On the Troll-Trust Model for Edge Sign Prediction in Social Networks

Géraud Le Falher, Nicolo Cesa-Bianchi, Claudio Gentile, Fabio Vitale
; Proceedings of the 20th International Conference on Artificial Intelligence and Statistics, PMLR 54:402-411, 2017.

Abstract

In the problem of edge sign prediction, we are given a directed graph (representing a social network), and our task is to predict the binary labels of the edges (i.e., the positive or negative nature of the social relationships). Many successful heuristics for this problem are based on the troll-trust features, estimating at each node the fraction of outgoing and incoming positive/negative edges. We show that these heuristics can be understood, and rigorously analyzed, as approximators to the Bayes optimal classifier for a simple probabilistic model of the edge labels. We then show that the maximum likelihood estimator for this model approximately corresponds to the predictions of a Label Propagation algorithm run on a transformed version of the original social graph. Extensive experiments on a number of real-world datasets show that this algorithm is competitive against state-of-the-art classifiers in terms of both accuracy and scalability. Finally, we show that troll-trust features can also be used to derive online learning algorithms which have theoretical guarantees even when edges are adversarially labeled.

Cite this Paper


BibTeX
@InProceedings{pmlr-v54-falher17a, title = {{On the Troll-Trust Model for Edge Sign Prediction in Social Networks}}, author = {Géraud Le Falher and Nicolo Cesa-Bianchi and Claudio Gentile and Fabio Vitale}, booktitle = {Proceedings of the 20th International Conference on Artificial Intelligence and Statistics}, pages = {402--411}, year = {2017}, editor = {Aarti Singh and Jerry Zhu}, volume = {54}, series = {Proceedings of Machine Learning Research}, address = {Fort Lauderdale, FL, USA}, month = {20--22 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v54/falher17a/falher17a.pdf}, url = {http://proceedings.mlr.press/v54/falher17a.html}, abstract = {In the problem of edge sign prediction, we are given a directed graph (representing a social network), and our task is to predict the binary labels of the edges (i.e., the positive or negative nature of the social relationships). Many successful heuristics for this problem are based on the troll-trust features, estimating at each node the fraction of outgoing and incoming positive/negative edges. We show that these heuristics can be understood, and rigorously analyzed, as approximators to the Bayes optimal classifier for a simple probabilistic model of the edge labels. We then show that the maximum likelihood estimator for this model approximately corresponds to the predictions of a Label Propagation algorithm run on a transformed version of the original social graph. Extensive experiments on a number of real-world datasets show that this algorithm is competitive against state-of-the-art classifiers in terms of both accuracy and scalability. Finally, we show that troll-trust features can also be used to derive online learning algorithms which have theoretical guarantees even when edges are adversarially labeled.} }
Endnote
%0 Conference Paper %T On the Troll-Trust Model for Edge Sign Prediction in Social Networks %A Géraud Le Falher %A Nicolo Cesa-Bianchi %A Claudio Gentile %A Fabio Vitale %B Proceedings of the 20th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2017 %E Aarti Singh %E Jerry Zhu %F pmlr-v54-falher17a %I PMLR %J Proceedings of Machine Learning Research %P 402--411 %U http://proceedings.mlr.press %V 54 %W PMLR %X In the problem of edge sign prediction, we are given a directed graph (representing a social network), and our task is to predict the binary labels of the edges (i.e., the positive or negative nature of the social relationships). Many successful heuristics for this problem are based on the troll-trust features, estimating at each node the fraction of outgoing and incoming positive/negative edges. We show that these heuristics can be understood, and rigorously analyzed, as approximators to the Bayes optimal classifier for a simple probabilistic model of the edge labels. We then show that the maximum likelihood estimator for this model approximately corresponds to the predictions of a Label Propagation algorithm run on a transformed version of the original social graph. Extensive experiments on a number of real-world datasets show that this algorithm is competitive against state-of-the-art classifiers in terms of both accuracy and scalability. Finally, we show that troll-trust features can also be used to derive online learning algorithms which have theoretical guarantees even when edges are adversarially labeled.
APA
Falher, G.L., Cesa-Bianchi, N., Gentile, C. & Vitale, F.. (2017). On the Troll-Trust Model for Edge Sign Prediction in Social Networks. Proceedings of the 20th International Conference on Artificial Intelligence and Statistics, in PMLR 54:402-411

Related Material