MAF stands for Morpho-syntactic Framework. The goal of this framework is to standardize the morpho-syntactic annotation of textual data, such as the properties of each word, for instance part of speech information (e.g. noun, adjective, verb), morphological structure (i.e. the internal structure of words) and grammatical forms (e.g number, gender, person). This morpho-syntactic annotation will be represented in a data model.

There are two guiding principles in MAF: segmentation and annotation of textual data. In MAF, an annotated document consists of the raw original document and a set of its annotation. MAF provides the data model and the annotation format. The morpho-syntactic tag sets are not specified and they can be created individually. However, they should be based on data categories from the DCR (ISO 12620 Data Category Registry). The MAF also provides a generic solution for the representation of linguistic ambiguities, such as morphological lexical or structural.