We study estimator selection and hyper-parameter tuning in off-policy evaluation. Although cross-validation is the most popular method for model selection in supervised learning, off-policy evaluation relies mostly on theory, which provides only limited guidance to practitioners. We show how to use cross-validation for off-policy evaluation. This challenges a popular belief that cross-validation in off-policy evaluation is not feasible. We evaluate our method empirically and show that it addresses a variety of use cases.

Off-policy learning is a framework for optimizing policies without deploying them, using data collected by another policy. In recommender systems, this is especially challenging due to the imbalance in logged data: some items are recommended and thus logged more frequently than others. This is further perpetuated when recommending a list of items, as the action space is combinatorial. To address this challenge, we study pessimistic off-policy optimization for learning to rank. The key idea is to compute lower confidence bounds on parameters of click models and then return the list with the highest pessimistic estimate of its value. This approach is computationally efficient, and we analyze it. We study its Bayesian and frequentist variants and overcome the limitation of unknown prior by incorporating empirical Bayes. To show the empirical effectiveness of our approach, we compare it to off-policy optimizers that use inverse propensity scores or neglect uncertainty. Our approach outperforms all baselines and is both robust and general.

In this article, we present results of an auditing study performed over YouTube aimed at investigating how fast a user can get into a misinformation filter bubble, but also what it takes to “burst the bubble,” i.e., revert the bubble enclosure. We employ a sock puppet audit methodology, in which pre-programmed agents (acting as YouTube users) delve into misinformation filter bubbles by watching misinformation-promoting content. Then they try to burst the bubbles and reach more balanced recommendations by watching misinformation-debunking content. We record search results, home page results, and recommendations for the watched videos. Overall, we recorded 17,405 unique videos, out of which we manually annotated 2,914 for the presence of misinformation. The labeled data was used to train a machine learning model classifying videos into three classes (promoting, debunking, neutral) with the accuracy of 0.82. We use the trained model to classify the remaining videos that would not be feasible to annotate manually. Using both the manually and automatically annotated data, we observe the misinformation bubble dynamics for a range of audited topics. Our key finding is that even though filter bubbles do not appear in some situations, when they do, it is possible to burst them by watching misinformation-debunking content (albeit it manifests differently from topic to topic). We also observe a sudden decrease of misinformation filter bubble effect when misinformation-debunking videos are watched after misinformation-promoting videos, suggesting a strong contextuality of recommendations. Finally, when comparing our results with a previous similar study, we do not observe significant improvements in the overall quantity of recommended misinformation content.

In this paper, we describe a black-box sockpuppeting audit which we carried out to investigate the creation and bursting dynamics of misinformation filter bubbles on YouTube. Pre-programmed agents acting as YouTube users stimulated YouTube's recommender systems: they first watched a series of misinformation promoting videos (bubble creation) and then a series of misinformation debunking videos (bubble bursting). Meanwhile, agents logged videos recommended to them by YouTube. After manually annotating these recommendations, we were able to quantify the portion of misinformative videos among them. The results confirm the creation of filter bubbles (albeit not in all situations) and show that these bubbles can be bursted by watching credible content. Drawing a direct comparison with a previous study, we do not see improvements in overall quantities of misinformation recommended.

The negative effects of misinformation filter bubbles in adaptive systems have been known to researchers for some time. Several studies investigated, most prominently on YouTube, how fast a user can get into a misinformation filter bubble simply by selecting “wrong choices” from the items offered. Yet, no studies so far have investigated what it takes to “burst the bubble”, i.e., revert the bubble enclosure. We present a study in which pre-programmed agents (acting as YouTube users) delve into misinformation filter bubbles by watching misinformation promoting content (for various topics). Then, by watching misinformation debunking content, the agents try to burst the bubbles and reach more balanced recommendation mixes. We recorded the search results and recommendations, which the agents encountered, and analyzed them for the presence of misinformation. Our key finding is that bursting of a filter bubble is possible, albeit it manifests differently from topic to topic. Moreover, we observe that filter bubbles do not truly appear in some situations. We also draw a direct comparison with a previous study. Sadly, we did not find much improvements in misinformation occurrences, despite recent pledges by YouTube.

Recommender systems generate items that should be interesting for the customers. However, recommenders usually fail in the cold-start scenario - when a new item or a new customer appears. In our work, we study the cold-start problem for a new customer. For a cold-start customer we find the most similar customers and use a “their” pre-trained collaborative filtering model to recommend. We compare several recommendation approaches and similarity metrics to analyze the accuracy and computational performance.

User behaviour in data intensive applications such as the Web-based applications represents a complex set of actions influenced by plenty of factors.Thanks to this complexity, it is extremely hard for human to be able to understand all its aspects.Despite of this, by observing user actions from multiple views, we are able to extract and to model typical behaviour and its deviations on the Web.The website itself, together with transaction server logs, includes information about the site structure, content and about the actual user actions (clicks) within the site.User actions logically reflect the behaviour, while other sources indicate his/her context.Combination of these data sources allows to model the typical user behaviour and his/her preferences.The longterm behaviour describes relatively stable user preferences based on extensive user history.As the Web has become more and more dynamic, modelling user behaviour from the long-term perspective does not satisfy requirements of current Web based applications.On the other side, the short-term behaviour describes current user activity and his/her actual intent.However, this source of information is often noisy.To address these shortcomings the state-ofthe-art combines both perspectives, which allows to meaningful and timely modelling of user behaviour.In this paper, we provide a comprehensive survey