I’ve just finished to give a presentation. I talked about old stuff – TRULLO (pdf, post) and distributed trust propagation (pdf, post). So I recycled old slides – only the first 20 slides (below) were brand new Thanks to Neal and Elisa!
Situation: Using mobile devices, such as smart phones, people may create and distribute different types of digital content (e.g., photos, videos). One of the problems is that digital content, being easy to create and replicate, may likely swamp users rather than informing them. To avoid that, users may run trust models on their mobile devices. A trust model is a piece of software that keeps track of who provides quality content and who does not.
Problem: Devices should be able to set their initial trust for other devices. One way of doing so is for devices to learn from their own past experiences. To see how, consider the following quotes about human trust: “We may initially trust or not trust those involved on our projects based on past experience”, and “If your boyfriend is unfaithful, you won’t initially trust the next man you date” Algorithms that model human trust on pervasive devices, one might say, ought to do the same thing – they should assign their initial trust upon `similar’ past experiences.
Existing Solutions: Existing solutions usually require an ontology upon which they decide which past experiences are similar, and, in so doing, they require both that the same ontology is shared by all users (which is hardly the case in reality) and that users agree on that ontology for good (ie, the ontology is not supposed to change over time)
Proposal: TRULLO gathers ratings of past experiences in a matrix, learns staticial “features” from that matrix, and combines those features to set initial trust values. It works quite well in a simulated antique market and its implementation is reasonably fast on a Nokia mobile phone.
Future: TRULLO does not work if one does not have past experiences. That is why we will propose a distributed trust propadation algorithm (pdf).
Consider a community of users who share digital content through their handheld devices. If there are many such users, and no (or a very poor) means of filtering, they may suffer from content overload. To avoid this, it would be preferable if users could effectively select content that both lies within their range of interests and comes from sources that are trustworthy. Users may do so by running trust models on their devices. A trust model is a piece of software that keeps track of which devices are trusted and which are not.
This talk will look at how a trust model running on device A determines the extent to which A should initially trust device B in a given context (content category). It does so by considering two cases: in the first, A does not know B at all; in the second case, A knows B but in contexts other than that of interest. For each of those two cases, this talk will discuss the most recent proposal that improves on existing solutions (TRULLO and distributed propagation), and will also attempt to suggest new research directions (such as private collaborative filtering – post & more).