A literature-based framework for semantic description of e-Science resources is presented. The main hypothesis is that the scientific literature can be used to mine semantic descriptions to facilitate semi-automatic annotation of e-Science resources. The approach includes text-mining methodologies to automatically learn domain annotation tag-sets; identify descriptive sentences and extract semi-formal functional descriptions of e-Science resources from the literature; link similar resources into semantic networks to enhance the resource discovery process. This description may be from another edition of this product.