Sequential recommenders use the ordering of user-item interactions to perform next-item prediction. Several studies have attempted to estimate how much sequential information is available in data sets used for the offline evaluation of sequential recommenders by randomly shuffling users' interaction histories and breaking the sequential dependencies between interactions. However, random shuffling fails to distinguish between sequential patterns from user behaviour (users consuming items based on previous interactions) and item availability (when items enter the system and become available for user consumption).In this article, we analyse several widely used data sets in sequential recommendation studies using two shuffling techniques: random shuffling and constrained shuffling. While random shuffling reorders interactions arbitrarily, constrained shuffling does not allow user-item interactions to occur prior to the item's first appearance in the data set. Our experiments show that sequential information can either come exclusively from user behaviour patterns or item availability, or from a combination of the two. These findings have implications for understanding evaluation results in sequential recommendation and highlights why some data sets may be less appropriate for offline evaluation given how little sequential information comes from user behaviour.

Recommender systems typically operate within a single domain, for example, recommending books based on users' reading habits. If such data is unavailable, it may be possible to make cross-domain recommendations and recommend books based on user preferences from another domain, such as movies. However, despite considerable research on cross-domain recommendations, no studies have investigated their impact on users' behavioural intentions or system perceptions compared to single-domain recommendations. Similarly, while single-domain explanations have been shown to improve users' perceptions of recommendations, there are no comparable studies for the cross-domain case.

Current practices in offline evaluation use rank-based metrics to measure the quality of top-n recommendation lists. This approach has practical benefits as it centres assessment on the output of the recommender system and, therefore, measures performance from the perspective of end-users. However, this methodology neglects how recommender systems more broadly model user preferences, which is not captured by only considering the top-n recommendations. In this article, we use item response theory (IRT), a family of latent variable models used in psychometric assessment, to gain a comprehensive understanding of offline evaluation. We use IRT to jointly estimate the latent abilities of 51 recommendation algorithms and the characteristics of 3 commonly used benchmark data sets. For all data sets, the latent abilities estimated by IRT suggest that higher scores from traditional rank-based metrics do not reflect improvements in modeling user preferences. Furthermore, we show that the top-n recommendations with the most discriminatory power are biased towards lower difficulty items, leaving much room for improvement. Lastly, we highlight the role of popularity in evaluation by investigating how user engagement and item popularity influence recommendation difficulty.

Negative item sampling in offline top-n recommendation evaluation has become increasingly wide-spread, but remains controversial. While several studies have warned against using sampled evaluation metrics on the basis of being a poor approximation of the full ranking (i.e. using all negative items), others have highlighted their improved discriminative power and potential to make evaluation more robust. Unfortunately, empirical studies on negative item sampling are based on relatively few methods (between 3-12) and, therefore, lack the statistical power to assess the impact of negative item sampling in practice.

Image retrieval from generative adversarial networks (GANs) is challenging for several reasons. First, there are no clear mappings between the GAN's latent space and useful semantic features, making it difficult for users to navigate. Second, the number of unique images that can be generated is exceptionally high, taxing the scaling properties of existing search algorithms. In this article, we present ROGUE, a system to support exploratory search of images generated from GANs. We demonstrate how to implement features that are commonly found in exploratory search interfaces, such as faceted search and relevance feedback, in the context of GAN search. We additionally use reinforcement learning to help users navigate the image space [8], trading off exploration (showing diverse images) and exploitation (showing images predicted to receive positive relevance feedback). Finally, we present a usability study where participants were situated in the role of a casting director who needs to explore actors' headshots for an upcoming movie. The system obtained an average SUS score of 72.8 and all participants reported being either satisfied or very satisfied with the images they identified with the system. The system is shown in this accompanying video: https://vimeo.com/680036160.

Professional discourse is the language used by specialists, such as lawyers, doctors and academics, to communicate the knowledge and assumptions associated with their respective fields. Professional discourse can be especially difficult for non-specialists to understand due to the lexical ambiguity of commonplace words that have a different or more specific meaning within a specialist domain. This phenomena also makes it harder for specialists to communicate with the general public because they are similarly unaware of the potential for misunderstandings. In this article, we present an approach for detecting domain terms with lexical ambiguity versus everyday English. We demonstrate the efficacy of our approach with three case studies in statistics, law and biomedicine. In all case studies, we identify domain terms with a [email protected] greater than 0.9, outperforming the best performing baseline by 18.1–91.7%. Most importantly, we show this ranking is broadly consistent with semantic differences. Our results highlight the difficulties that existing semantic difference methods have in the cross-domain setting, which rank non-domain terms highly due to noise or biases in the data. We additionally show that our approach generalizes to short phrases and investigate its data efficiency by varying the number of labeled examples.