Online Matrix Completion for Signed Link Prediction

This work studies the binary matrix completion problem underlying a large body of real-world applications such as signed link prediction and information propagation. That is, each entry of the matrix indicates a binary preference such as "like" or "dislike", "trust" or "distrust". However, the performance of existing matrix completion methods may be hindered owing to three practical challenges: 1) the observed data are with binary label (i.e., not real value); 2) the data are typically sampled non-uniformly (i.e., positive links dominate the negative ones) and 3) a network may have a huge volume of data (i.e., memory and computational issue). In order to remedy these problems, we propose a novel framework which {i} maximizes the resemblance between predicted and observed matrices as well as penalizing the logistic loss to fit the binary data to produce binary estimates; {ii} constrains the matrix max-norm and maximizes the F-score to handle non-uniformness and {iii} presents online optimization technique, hence mitigating the memory cost. Extensive experiments performed on four large-scale datasets with up to hundreds of thousands of users demonstrate the superiority of our framework over the state-of-the-art matrix completion based methods and popular link prediction approaches.