The first meta-concept reflects a specification for data as a whole. We call this a Data Model. Good examples are ontologies or dictionaries. In addition, we have Data Sets that contain individual data according to the Data Model. The next basic meta-concept inside such a data model is a Concept referring to abstract notions as types of things of interest. Next we have Attributes being able to describe intrinsic characteristics and Relations to describe extrinsic characteristics of concepts. Concepts can be instantiated with Individuals referring to real world things you can or could point at. Such instances get lexical or reference Values for attributes respectively relations. Lexical values can be classified according to some Value Type (like string, decimal, integer, boolean etc.). Concepts can have Constraints restricting the amount of values or the values themselves or both. Also attributes and relations can have restrictions with respect to their source or target concepts (in case of relations) or value type (in case of attributes). Finally we have Derivations for concepts that tell us how new values for attributes or relations can be inferred from existing asserted values.We define three mechanisms as specific relations:1. Classification (inverse: instantiation), from ‘concrete’ to ‘abstract’2. Generalization (inverse: specialization), from ‘specific’ to ‘generic’3. Composition (inverse: decomposition), from ‘detailed’ to ‘global’These three mechanisms generate three hierarchy types namely a typology (of concepts), a taxonomy (of concepts, attributes or relations) and a meronomy (of concepts) respectively. Concepts are themselves instances of ‘Concept’ and can be instantiated in instances; Value Types are themselves instances of ‘Value Type’ and can be instantiated in values. Concepts can be specialized in other Concepts; attributes can be specialized in other attributes, and relations can be specialized in other relations. Specialized concepts, attributes and relationships inherit all constraints and derivations of the concepts, attributes, relationships they are specialized from. Concepts can be decomposed in concepts; Instances can be decomposed in other instances.Beside the three abstraction mechanisms we also provide other relation types:• Grouping, like for grouping all concepts, attributes, relations etc. into one data model,• Characterisation, indicating how attributes, relations, constraints and derivations relate to concepts/individuals on concept/instance level, and• The general associations on both concept and individual level.