530 likes | 557 Views
This research explores a new approach for propagating trust in mobile reputation systems to combat content overload and accept content only from reputable sources. The proposed method involves building a trust graph and defining a function that rates the trust relationships between nodes.
E N D
Lightweight DistributedTrust Propagation(daniele quercia & stephen hailes & licia capra) UCL
Daniele Quercia Situation: People exchange digital content
Daniele Quercia Problem: Content overload
Daniele Quercia Proposal:Accept content onlyfrom reputable people
Daniele Quercia Take two devices… A B
Daniele Quercia Shall I accept content from B? A B
Daniele Quercia A should set its initial trust for B A B
Daniele Quercia A should set its initial trust for B A B Yes, but how?
Daniele Quercia • Traditional way: • Trust propagation C ? A B
Daniele Quercia • That way works on the Web and has been tested on “binary” ratings
Daniele Quercia • If that way is tested on • 3-level ratings
Conclusion:A needs a new way for propagating its trust in B
Conclusion:A needs a new way for propagating its trust in B But which way?
Daniele Quercia AB AC CB C 1 2 ? A B
Daniele Quercia AB AC CB 1 2 Put trust relationships in a new graph & label graph nodes ? C 1 2 ? A B
Daniele Quercia AB AC CB 1 2 Define a “good” function that rates AB Put trust relationships in a new graph & label graph nodes ? C 1 f 2 ? A B
Daniele Quercia AB AC CB 1 2 find_f build_graph(trust_relationships, AB) ? C 1 f 2 ? A B
? A B 1 A B 1 0.5 1 2 3 ? ? 2 D C M M 1 2 3
? A B 1 A B 1 0.5 1 2 3 ? ? 2 D C M M 1 2 3 What are those ?
Remember, we’ll define f We will see that frates neighbouring nodes alike It’s a neighbour of AB; therefore, it rates as AB does ?
Remember, we’ll define f We will see that frates neighbouring nodes alike It’s a neighbour of AB; therefore, it rates as AB does ? Find relationships that rate as AB does
Remember, we’ll define f We will see that frates neighbouring nodes alike It’s a neighbour of AB; therefore, it rates as AB does ?
2 Types of those relationships
1. The relationships with same rater (A) A B 1 2 3 2 D C 1 As long as B and Dbehave alike
2. The relationships with same rated node (B) A B 1 2 3 2 D C 1 As long as A and Crate alike
OK, 2 Types
Taking those relationships, one comes up with …
…a graph! A C 1 A B 0.5 1 A D C B M M 2 3
A C 1 A B 0.5 1 A D C B M M 2 3
Take f that returns • the ratings already there ( ) • & similar ratings for neighbouring nodes A C 1 A B 0.5 1 A D C B M M 2 3 2 3
Daniele Quercia AB AC CB 1 2 find_f build_graph ? C 1 f 2 ? A B
Does it work? Useful, Robust, Fast, “Light”?
Daniele Quercia • Useful? Tested on real data • (Advogato: > 55K user ratings)
Daniele Quercia • Useful? Tested on real data • (Advogato: > 55K user ratings)
Daniele Quercia Robust?
Daniele Quercia Robust?
Daniele Quercia Fast and “Light”?
Daniele Quercia Fast and “Light”? • For propagating AB • (worst case) • Transmit 30KB • & run for 2.8ms