Deanonymizing Social Networks Using Structural Information

Authors: Ioannis Caragiannis, Evanthia Tsitsoka

IJCAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results show that by combining the two algorithms, we can tolerate noise at the level of up to 10% and deanonymize correctly large fractions of the network nodes.
Researcher Affiliation Academia Ioannis Caragiannis and Evanthia Tsitsoka University of Patras, Greece {caragian,tsitsoka}@ceid.upatras.gr
Pseudocode Yes Algorithm 1 The local search algorithm
Open Source Code No The paper states 'We have implemented in C both the NDSD and the local search algorithm and have used them in all our experiments.' but does not provide a link or explicit statement about public code availability.
Open Datasets Yes The real-world instances include the ego-facebook graph from the SNAP collection of social networks [Leskovec and Krevl, 2014] as well as sixteen facebook graphs from the Network Repository [Rossi and Ahmed, 2015].
Dataset Splits No The paper describes generating noisy graphs (H) from original graphs (G) using parameters δ and ϵ to simulate real-world conditions for testing, but it does not specify traditional train/validation/test dataset splits, percentages, or cross-validation strategies needed for reproducibility of data partitioning.
Hardware Specification Yes Our computational results (including what we report here as well as additional ones that we omit due to lack of space) have been obtained using a desktop PC with an Intel i7-4790/3.6GHz processor with 8GB of RAM, running a Slackware64 operating system.
Software Dependencies No The paper mentions implementation in C, use of NetworkX, and 'running a Slackware64 operating system', but it does not provide specific version numbers for these software components.
Experiment Setup Yes In particular, we build H from G using two parameters, δ and ϵ. The parameter δ indicates the fraction of nodes that are removed from graph G together with their incident edges. Next, we remove an ϵ fraction of the edges that survived and replace each of them by a random (new) edge in the graph. In our experiments with real-world data from the Network Repository, we have used different values between 0 and 20% for δ and between 0 and 10% for ϵ.