Frameworks and toolsets

A major challenge in research scenarios in the domain of linguistic natural language processing is the annotation of language data at different levels of linguistic organization. Different levels of linguistic organization are typically analyzed and annotated by means of different tools, e.g. different part of speech taggers, named entity recognizers, syntactic parsers, coreference resolution etc. In most cases, these tools work separately from one another and require specific input formats and produce output formats that are an obstacle to combining these tools without extensive format transformations as inbetween steps. There are several reasons for this situation, e.g. a lack of standardization and a lack of process documentation defining the processes producing and further managing these formats. In order to joint several processing tools together into processing pipelines or scenarios, different toolsets and frameworks exist that overcome these obstacles or offer ways towards a solution.