Graph data appear in almost all disciplines. Developing machine learning algorithms for graphs is a crucial task, with a plethora of interdisciplinary applications. As a prominent paradigm, network representation learning aims to embed nodes in a low-dimensional space, preserving the structural properties of the network. Nevertheless, this is a challenging task for various reasons. In the era of data deluge, we need to handle massive graphs. Besides, the models should capture the inherent complex structure of the data involved. Furthermore, we often have to deal with unreliable data sources. To address such challenges, GraphIA aims to design expressive representation learning models capable of leveraging rich structural and information semantics towards improving learning performance, while also ensuring scalability and robustness. Two interdisciplinary applications will be addressed: detection of influential spreaders in complex networks and analysis of network data in bioinformatics.