Link Prediction in Multi-relational Graphs using Additive Models

Download Now
Provided by: RWTH Aachen University
Topic: Data Management
Format: PDF
In this paper, the authors present a general and novel framework for predicting links in multi-relational graphs using a set of matrices describing the various instantiated relations in the knowledge base. They construct matrices that add information further remote in the knowledge graph by join operations and they describe how unstructured information can be integrated in the model. They show that efficient learning can be achieved using an alternating least squares approach exploiting sparse matrix algebra and low-rank approximations.
Download Now

Find By Topic