LINK PREDICTION USING TENSOR DECOMPOSITION

In recent years, tensor decomposition has gained increasing interest in the field of link prediction, which aims to estimate likelihood new connections forming between nodes a network. This study highlights potential Canonical Polyadic enhancing prediction complex networks. It suggests effective algorithms that not only take into account structural characteristics network but also its temporal evolution. During process decomposition, initial is decomposed two-way tensors, known as factor matrices, representing different modes data. These matrices capture underlying patterns or relationships within network, providing insights structure and dynamics For evaluation, we examine dataset derived from WSDM. After preprocessing, data represented multi-way tensor, with each mode aspects such users, items, time. Our primary objective make precise predictions about links users items specific time periods. The experimental results demonstrate our approach significantly improves accuracy for evolving networks, measured by AUC.