What am I not Seeing? An Interactive Approach to Social Content Discovery in Microblogs

Kang, Byungkyu; Tintarev, Nava; Höllerer, Tobias; O’Donovan, John

doi:10.1007/978-3-319-47874-6_20

Cited by 8 publications

(7 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this regard, the work of [9] is pertinent, showing how visualisation could increase user awareness of the filter bubble, understandability of the filtering mechanism, and a user's sense of control over their data stream. In another study users were able to control which people in their immediate and extended network contributed to their information feed on Twitter, and the findings suggest that the interface increased users' sense of transparency and control [8].…”

Section: Related Workmentioning

confidence: 98%

“…This work addresses the issue of filter bubbles by helping users understand not only why a recommendation was made, but also convey the limits of this recommendation. Previous work on answering the why question has led to considerable recent research on explaining recommendations (e.g., [3,8,11]), but the issue of framing the limits of a recommendation is relatively under-examined.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Knowing the unknown

Tintarev

Rostami

Smyth

2018

Proceedings of the 33rd Annual ACM Symposium on Applied Computing

Self Cite

View full text Add to dashboard Cite

In this paper we consider how to help users to better understand their consumption profiles by examining two approaches to visualising user profileschord diagrams, and bar charts-aimed at revealing to users those regions of the recommendation space that are unknown to them, i.e. blind-spots. Both visualisations do this by connecting profile preferences with a filtered recommendation space. We compare and contrast the two visualisations in a live user study (n = 70). The results suggest that, although users can understand both visualisations, chord diagrams are particularly effective in helping users to identify blind-spots, while simpler bar charts are better for conveying what was already known in a profile. Evaluating the understandability of blind-spot visualizations is a first step toward using visual explanations to help address a criticism of recommender systems: that personalising information creates filter bubbles. CCS CONCEPTS • Information systems → Decision support systems; • Humancentered computing → Human computer interaction (HCI);

show abstract

Section: Related Workmentioning

confidence: 98%

Section: Related Workmentioning

confidence: 99%

Knowing the unknown

Tintarev

Rostami

Smyth

2018

Proceedings of the 33rd Annual ACM Symposium on Applied Computing

Self Cite

View full text Add to dashboard Cite

show abstract

“…Webster and Vassileva (2007) propose a news recommender system with a twist: it is accompanied by a visual map of yourself in relation to other users in your social network, whom you can place closer or farther away from yourself to change their own influence on what you see, as well as take a glimpse into the kinds of recommendations they will be getting. Kang et al (2016) propose a similar system for Twitter users to explore content 'popular just beyond a user's typical information horizon' . Here, the focus is less on 'why did I get this recommendation?'…”

Section: Passive Interventions: Profiles In Perspectivementioning

confidence: 99%

Smart Technologies and Our Sense of Self: Going Beyond Epistemic Counter-Profiling

Delacroix

Veale

2019

Preprint

View full text Add to dashboard Cite

This chapter focuses on the extent to which sophisticated profiling techniques may end up undermining, rather than enhancing, our capacity for ethical agency. This capacity demands both opacity respect—preserving a gap between the self we present and the self we conceal—and an ability to call into question practices that are ethically wanting. Pushed to its limit, the smooth optimisation of our environment may prevent us from experiencing many of the tensions that otherwise prompt us to reconsider accepted practices. An optimally personalised world may not ever call for any ‘action’ as Hannah Arendt describes it.Can systems be designed to personalise responsibly? Greater time and research needs to be invested in designing a range of viable ‘perspective widening’ tools, as many such tools either burden users with little guarantee of meaningful engagement, or underestimate the extent to which individuals’ preferences are themselves malleable. Any approach that tries to predict what users might like, or what might change their views, risks the same pitfalls as any other form of personalisation. Instead, we argue that the most promising avenue is to push for diverse uses of newly developed systems, and measure those systems’ success at least partly on that basis. Inviting appropriation and repurposing would help keep users engaged in systems of data collection and profiling. This will not be a straightforward task: sometimes it will be in tension with traditional measures of success and performance. Yet the increasing integration of algorithmic systems in society requires us to widen our understanding of agency beyond a narrow, decontextualised focus on passive consumption preferences.

show abstract

“…We will also continue to build on our previous work on explanation interfaces that used weak ties to support content discovery [14,23], to study the role of item positions in relation to perceptions of diversity. By defining diversity in a way that is understandable and acceptable to users, it becomes possible to move research on explanation-aware recommendation to the next level: how we present diverse items in recommender systems can help users not only to understand the recommendations, but also themselves and their own biases.…”

Section: Discussionmentioning

confidence: 99%

“…The first approach is to help users to better understand the recommended items relative to a wider set of candidate items. Taking this approach, we have found that helping users control which people contributed to their information feed on Twitter increased their sense of transparency and control [14,23]. However, we also found that users had a poor mental model for the degree of novel content discovered when presented with non-personalized tweets, and thus potentially more challenging, information.…”

Section: Related Workmentioning

confidence: 99%

Presenting Diversity Aware Recommendations: Making Challenging News Acceptable

Tintarev¹

Boise State ScholarWorks

View full text Add to dashboard Cite

Recommender systems find relevant content for us online, including the personalized news we increasingly receive on Twitter and Facebook. As a consequence of personalization, we increasingly see content that agrees with our views, we cease to be exposed to views contrary to our own. Both algorithms and the users themselves filter content, and this creates more polarized points of view, so called "filter bubbles" or "echo chambers". This paper presents a vision of a diversity aware recommendation model, for the selection and presentation of a diverse selection of news to users. This diversity aware recommendation model considers that: a) users have different requirements on diversity (e.g., challenge-averse or diversity seeking), and that b) items will satisfy these requirements to different extents (e.g., liberal or conservative news). By considering both item and user diversity this model aims to maximize the amount of diverse content that users are exposed to, without damaging system reputation.

show abstract

What am I not Seeing? An Interactive Approach to Social Content Discovery in Microblogs

Cited by 8 publications

References 23 publications

Knowing the unknown

Knowing the unknown

Smart Technologies and Our Sense of Self: Going Beyond Epistemic Counter-Profiling

Presenting Diversity Aware Recommendations: Making Challenging News Acceptable

Contact Info

Product

Resources

About