Broad Learning Through Fusions

This textbook provides state-of-art research development about data mining and machine learning works on fusion learning. Fusion learning is a new learning task, whose objective is to combine multiple different data fragments of the same information entities together for knowledge discovery. The fusion learning problems across multiple online social networks is a novel research domain introduced by the authors. How to effectively combine multiple social networks, and how to make use of the combined aligned networks, are both open questions at that time. Fusion learning is a novel yet important area, and there are adequate opportunities there to be discovered. Current research fusion learning works suffer from several big problems, which will compose the future research directions as well. The first part of the book provides some basic background knowledge of Fusion Learning , Machine Learning and Social Networks to make this book self- contained. The second part of the book includes , depending on the availability of training data, different categories of network alignment models have been proposed. By aligning different social networks together, various application services can benefit from the information transferred from different social networks. The third part will introduce five different traditional knowledge discovery applications across aligned social networks. Fusion learning is a novel yet important area, and there are adequate opportunities there to be discovered. Current research fusion learning works suffer from several big problems, which will compose the future research directions as well.