states and a cache state, where the cache state represents whether the recommender system performs recommendations by real-time computation or by the cache. The computational load of the recommender system determines the cache state. We perform reinforcement learning based on such a model to improve user engagement over multiple requests. Moreover, we show that the cache will introduce a challenge called critic dependency, which deteriorates the performance of reinforcement learning. To tackle this challenge, we propose an eigenfunction learning (EL) method to learn independent critics for CARL. Experiments show that CARL can significantly improve the users' engagement when considering the result cache. CARL has been fully launched in Kwai app, serving over 100 million users.

Voice is a natural mode of expression offered by modern computer-based systems. Qualitative perspectives on voice-based user experiences (voice UX) offer rich descriptions of complex interactions that numbers alone cannot fully represent. We conducted a systematic review of the literature on qualitative approaches to voice UX, capturing the nature of this body of work in a systematic map and offering a qualitative synthesis of findings. We highlight the benefits of qualitative methods for voice UX research, identify opportunities for increasing rigour in methods and outcomes, and distill patterns of experience across a diversity of devices and modes of qualitative praxis.

Recommender systems have become an integral part of online services to help users locate specific information in a sea of data. However, existing studies show that some recommender systems are vulnerable to poisoning attacks, particularly those that involve learning schemes. A poisoning attack is where an adversary injects carefully crafted data into the process of training a model, with the goal of manipulating the system's final recommendations. Based on recent advancements in artificial intelligence, such attacks have gained importance recently. While numerous countermeasures to poisoning attacks have been developed, they have not yet been systematically linked to the properties of the attacks. Consequently, assessing the respective risks and potential success of mitigation strategies is difficult, if not impossible. This survey aims to fill this gap by primarily focusing on poisoning attacks and their countermeasures. This is in contrast to prior surveys that mainly focus on attacks and their detection methods. Through an exhaustive literature review, we provide a novel taxonomy for poisoning attacks, formalise its dimensions, and accordingly organise 30+ attacks described in the literature. Further, we review 40+ countermeasures to detect and/or prevent poisoning attacks, evaluating their effectiveness against specific types of attacks. This comprehensive survey should serve as a point of reference for protecting recommender systems against poisoning attacks. The article concludes with a discussion on open issues in the field and impactful directions for future research. A rich repository of resources associated with poisoning attacks is available at https://github.com/tamlhp/awesome-recsys-poisoning.

When it comes to a personalized item recommendation system, It is essential to extract users' preferences and purchasing patterns. Assuming that users in the real world form a cluster and there is common favoritism in each cluster, in this work, we introduce Co-Clustering Wrapper (CCW). We compute co-clusters of users and items with co-clustering algorithms and add CF subnetworks for each cluster to extract the in-group favoritism. Combining the features from the networks, we obtain rich and unified information about users. We experimented real world datasets considering two aspects: Finding the number of groups divided according to in-group preference, and measuring the quantity of improvement of the performance.

Eury Jones was 58 the first time she pegged a man. She was coming out of a long-term relationship and entering a new life chapter. "I did a bit of Googling about threesomes and chanced upon Feeld," she says of the dating app for free-spirited relationship structures. But Jones, who lives in London and has a generous smile, wasn't accustomed to the circus of online dating: maneuvering fake profiles, encountering people who only want to swap photos and never meet. Gradually, she says, "I knew not to entertain that nonsense." Then she met him--a cyclist in a Che Guevara T-shirt.

A series of graph filtering (GF)-based collaborative filtering (CF) showcases state-of-the-art performance on the recommendation accuracy by using a low-pass filter (LPF) without a training process. However, conventional GF-based CF approaches mostly perform matrix decomposition on the item-item similarity graph to realize the ideal LPF, which results in a non-trivial computational cost and thus makes them less practical in scenarios where rapid recommendations are essential. In this paper, we propose Turbo-CF, a GF-based CF method that is both training-free and matrix decomposition-free. Turbo-CF employs a polynomial graph filter to circumvent the issue of expensive matrix decompositions, enabling us to make full use of modern computer hardware components (i.e., GPU). Specifically, Turbo-CF first constructs an item-item similarity graph whose edge weights are effectively regulated. Then, our own polynomial LPFs are designed to retain only low-frequency signals without explicit matrix decompositions. We demonstrate that Turbo-CF is extremely fast yet accurate, achieving a runtime of less than 1 second on real-world benchmark datasets while achieving recommendation accuracies comparable to best competitors.

A recent study has shown that diffusion models are well-suited for modeling the generative process of user-item interactions in recommender systems due to their denoising nature. However, existing diffusion model-based recommender systems do not explicitly leverage high-order connectivities that contain crucial collaborative signals for accurate recommendations. Addressing this gap, we propose CF-Diff, a new diffusion model-based collaborative filtering (CF) method, which is capable of making full use of collaborative signals along with multi-hop neighbors. Specifically, the forward-diffusion process adds random noise to user-item interactions, while the reverse-denoising process accommodates our own learning model, named cross-attention-guided multi-hop autoencoder (CAM-AE), to gradually recover the original user-item interactions. CAM-AE consists of two core modules: 1) the attention-aided AE module, responsible for precisely learning latent representations of user-item interactions while preserving the model's complexity at manageable levels, and 2) the multi-hop cross-attention module, which judiciously harnesses high-order connectivity information to capture enhanced collaborative signals. Through comprehensive experiments on three real-world datasets, we demonstrate that CF-Diff is (a) Superior: outperforming benchmark recommendation methods, achieving remarkable gains up to 7.29% compared to the best competitor, (b) Theoretically-validated: reducing computations while ensuring that the embeddings generated by our model closely approximate those from the original cross-attention, and (c) Scalable: proving the computational efficiency that scales linearly with the number of users or items.

Cold-start item recommendation is a long-standing challenge in recommendation systems. A common remedy is to use a content-based approach, but rich information from raw contents in various forms has not been fully utilized. In this paper, we propose a domain/data-agnostic item representation learning framework for cold-start recommendations, naturally equipped with multimodal alignment among various features by adopting a Transformer-based architecture. Our proposed model is end-to-end trainable completely free from classification labels, not just costly to collect but suboptimal for recommendation-purpose representation learning. From extensive experiments on real-world movie and news recommendation benchmarks, we verify that our approach better preserves fine-grained user taste than state-of-the-art baselines, universally applicable to multiple domains at large scale.

Watching movies is one of the social activities typically done in groups. Emotion is the most vital factor that affects movie viewers' preferences. So, the emotional aspect of the movie needs to be determined and analyzed for further recommendations. It can be challenging to choose a movie that appeals to the emotions of a diverse group. Reaching an agreement for a group can be difficult due to the various genres and choices. This paper proposes a novel approach to group movie suggestions by examining emotions from three different channels: movie descriptions (text), soundtracks (audio), and posters (image). We employ the Jaccard similarity index to match each participant's emotional preferences to prospective movie choices, followed by a fuzzy inference technique to determine group consensus. We use a weighted integration process for the fusion of emotion scores from diverse data types. Then, group movie recommendation is based on prevailing emotions and viewers' best-loved movies. After determining the recommendations, the group's consensus level is calculated using a fuzzy inference system, taking participants' feedback as input. Participants (n=130) in the survey were provided with different emotion categories and asked to select the emotions best suited for particular movies (n=12). Comparison results between predicted and actual scores demonstrate the efficiency of using emotion detection for this problem (Jaccard similarity index = 0.76). We explored the relationship between induced emotions and movie popularity as an additional experiment, analyzing emotion distribution in 100 popular movies from the TMDB database. Such systems can potentially improve the accuracy of movie recommendation systems and achieve a high level of consensus among participants with diverse preferences.