Node Similarity with q -Grams for Real-World Labeled Networks - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year :

Node Similarity with q -Grams for Real-World Labeled Networks


We study node similarity in labeled networks, using the label sequences found in paths of bounded length q leading to the nodes. (This recalls the q-grams employed in document resemblance, based on the Jaccard distance.) When applied to networks, the challenge is twofold: the number of q-grams generated from labeled paths grows exponentially with q, and their frequency should be taken into account: this leads to a variation of the Jaccard index known as Bray-Curtis index for multisets. We describe nSimGram, a suite of fast algorithms for node similarity with q-grams, based on a novel blend of color coding, probabilistic counting, sketches, and string algorithms, where the universe of elements to sample is exponential. We provide experimental evidence that our measure is effective and our running times scale to deal with large real-world networks.

Dates and versions

hal-01964716 , version 1 (03-01-2019)



Alessio Conte, Gaspare Ferraro, Roberto P Grossi, Andrea Marino, Kunihiko Sadakane, et al.. Node Similarity with q -Grams for Real-World Labeled Networks. KDD 2018 - ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Aug 2018, London, United Kingdom. pp.1282-1291, ⟨10.1145/3219819.3220085⟩. ⟨hal-01964716⟩


35 View
0 Download



Gmail Facebook Twitter LinkedIn More