This paper investigates the prediction of positive and negative links in online social networks, using datasets from Epinions, Slashdot, and Wikipedia. The authors propose a machine learning approach to predict the sign of an edge in a network, based on the signs of other edges. They find that their models can accurately predict the sign of an edge, even when the sign is hidden. The models are evaluated on three different datasets, and the results show that the models perform well across all three domains. The authors also compare their results to social psychological theories of balance and status, and find that their models align with these theories in some cases and differ in others. The paper also discusses the generalization of the models across different datasets, and finds that the models can be applied to a wide range of social network data. The authors conclude that their models provide insight into the formation of signed links in social networks, and that they can be used to infer the attitudes of users in online systems.This paper investigates the prediction of positive and negative links in online social networks, using datasets from Epinions, Slashdot, and Wikipedia. The authors propose a machine learning approach to predict the sign of an edge in a network, based on the signs of other edges. They find that their models can accurately predict the sign of an edge, even when the sign is hidden. The models are evaluated on three different datasets, and the results show that the models perform well across all three domains. The authors also compare their results to social psychological theories of balance and status, and find that their models align with these theories in some cases and differ in others. The paper also discusses the generalization of the models across different datasets, and finds that the models can be applied to a wide range of social network data. The authors conclude that their models provide insight into the formation of signed links in social networks, and that they can be used to infer the attitudes of users in online systems.