Exposure diversity as a design principle for recommender systemsHelberger, N.; Karpinnen, K.; D'Acunto, L. General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons). Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: http://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible. ABSTRACTPersonalized recommendations in search engines, social media and also in more traditional media increasingly raise concerns over potentially negative consequences for diversity and the quality of public discourse. The algorithmic filtering and adaption of online content to personal preferences and interests is often associated with a decrease in the diversity of information to which users are exposed. Notwithstanding the question of whether these claims are correct or not, this article discusses whether and how recommendations can also be designed to stimulate more diverse exposure to information and to break potential 'filter bubbles' rather than create them. Combining insights from democratic theory, computer science and law, the article makes suggestions for design principles and explores the potential and possible limits of 'diversity sensitive design'. IntroductionRecommendation systems increasingly influence our information choices: the information that is ultimately being presented to us has been filtered through the lens of our personal preferences, our previous choices and the preferences of our friends. There are also commercial and strategic decisions behind the algorithms that determine which information we will see, which information is prioritized and which information is excluded (Bozdag, 2013;Foster, 2012;Schulz, Dreyer, & Hagemeier, 2012;Webster, 2010). Search engines and social networks, in particular, increasingly rely on recommender systems, which are a class of information filtering systems that study patterns of user behaviour to determine what someone will prefer from among a collection of 'information'. By doing so, recommender systems essentially personalize the list of content that is offered to a user. With the emerging trends of higher interactivity and user orientation, the use of recommender systems is not limited to search engines and social networks, as media organizations are also increasingly incorporating recommenders into their own services.The impact of personalized recommendations on the realization of media and information diversity is currently a central questi...

The enormous popularity of Video on Demand (VoD) has attracted substantial research attention into the effective use of peer-to-peer (P2P) architectures to provide solutions at large-scale. In particular, the high efficiency of BitTorrent has inspired many P2P protocols for VoD. However, these protocols use different approaches to adapt the design of Bittorrent to VoD, and in most cases their performance has been evaluated separately and in limited scenarios. As a consequence, the research community still lacks a clear understanding of how these protocols compare against each other and how well each of them would work in real world conditions, where, for instance, peers have heterogeneous bandwidths, may freeride or may be located behind NAT/firewall.In this paper, we propose a simulation based methodology which aims at putting forward a common base for comparing the performance of these different protocols under a wide range of conditions. We show that, despite their considerable differences, (i) existing BitTorrent-like VoD approaches all share some characteristics, such as that their bandwidth reciprocity based methods to incentivize cooperation do not always yield an optimal overall performance. Furthermore, we demonstrate that (ii) in these protocols there is a trade-off between QoS and resilience to freeriding and malicious attacks. We also discover that, (iii) when peers doing streaming coexist with peers doing traditional file transfer, the latter actually benefit from this coexistence, at the expenses of the former. Finally, we show that (iv) early departures of peers from the system do not significantly affect the QoS de- livered, while jumping to a different position in the file has a bigger negative impact. Overall, our findings provide important implications for both VoD service providers and future system designers. On the one hand, our results can guide VoD service providers in selecting the most appropriate protocol for a given environment. On the other hand, exposing the flaws of current approaches will help researchers in improving them and/or designing better ones.

The efficiency of BitTorrent in disseminating content has inspired a number of P2P protocols for on-demand video streaming (VoD). Prior work on adapting BitTorrent to VoD mainly focused on the piece selection policy, since streaming requires a somewhat "in order" download progress. Conversely, not much effort has been spent into adapting BitTorrent's peer selection policy, where nodes mainly serve those that have recently uploaded to them at the highest rates. This mechanism incentivizes cooperation among peers but, in a heterogeneous system (i.e. where peers have different bandwidth capacities), it causes faster peers to receive higher download speeds than slower peers. This might hurt the system's ability of providing as many nodes as possible with the minimum download speed necessary to sustain the video playback rate. Furthermore, peers gain little utility in downloading at rates much higher than the video playback rate. Inspired by these observations, in this work, we extend the peer selection mechanism of an existing BitTorrentlike VoD protocol, give-to-get (G2G), with techniques that allow peers to relax their reciprocity-based peer selection and choose more random nodes when their current QoS is high. In this way, more peers can be granted a good QoS and free-riding is tolerated only when bandwidth resources are abundant. To demonstrate the benefits of our approach, we present extensive simulations of the introduced techniques.

No abstract

Abstract-Enhancing reciprocity has been one of the primary motivations for the design of incentive policies in BitTorrent-like P2P systems. Reciprocity implies that peers need to contribute their bandwidth to other peers if they want to receive bandwidth in return. However, the over-provisioning that characterizes today's BitTorrent communities and the development of many next-generation P2P systems with real-time constraints (e.g., for live and on-demand streaming) suggest that more effort can be devoted to reducing the inequity (i.e., the difference of service received) among peers, rather than only enhancing reciprocity. Inspired by this observation, in this work we analyze in detail several incentive mechanisms that are used in BitTorrent systems, and explore several strategies that influence the balance between reciprocity and equity. Our study shows that (i) reducing inequity leads to a better overall system performance, and (ii) the behavior of seeders (i.e., peers that hold a complete copy of the file and upload it for free) influences whether reciprocity is enhanced or inequity reduced.