Situated at the interface between the philosophy of science computer science, the development of innovative algorithmic solutions will be informed by the
combination of three research perspectives. First, we will examine the use of research data in publications, in different disciplines, and for different research tasks, so as to understand to which extent scientific discoveries are based on data availability and how they are affected by data sharing cultures. Second, we will contribute new semantic technologies to support dataset search, to match research data with different user groups and scientific communities, and to generate research dataset search engine result pages. Third, using key insights into research dataset use and semantics, we will develop information retrieval algorithms for unsupervised dataset search and for predicting user interactions with dataset search engine results. These two will be combined into a self-learning method for retrieving and recommending research datasets.