Our main goal is to develop geoscience-driven Machine Learning (ML) applications (e.g. plug-ins to existing software packages, standalone, possibly web-based software) that will help academic and industry geoscientists: (i) better understand the tectono-stratigraphic development of sedimentary basins; and (ii) more accurately and quickly predict the nature and occurrence of hydrocarbons in mature and immature areas of petroliferous sedimentary basins. More specifically, we will focus on developing applications that will enable geoscientists to apply more quantitative techniques to very large subsurface datasets, thereby facilitating a better understanding of the multidimensional and non-linear relationships existing between some of the key geological properties (e.g. lithology distribution, physical properties). Our project thus has clear benefits for business (e.g. software development, marketing and sales), the exploration and production (E&P) industry (e.g. increased efficiency, and prediction accuracy in exploration for oil and gas) and academia (addressing multiple geoscience disciplines, and not least the integration between them). Furthermore, because integrated, geoscience-driven, ML applications for basin analysis and play evaluation do not yet exist, there is a significant academic challenge embedded in this computer science-based research problem.